To my father, mother, sisters and brother

To someone who taught me to believe in dreams

Akram

To my beloved family and true friends,

Ramadan
 Fixed-Network
Automatic Meter Reading
(AMR) System
Preface
Acknowledgements
First, we would like to thank our supervisor Dr. Gheith Abandah for his
kindness and advice throughout the year.

The kindly notes and tips from engineer Zyad Al-Khatib were very helpful;
thus we give him our deepest gratitude.

For Salam and Camelia who started the AMR last year, thank you.

Ala’, Majed and Faisal for your help, thanks a lot.

For all the telephone calls and mobile messages from friends and
colleagues, we send them “AT+CMGS=Thanks”.

To Thamer, Lahham and Ali, thanks for the great 4.5 years.

Without the encouragement, caring, patience, understanding, and prayers

from our mothers, fathers, sisters, and brothers none of this would have come
true; we love you all.

Finally, to all who helped or just loved, thanks.

Akram Jamal Awad

Ramadan Tyseer Abu Ghosh

II
Abstract
Fixed-Network Automatic Meter Reading (AMR) is the latest advance in
the telemetry field. It utilizes already available fixed communication networks
(e.g., the cellular network) for exchanging data to minimize cost and human effort.

The purpose of this project is to introduce a Fixed-Network AMR design

that manages the reading of the electricity meters at the consumers' side. This
design is intended to replace the exciting manual methods of gathering data.

The approach to the solution for this problem is made with the use of the
GSM network and a custom RF solution. Different hardware modules are
introduced to help exchange data between a central office and any node in the
system (i.e., customer's side).

The achieved results of the project are encouraging. Various data was
gathered and certain wireless processes were controlled. The GSM and RF
communication media were fully utilized by introducing our own GSM and RF
protocols. A computer software was developed to run the overall process.

The results obtained stand as a proof of concept for the credibility of

implementing such AMR system. They also show that even if the initial cost may
be relatively high, it is considered cheaper on the long-run.

Yet, the results still need further analysis and improvements to be made.
Using other chips and advanced microcontrollers can help in gathering more data
and control further processes.

III
Table of Contents
Preface I

Table of Contents IV

Ch 1: Introduction 1
1.1 Why Distributed Electrical Energy? 2
1.2 Why AMR? 3
1.3 Problem and Solution 6
1.3.1 Problem Statement 6
1.3.2 Solution Approach 6
1.4 Glossary 8
1.5 Symbols Convention 9
1.6 Report Outline 9

Ch 2: System Infrastructure 10
2.1 Microchip PIC18C452 11
2.1.1 Interrupts 12
2.1.2 I/O Ports 13
2.1.3 Addressable USART 13
2.1.4 Master Synchronous Serial Port (MSSP) 15
2.1.5 Timer0 16
2.2 Siemens M20 Terminal 18
2.2.1 AT Commands 18
2.2.2 M20 Interface 19
2.2.3 SMS with the M20 19
2.3 RF Modules 21
2.3.1 FM-RTFQ1-433 21
2.3.2 FM-RTFQ1-433 22
2.4 Analog ADE7756 23
2.4.1 Basic Operation 23
2.4.2 Adapting the Input Signals 24
2.4.3 Communicating with the ADE7756 24
2.4.4 Special Features 25
2.5 93C56, Serial EEPROM 27
2.5.1 Using the 93C56 EEPROM 27
2.5.2 Adapting Data to and from the 93C56 28
2.6 Supplementary Hardware 29
2.6.1 Line Driver (MAX232) 29
2.6.2 Voltage Regulator (LM317) 29
2.6.3 Open Collector Buffer 29

Ch 3: GSM Link 30
3.1 The PIC and the M20 31
3.1.1 Hardware Interface 31
3.1.2 AT Command Interface 31
3.2 GSM Link Protocol 36
3.2.1 Protocol Main Rules 36
3.2.2 Invalidity Reports 37
3.2.3 An Example 37

V
Ch 4: RF Link 39
4.1 Need Another USART Module 40
4.1.1 The Slave PIC 40
4.1.2 The Master PIC and the Slave PIC 40
4.1.3 How It Works 42
4.2 Adapting the RF Transmitter 43
4.2.1 Power Supply 43
4.2.2 Input Data Level 43
4.2.3 Transmitter's Phased Locked Loop (PLL) 44
4.3 The PIC and the RF Modules 45
4.3.1 Hardware Interface 45
4.3.2 How It Works 45
4.4 RF Link Protocol 46

Ch 5: Data Gathering & Process Controlling 47

5.1 Data Gathering 48
5.5.1 Calculate Digitally with the ADE7756 48
5.5.2 The PIC18C452 and the ADE7756 49
5.5.3 Calculating the Consumed Energy 50
5.5.4 Calculating the Power Factor 54
5.5.5 Reading the Status of a Switch 55
5.2 Process Controlling 55

Ch 6: Central Office 57
6.1 What is a Central Office? 58
6.2 System Overview 58
6.3 System Walkthrough 59
6.3.1 Initializing the Software 60
6.3.2 Sending Tasks 61
6.3.3 Receiving Tasks 62
6.3.4 Debugging and Troubleshooting 64

Ch 7: Total System Implementation 66

7.1 System Startup 67
7.1.1 NCN Startup 67
7.1.2 CCU Startup 67
7.1.3 Central Office Startup 67
7.2 System Operation 68
7.3 The System Through an Example 71

Ch 8: Conclusion 73
8.1 Work Done 74
8.2 Problems Faced 74
8.3 Further Suggestions 75

References 76

Appendix A 78
A.1 NCN Schematic Diagram 79
A.2 CCU Schematic Diagram 80
A.3 ADE7756 Calibration Schematic Diagram 81 VI
Introduction
1.1 Why Distributed Electrical Energy?
Since its "discovery", electricity has been always one of the fundamental requirements
for any modern civilization and its development. It is now at the heart of any property, whether
it is residential, commercial or industrial. Thinking for a while, we can imagine how deep
electricity goes through our lives. Factories, mills, laboratories, traffic lights, televisions, PCs
and cell phones are just few examples on the present situation of Electricity Empire.

For many decades, Studies in the field of electrical energy used to concentrate on
electricity generation, power transmission methods, transmission networks, line losses, etc…
Nevertheless, in all generations of electrical power technologies no significant changes in
electrical energy meters were remarked. It used to be the very basic and traditional meter either
in its shape or functionality, and it had nothing to do as a part of the system except electrical
energy consumption calculation.

But the scene turned far different in the last few years. Metering technology has
advanced wide steps to achieve the so called (new age of metering), which made benefit from
every possible technology. Latest models of electrical meters utilize the advantages of modern
communication systems, data transfer techniques and circuitry integration technologies.
Therefore, many new functions were added to the electrical meter and many other are not
farther than a single IC addition. All such advances bring the electrical energy meter much
nearer to its goal, not only to calculate energy consumption but also to be able to track the
quality of electricity and have full system control at the customer side.

But, are these technologies limited for electrical energy metering purposes only?
Fortunately, the answer is No! The electrical energy can be considered as one of the extremely
difficult quantities to be measured since it is a combination of two "invisible" quantities, voltage
and current. Approving on this statement, it will be clear to realize that once a technology
works properly for electrical energy metering purposes, the system can be modified or even
simplified to match other similar purposes such as water or gas consumption metering.

2
1.2 Why AMR?
Electrical energy meter reading has been always done manually by a staff of meter
readers who used to visit every meter location periodically and enter the meter reading by hand.
These readings were then added to previously read values or what can be called (the customer
consumption registry), from which the cost of the consumed energy during the latest cycle
(usually a month) can be calculated. This could be either a paper or computerized job. In some
cases where consumption rationalization strategies are followed; the consumer is charged for
the energy unit proportionally to his/her total consumption. It can be noticed that for this
method of meter reading, the previous feature is the only luxury of such a limited system. In
addition, many problems usually face the reader’s job like being unable to reach the meter
location, the absence of the property owners at the visit time or simply the human error in
recording the meter reading.

Various methods were and are developed to overcome the old traditional method
limitations, trying to make electricity metering in general and meter reading in particular more
accurate, reliable and functional. The new meter-reading methods can be classified into the
following categories [1]:

• In-Site Meter Reading

- Electronic Meter Reading (EMR)
- Remote Meter Reading (RMR)

• Automatic Meter Reading (AMR)

- Off-site Meter Reading (OMR)
- Mobile AMR
- Fixed-network AMR

A brief of every method is given hereby.

Electronic Meter Reading (EMR)

This method only differs from the recording book method in that readings are entered
directly through a keyboard to a mini-computer where data can be loaded easily to the system’s
main server at the utility office.

Remote Meter Reading (RMR)

A typical hand-held device with an optical-sensing or infra-red port is used here to
automatically read and store the meter reading once the reader holds his device and directs its
port to the meter side. In this method the probability of human error in recording the readings
is effectively reduced.

Off-site Meter Reading (OMR)

Unlike the previous illustrated systems, this is the first step in migration to the AMR
technology. It consists of a Portable Radio Network installed in handheld devices. In the other
side, a transmitter, or a transceiver, is installed in the meter. The handheld device will then
receive the meter reading while walking or slowly driving by the reading route. Although there
is no need in this method to visit the meter location itself, it is still mandatory that the meter
area is personally visited by the reader. This is mainly because the wireless communication
between the meter and hand-held device is of the short-range and minimum-power type due to
size limitations of both devices, as well as the highly-lossy medium of the communication
channel by nature (buildings, trees, metal gates, etc…).

Mobile AMR
The only human effort in meter reading is the driver’s effort in this method. A special-
purpose vehicle is prepared with equipment that can communicate with the meters using
relatively high antenna connected to a sensitive radio transceiver. The meter readings are
gathered by the vehicle while driving the route only. A computer installed in the vehicle 3
processes the data received from the meters and can either keep them to be loaded manually at
the utility or send them directly to the utility by means of a long-range communication network
such as cellular or satellite. It is clear that this method benefits as much as it can from the
utility's vehicle capabilities to reduce the losses, size and time limitations of the previous system,
however a new disadvantage arises which is the inability of the vehicle to reach the range of the
meter transmission in crowded or randomly-built areas.

Fixed-Network AMR
Of all the AMR implementation methods this can be called a "revolutionary method" in
the way that it completely eliminates many of the problems or limitations faced in the other
systems and opens the door for new functions, applications and developments to be added to
the system. Figure 1.1 shows the main structure of the Fixed-Network AMR system.

CCU
(Cell Control Unit)

Electricity Meters

Cell NCN Central Office

(Network Control Node)

CCU
(Cell Control Unit)

Figure 1.1: Fixed-Network AMR System Structure

The system’s network components include:

• Cell Control Unit (CCU)
• Network Control Node (NCN)
• Utility’s Host Computer

The CCU is connected directly to one or more electrical meters. It reads meters,
processes data for a variety of applications, stores data temporarily and transports the data to
the Host Computer when required.

The NCN is a regional concentrator and routing device that is installed on high points
like power poles or street light arms clearing all the surrounding area. The primary functions of
the NCN are data transfer and information routing between CCUs and the utility’s Host
Computer.

The Host Computer at the Central Office; which manages the collection of data
from the network devices and facilitates the download of any application information to
appropriate network devices. It also transfers the data to a database for storage and retrieval [2].

A look at the system as a whole will show that a communication channel between the
utility and any meter can be established 24 hours a day. Although this may seem to be very
costly initially; significant saving in terms of time, efficiency and accuracy of reading should
compensate after the system activation. This statement would look truer after reviewing the
coming points which are just some of the Fixed-Network benefits:

• Reduction in meter reading staff

• Improving accuracy and billing efficiency
• Eliminating the need to access premises
• Improving off-cycle reads
• Enabling remote line connection/disconnection process
• Theft identification
• Improving outage detection and restoration reporting 4
 • Improving reliability
• Improving distribution system planning
• Delivery of new services to customers
• Improving customer satisfaction

That is why our decision was made to utilize the benefits of the Fixed-Network AMR
method in our project as will be discussed in the following chapters.

5
1.3 Problem and Solution

1.3.1 Problem Statement

It is desired to build an automatic electrical-energy meter reading system using the
Fixed-Network AMR technology. The new system is intended to be reliable, efficient and cost
effective. It should satisfy the local-market special needs but also be flexible, scalable and
expandable so that it can be used successfully in any other market or can be easily modified
when a system updating plan is proposed.

Taking the decision to build a Fixed-Network AMR system immediately divides our
problem into subsystem problems. These are:

The Consumer-Side Terminal which consists of the meter itself and the Cell
Control Unit (CCU). Problems arise here are the type of meter to be used in the system and
the best implementation of the CCU, its structure, functions and the interface between the unit
and the meters in its cell. For this purpose all the meters connected directly to a CCU, with the
CCU itself, will be considered a “Cell”.

The Network Control Node (NCN). As it serves as a transponder in the system,

the type of the link between the NCN and the utility’s Host Computer should be determined
taking into consideration what is locally most available and feasible. Another link to be specified
is that between the NCN and the CCU. This link should be as cheap as could since it is
essential for every cell, however it should cover wide area of cells so that the NCN can serve the
maximum number of cells. An NCN with the cells it covers are said to construct a “Region”.
Additional point that should be studied is the protocols between devices in both links and the
way the NCN controls the job of CCUs in its region.

The Central Office which contains the Host Computer located at the utility’s offices
and the database that stores the customers’ information and consumption history. The billing
system can be also considered a part of the whole system as it uses the data gathered by the
Host Computer in issuing bills.

1.3.2 Solution Approach

Studying the available resources and approaches, the solution shown in figure 1.2 has
been adopted.

NVR NVR
ADE7756 M20 M20

Database
PIC PIC

ADE7756

ADE7756 ADE7756 Central Office

Cell
NCN

PIC ADE7756 NVR M20

93C56
Microcontroller Digital Meter GSM Modem RF Module Host Processor
EEPROM

Figure 1.2: Solution System Block Diagram

Concisely, an active-energy metering IC (Analog ADE7756) [4] receives two signals

proportional to the line voltage and current from which it calculates the consumed energy. This 6
energy is periodically read, accumulated and stored at a high-performance microcontroller
(Microchip PIC18C452) [3]. The RF transmitter (RF RTFQ1) and receiver (RF RRFQ1) [6]
provide the window for the CCU to communicate with NCN. Both microcontroller and RF
Tx\Rx form a CCU. The CCU can include more components depending on the additional
functions added to the system.

To enhance the system reliability; a non-volatile EEPROM (93C56) [7] is used to store
the meter reading and other important data at a CCU or NCN; and reload the stored data at
startup.

At the NCN, the utility’s orders to the system are received as Short-Messaging-Service
(SMS) messages through a GSM modem (M20 terminal) [5]. A microcontroller analyzes the
message and sends the proper commands to all CCUs in its region or to a specific CCU only.
Any readings, alerts or reports sent by the CCU are received in the NCN, reshaped to an SMS-
message form and sent again to the Host Computer.

7
1.4 Glossary
The following table describes some technical terms that are used in this report for
clarification.

Term Description
ADE7756 The ADE7756 active-energy metering IC produced by Analog Devices.
It takes the current and voltage signals from the power line and
produces digital streams representing the consumed energy and other
power-line readings
AMR Automatic Meter Reading
CCU Cell Control Unit: It is the PIC and RF modules connected to the
meters at the consumer side. It communicates with the NCN through
the RF modules
Cell The CCU and meters connected to it
Central Office The utility office that contains the Host Computer and any other
supplementary devices
Line Driver The MAX232 low-power +5 TIA/EIA-232 dual driver/ receiver. It is
used between M20 and the PIC
Host Computer The mainframe computer at the utility that controls the system
operations
M20 The Siemens M20 Terminal which is the GSM modem of the system
NCN Network Control Node: It is a PIC connected to both the M20 and the
RF modules. It represents the linking node between the utility's central
office and the CCUs
NVR The 93C56 non-volatile EEPROM chip used as a storage device for the
NCN and the CCU.
PIC The PIC18C452 high-performance microcontroller produced by
Microchip and works as the heart of every system part
Region The group of CCUs covered by one NCN, with the NCN itself
RF Rx The FM RRFQ1-433 Radio-Frequency FM receiver working at 433
MHz. It is a production of RF Solutions
RF Tx The FM RTFQ1-433 Radio-Frequency FM transmitter working at 433
MHz. It is also produced by RF Solutions
SMS Short Messaging Service
SPI Serial Peripheral Interface, it is a serial data transfer protocol in which a
master device (SPI enabled) communicates with a slave device (SPI
enabled) through some data and control lines
USART Addressable Universal Synchronous Asynchronous Receiver
Transmitter module in the PIC18C452 microcontroller. It is used to
interface the PIC with other terminal equipments like a PC or the M20

8
1.5 Symbols Convention
The following symbols are used throughout this report to present the system
components.

PIC MAX232 ADE7756 NVR M20

RS232/TTL 93C56
PIC18C452 ADE7756 M20 RF Module Host Computer Electrical Load
Line Driver EEPROM

1.6 Report Outline

Chapter 1: Introduction. (Akram)

Chapter 2: System Infrastructure. In this chapter a detailed illustration of the main

components used in the structure of the project is given. The device features that are used in
the project are also previewed there. (Ramadan)

Chapter 3: GSM Link. The first part of this chapter describes the connection between
the M20 and the PIC in the NCN. It also discusses the method in which the SMS message is
analyzed in the PIC and how a new message is built and sent to the M20. The rest of this
chapter shows the GSM protocol which intends to found a shared language understood by both
the PIC and M20 to achieve the system’s goal and to insure secure data transfer. (Akram)

Chapter 4: RF Link. This chapter deals with the RF-related components in the
project, the communication between master and slave PICs in the NCN and the adaptation of
the RF Tx and Rx modules. (Akram)

Chapter 5: Data Gathering. In chapter 5 we investigate the way in which the

ADE7756 digitally calculates energy, as well as the meter data gathering process made by the
PIC at the CCU. The Power Factor calculation function is also introduced by the end of this
chapter. (Ramadan)

Chapter 6: Central Office. This chapter introduces a software program that simulates
the Central Office. (Ramadan)

Chapter 7: Total System Implementation. The operation of the system as a whole is

described here using a detailed example. (Akram)

Chapter 8: Conclusion. The final results of the project are shown in the last chapter.
Furthermore; suggestions regarding future developments of the project are discussed here.
(Akram & Ramadan)

9
System Infrastructure
 Our desired goal is simply to establish a communication link and to gather data. The
solution for our problem resides on different hardware components interacting with each other
to achieve this goal. This chapter introduces these hardware modules briefly and mentions
some of the good features they provide to our system*.

2.1 Microchip PIC18C452

It is simply a highly integrated chip that contains all the components comprising a
controller [3]. Typically this includes a CPU, RAM, ROM, I/O ports, Digital Communication
Modules, Timers and other integrated components. Unlike a general-purpose computer (which
also includes all of these components), a microcontroller is designed for a very specific task; to
control a particular system. As a result, the parts can be simplified and reduced, which cuts
down on production costs and reduces the need for many computers to implement the system.

RC4/SDI/SDA
RC7/RX/DT
RC6/TX/CK
RB3/CCP2

RD7/PSP7
RD6/PSP6
RD5/PSP5
RD4/PSP4

RD3/PSP3
RD2/PSP2
RC5/SDO
RB2/INT2
RB1/INT1
RB0/INT0
RB7
RB6
RB5
RB4

VDD
VSS
40
39
38
37
36
35
34
33
32
31
30
29
28
27
26
25
24
23
22
21
PIC18C45 2
10

12
13
14
15
16
17
18
19
20
11
1
2
3
4
5
6
7
8
9
MCLR/VPP

VSS
VDD
RA0/AN0

RC2/CCP1
RA1/AN1

RE0/RD/AN5

OSC2/CLKO/RA6
OSC1/CLKI
RE1/WR/AN6
RA2/AN2/VREF-

RE2/CS/AN7

RC3/SCK/SCL
RA4/T0CKI

RC0/T1OSO/T1CKI
RA3/AN3/VREF+

RD0/PSP0
RD1/PSP1
RC1/T1OSI/CCP2
RA5/AN4/SS/LVDIN

Figure 2.1: Pin Diagram for the Microchip PIC18C452

The PIC within our hands is from the PIC18CXXX family. It is a family of high
performance, CMOS, fully static, 16-bit MCUs (Microcontroller Units) with integrated
analog-to-digital (A/D) converter. All PIC18CXXX MCUs incorporate an advanced RISC
architecture. The PIC18CXXX has enhanced core features, 32 level-deep stack, and multiple
internal and external interrupts sources. The separate instruction and data busses of the
Harvard architecture allow a 16-bit wide instruction word with the separate 8-bit wide data. A
total of 77 instructions (reduced instruction set) are available. The PIC18CXXX family has
special features to reduce external components, thus reducing cost, enhancing system reliability
and reducing power consumption. The PIC18CXXX family is considered among the top of the
shelf considering Microchip products.

The C letter in the PIC18CXXX indicates that this family of MCUs is an OTP
program-memory (One-Time-Programmable), which means that the memory used for storing
the program can't be cleared electrically (not EEROM) but by using ultraviolet radiation (it is
EPROM).

Other Microchip Families (e.g. PIC16FXXX) are Flash MCUs, this means that they
can be programmed and erased electrically without the need to the ultraviolet radiation
(EEROM), this makes the development process faster, but since there were no PIC18FXXX
MCUs available, we had to use this family because of the more features it provides.

11

* Most of the figures in this chapter are extracted from the datasheets of each hardware module. One can find these datasheets on
the companion CD
 The PIC18C452 has 32K bytes of EPORM (Program memory) and 1536 bytes of
RAM (Scratch Board), this relatively huge-program-memory space helps us develop software
with this PIC that can't be developed with others. The PIC18C452 can operate at a frequency
up to 40 MHz.

Software development for the PIC18C452 (and any PIC from any family) can be done
using the MPLAB. The MPLAB is a complete Integrated Development Environment (IDE)
from Microchip, it allows the user to edit, debug, compile, download code and even simulate
from a single user interface. MPASM is a utility that is integrated with the MPLAB; it helps
simulating software behavior before downloading the final machine code to the PIC. The
PICSTART Plus is a serial hardware programmer that can be used within the MPLAB to
program the PIC. The development for the PIC18C452 can be either using the C language or
the assembly language; we have developed all the programs using the assembly language.

Figure 2.2: Microchip MPLAB IDE

In the next subsections we will introduce some of the main features of the PIC18C452
that will help build up the system. The PIC18C452 is the core of our system because it has to
interface almost all other modules.

2.1.1 Interrupts
Interrupt-driven software simplifies development and helps build a more rigid system.
Interrupts can come from many sources (internal and external). We shall be using interrupt-
driven architecture throughout the entire system. This means that each module in the system
has its own interrupt that the PIC sees and services.

There are two interrupt vectors. One interrupt vector is for high priority interrupts and
is located at address 000008h. The other interrupt vector is for low priority interrupts and is
located at address 000018h. We shall use interrupt priority because of the many interrupts
requesting servicing; and some interrupts are considered much more important than others.

12
 The PIC (in general) has four registers associated with interrupts. The INTCON
register contains the Global Interrupt Enable bit (GIE), as well as the Peripheral Interrupt
Enable bit (PEIE), the PIE/PIR register pair that enables the peripheral interrupts and displays
the interrupt flag status and the Interrupt Priority Register (IPR) that controls whether the
interrupt source is a high priority or low priority interrupt. There are other registers used to pull
interrupt statuses for other internal and external modules too.

When a valid interrupt occurs, program execution vectors to one of these interrupt
vector addresses and the corresponding Global Interrupt Enable bit (GIE, GIEH, or GIEL) is
automatically cleared (this is done to disable other interrupts from generating another interrupt
signal while servicing the current). In the interrupt service routine, the sources of the interrupt
can be determined by testing the interrupt flag bits. The interrupt flag bits must be cleared
before re-enabling interrupts to avoid infinite interrupt requests. Most flag bits are required to
be cleared by the application software. There are some flag bits that are automatically cleared
by the hardware. When an interrupt condition is met; that individual interrupt flag bit will be
set regardless of the status of its corresponding mask bit.

While like in normal operation one needs to push and pop the WREG and STATUS
registers; this is not necessary in our case because the PIC18C452 automatically pushes and
pops these registers at interrupts.

The “return from interrupt” instruction, RETFIE, can be used to mark the end of the
interrupt service routine. When this instruction is executed, the stack is “popped” and the GIE
bit is set (re-enable interrupts). For more information about interrupts, see the Interrupt Logic
Block Diagram in the PIC18C452 datasheets on the companion CD, it presents a good
understanding of the interrupt login in the PIC18C452.

2.1.2 I/O Ports

General purpose I/O pins can be considered the simplest of the peripherals the PIC
provides. They allow the PIC to monitor and control other devices. To add flexibility and
functionality to a device, some pins are multiplexed with other functions. These functions
depend on which peripheral features are on the device. In general, when a peripheral is
functioning, that pin may not be used as a general purpose I/O pin. For most ports, the I/O
pin’s direction (input or output) is controlled by the data direction register, called the TRIS
register. TRIS<x> controls the direction of PORT<x>. A ’1’ in the TRIS bit corresponds to
that pin being an input, while a ’0’ corresponds to that pin being an output. The PORT register
is the latch for the data to be output. When the PORT is read, the device reads the levels
present on the I/O pins (not the latch).

I/O ports were used here for many reasons; one reason is to let the PIC control other
devices (e.g. we will control another PIC in the NCN module), another reason is to simply use
LEDs to show the progress of the software on the PIC. Since these I/O pins are multiplexed
with other features; we have used many of the features through these pins.

The PIC18C452 has 5 I/O ports different in length; they are PORTA, PORTB,
PORTC, PORTD and PORTE. We will be utilizing some of the pins of all ports except
PORTE. For detailed information about these ports, refer to the PIC18C452 datasheets on the
companion CD.

2.1.3 Addressable USART

The Addressable Universal Synchronous Asynchronous Receiver Transmitter
(Addressable USART) module is one of the serial I/O modules available in the PIC18CXXX
family (another is the MSSP). The USART can be configured as a full duplex asynchronous
system that can communicate with peripheral devices, such as personal computers.
13
 The USART module can be setup through the TXSTA and RCSTA registers; setup
includes selecting the serial communication type (synchronous or asynchronous), number of
bits, baud rate… The PIC will use this module to communicate with the Siemens M20 and
with the other CCUs through the RF modules.

We shall be using the PIC in the asynchronous mode. Asynchronous mode is selected
by clearing the SYNC bit (TXSTA register). In this mode, the USART uses standard NRZ
format (one start bit, eight or nine data bits and one stop bit), we are using the 8 bits format.
An on-chip special 8-bit baud rate generator can be used to derive standard baud rate
frequencies from the oscillator. The USART transmits and receives the LSb first. The
USART’s transmitter and receiver are functionally independent, but use the same data format
and baud rate. The baud rate generator produces a clock either x16 or x64 of the bit shift rate,
depending on the BRGH bit (TXSTA register). The data on the RX pin is sampled three times
by a majority detect circuit to determine if a high or a low level is present at the RX pin.

The USART transmitter block diagram is shown beside. The heart of the USART
transmitter is the transmit shift register (TSR). The shift register obtains its data from the
read/write transmit buffer, TXREG. The TXREG register is loaded with data in software. The
TSR register is not loaded until the STOP bit has been transmitted from the previous load. As
soon as the STOP bit is transmitted (we have to keep pulling the TXSTA<TRMT> bit), the
TSR is loaded with new data from the TXREG register. Once the TXREG register transfers
the data to the TSR register, the TXREG register is empty and flag bit TXIF (PIR1<4>) is set.

Data Bus

TXIF TXREG Register

TXIE
8
MSb LSb
(7) 0 Pin Buffer
and Control
TSR Register RC6/TX/CK
Interrupt

TXEN Baud Rate CLK

TRMT SPEN
SPBRG

Baud Rate Generator

USART TRANSMIT BLOCK DIAGRAM

x64 Baud Rate CLK CREN

SPBRG 64 MSb RSR Register LSb

or
16
STOP 7 1 0 START
Baud Rate Generator

Pin Buffer FIFO

Data
and Control Recovery
RCREG Register
RC7/RX/DT

Interrupt RCIF
SPEN 8
RCIE
USART RECEIVE BLOCK DIAGRAM Data Bus

Figure 2.3: USART Transmit and Receive Block Diagrams

In the receiver block diagram. The data is received on the RC7/RX/DT pin and drives
the data recovery block. The data recovery block is actually a high speed shifter operating at
x16 times the baud rate; whereas the main receive serial shifter operates at the bit rate, or at
FOSC. To set up an Asynchronous Reception: Initialize the SPBRG register for the
appropriate baud rate. Enable the asynchronous serial port. If interrupts are desired (we will
use the receive interrupt later on); we must set the enable bit RCIE. We enable continuous
reception by setting bit CREN. Flag bit RCIF will be set when reception is complete and an
interrupt will be generated. The PIC goes into an ISR (interrupt service routine). The software
there pulls the RCIF bit and determines the source of the interrupt. Read the 8-bit received
data by reading the RCREG register.

The baud rate is generated by writing to the SPBRG register; the value in this register is
evaluated in a formula along with the clock frequency of the PIC to determine the desired baud 14
rate. Choosing a clock frequency of 3.6864 MHz was made to ensure no percentage of error is
in the desired baud rate. For more information about the USART, refer to the PIC18C452
datasheets on the companion CD.

2.1.4 Master Synchronous Serial Port (MSSP)

The Master Synchronous Serial Port (MSSP) module is a serial interface useful for
communicating with other peripherals or microcontroller devices. These peripheral devices may
be serial EEPROMs, shift registers, display drivers, A/D converters, etc. The SSP module can
operate in one of two modes:

• Serial Peripheral Interface (SPI): It is typically a 3-wire interface, with a data out
line (SDO), a data in line (SDI), and a clock line (SCK). Since the clock is present,
this is a synchronous interface
2
• Inter-Integrated Circuit (I C): This is a two wire communication interface

We shall use the MSSP in the SPI mode because all of the peripherals that the PIC has
to communicate with are SPI enabled. The PIC uses the SPI to communicate with the
ADE7756 and the Slave PIC; this is why we should thoroughly explain the SPI.

SPI operates in 4 different modes (2 master modes and 2 slave modes). SPI is a master-
slave data interface; this means that one part has to generate the clock (and optionally the Chip
Select) signal. We will be using the PIC in different SPI modes depending on the situation.
(Coming a head: we will connect two PICs together through the SPI to interchange data, this
means that one must be set to Master mode and the other must be set to Slave mode).

The SPI mode allows 8-bits of data to be synchronously transmitted and received
simultaneously. When initializing the SPI, several options need to be specified. This is done by
programming the appropriate control bits in the SSPCON register (SSPCON<5:0>) and
SSPSTAT<7:6>. These control bits allow the following to be specified:

• Master Mode (SCK is the clock output)

• Slave Mode (SCK is the clock input)
• Clock Polarity (Idle state of SCK)
• Clock edge (output data on rising/falling edge of SCK)
• Data Input Sample Phase
• Clock Rate (Master mode only)
• Slave Select Mode (Slave mode only)
Internal
The MSSP consists of a transmit/receive Shift Register Data Bus

(SSPSR) and a buffer register (SSPBUF). The SSPSR shifts the Read Write

data in and out of the device, MSb first. The SSPBUF holds SSPBUF reg

the data that was written to the SSPSR, until the received data
is ready. Once the 8-bits of data have been received, that byte is SSPSR reg
moved to the SSPBUF register. Then the buffer full detect bit, SDI bit0 Shift
Clock
BF (SSPSTAT<0>), and interrupt flag bit, SSPIF, are set. This SDO
double buffering of the received data (SSPBUF) allows the next
byte to start reception before reading the data that was just SS Control
Enable
received. Any write to the SSPBUF register during SS Edge
Select
transmission/reception of data will be ignored, and the write
collision detect bit, WCOL (SSPCON<7>), will be set. When 2
Clock Select
the application software is expecting to receive valid data, the SSPM3:SSPM0
SSPBUF should be read before the next byte of data to transfer SMP:CKE 4
2 (
TMR2 output
)
2
is written to the SSPBUF. Buffer full bit, BF (SSPSTAT<0>), Edge
Select Prescaler TOSC
indicates when SSPBUF has been loaded with the received data SCK 4, 16, 64

(transmission is complete). When the SSPBUF is read, the BF

15
Data to TX/RX in SSPSR
TRIS bit
bit is cleared. This data may be irrelevant if the SPI is desired
Figure 2.4: SPI Block Diagram
only to transmit. Generally the SSP Interrupt is used to determine when the
transmission/reception has completed. The SSPBUF must be read and/or written. If the
interrupt method is not going to be used, then software polling can be done to ensure that a
write collision does not occur. We shall be using the interrupt method.

In the Master mode: the master can initiate the data transfer at any time because it
controls the SCK. The master determines when the slave (Processor 2) is to broadcast data (the
slave can only give the data when the master enables the clock). Data is transmitted / received
as soon as the SSPBUF register is written to. As each byte is received, it will be loaded into the
SSPBUF register as if a normal received byte (interrupts and status bits appropriately set). The
clock polarity is selected by appropriately programming bit CKP (SSPCON<4>). The SPI
clock rate (bit rate) is user programmable to be one of the following:

• FOSC/4 (or TCY)

• FOSC/16 (or 4 • TCY)
• FOSC/64 (or 16 • TCY)

This allows a maximum data rate of 5 Mbps (at 20 MHz).

In slave mode: the data is transmitted and received as the external clock pulses appear
on SCK (generated by the master device). When the last bit is latched, the interrupt flag bit
SSPIF is set.

When in slave select mode, the SS pin allows multi-drop for multiple slaves with a
single master (we will use this mode because the master PIC is going to interface both the
ADE7756 and the slave PIC).

To enable the serial port the SSP Enable bit, SSPEN (SSPCON<5>), must be set. To
reset or reconfigure SPI mode, clear the SSPEN bit which re-initializes the SSPCON register,
and then set the SSPEN bit. This configures the SDI, SDO, SCK, and SS pins as serial port
pins. For the pins to behave as the serial port function, they must have their data direction bits
(in the TRIS register) appropriately programmed. That is:

• SDI must have the TRIS bit set

• SDO must have the TRIS bit cleared
• SCK (Master mode) must have the TRIS bit cleared
• SCK (Slave mode) must have the TRIS bit set
• -SS have the TRIS bit set

For more information about the SPI, refer to the PIC18C452 datasheets on the
companion CD.

2.1.5 Timer0
The Timers modules embedded within the PIC present a simple counter/timer for the
use of counting external pulses or timing a periodic task to happen.

We need to use the timer module because we have to perform certain periodic tasks
(e.g. reading the accumulated energy from the ADE7756). Timer0 helps doing so by
generating an interrupt every 2 seconds (programmable).

Timer mode is selected by clearing the T0CS bit (T0CON register). In timer mode, the
Timer0 module will increment every instruction cycle. Timer0 can be configured as an 8-bit or
a 16-bit timer. To configure the timer as a 16-bit counter, the T08BIT bit (T0CON register)
must be cleared. We are using the timer mode not the counter mode because we want to
generate a periodic interrupt.
16
 The TMR0 interrupt flag bit is set when the TMR0 register overflows. When TMR0 is
in 8-bit mode, this means the overflow from FFh to 00h. When TMR0 is in 16-bit mode, this
means the overflow from FFFFh to 0000h. This overflow sets the TMR0IF bit (INTCON
register). The interrupt can be disabled by clearing the TMR0IE bit (INTCON register). The
TMR0IF bit must be cleared in software by the interrupt service routine.

TIMER0 BLOCK DIAGRAM IN 16-BIT MODE

FOSC/4 0
0
Sync with Set Interrupt
1 Internal TMR0L TMR0
Clocks High Byte Flag bit TMR0IF
T0CKI pin Programmable 1 on Overflow
Prescaler 8
T0SE (2 TCY delay)
3
Read TMR0L
T0PS2, T0PS1, T0PS0
T0CS PSA Write TMR0L
8
8
TMR0H

Data Bus<7:0>

Figure 2.5: Timer0 Block Diagram in 16-bit Mode

There is a prescaler for Timer0, this prescaler is enabled or disabled in software by the
PSA bit (T0CON register). Setting the PSA bit will enable the prescaler. The prescaler can be
modified under software control through the T0PS2:T0PS0 bits. When the prescaler is
enabled, prescaler values of 1:2, 1:4... 1:256 are selectable.

We want to generate an interrupt every 2 seconds; this is done by setting Timer0 to 16-
bit timer mode (with internal clock), and writing E3E0h to the Timer0 high and low bytes
(TMR0H and TMR0L). This means that Timer0 will overflow from FFFFh to 0000H after
7.8125 msec. This 7.8125 msec is multiplied by the prescaler 256 to give a total of 2 seconds.
For more information about Timer0, refer to the PIC18C452 datasheets on the companion
CD.

Remember
The PIC18C452 will be referred to throughout the
PIC
entire text as the symbol shown.

17
2.2 Siemens M20 Terminal
The Siemens M20 Terminal [5] is a GSM modem that combines all the features
required by developers and users to gain the power of GSM in their communication system. It
is designed for handling complex industrial applications such as telemetry, and for integration
in stationary or mobile fields all over the world.

Operating status LED Mounting holes Handset interface

bottom top

V.24 interface

M 20 Terminal
Sub-D socket

Power supply
(8-28,8V)
and audio interface

FME
antenna connector SIM card reader

Figure 2.6: Siemens M20 Terminal

The Siemens M20 Terminal is a GSM900 Phase II voice, data, fax and SMS terminal
device. This device is intended for universal use in various areas of application.

The most important features of the M20 are:

• SDI must have the TRIS bit set

• Data, voice, fax and SMS services
• GSM 900 Phase II
• Data transmission rate up to 9600 bps
• SMS (Text mode, PDU mode)
• Sensitivity –108 dBm
• Integrated echo suppression and noise reduction (for handset)
• Digital audio interface
• Mini SIM card reader with integral drawer (3V)
• Reloadable software
• Software interface with commands as AT-Hayes
• 2W power
• Power supply (8 V to 28.8 V)
• Terminal equipment (TE) interface using RS232: Baud rates: 300-57600, default
19200.
• Weight 145 g
• Dimensions (max) LxWxH = 107.0 x 63.5 x 31.3 mm

We are using the M20 to establish the link between the Central Office and the NCN.
The broadness of the GSM network will give our system the great benefit of covering the
largest area possible.

2.2.1 AT Commands
The operating functions of the M20 Terminal are accessed through the use of AT
commands; these AT commands are conforming to GSM 07.07 and GSM 07.05 standards.
AT commands are available via the serial interface of the M20 for function implementation.

The AT standard is a line-oriented command language. Each command is made up of

three elements: the prefix, the body, and the termination character.
18
 • The prefix consists of the letters “AT“
• The body is made up of individual characters; it consists of a name and (if
applicable) associated values
• The default termination character is “<CR>“(= 0x0D)

Example: AT+CMGR="12"
This command reads the message with index 12 stored in the default SIM card place.

Commands for the M20 can be categorized in two categories:

• Setup commands: used to set the different parameters of the M20 (e.g. Baud
rate, response type…)
• Communication Commands: used to establish communication using the M20
(e.g. Read & Send SMS)

The M20 implements what is called User Profiles; they are non volatile memory (NVR)
used to store the parameters the user sets for the M20. When the M20 is operated the next
time, it will load the defined user profile at start up. This helps reduce AT commands used to
setup the M20 on each startup.

Yet, not every setting for a parameter can be saved in a user profile; that's why we need
to initialize these parameters on every startup (e.g. to disable echoing characters back on the
serial interface, we must send the ATE0 command to the M20 each time we start the M20 up).

The specific AT commands that are needed for out system implementation shall be
mentioned when we talk about connecting the M20 with the PIC18C452. For a complete list
all AT commands for the M20, refer to the datasheets on the companion CD.

2.2.2 M20 Interface

As mentioned before; the M20 is accessed through the serial interface using RS232. We
are using the DB9 connector.

The DB9 connector has 9 signals, but we are not using all these signals; we are only
using the basic TXD, RXD and GND signals to establish the interface. The M20 supports
handshaking through the serial interface, but we are not using this feature because the M20 has
to be connected to the PIC18C452; and the PIC doesn't support handshaking though its
USART module.

Pin Description
5
1 Data Carrier Detect (DCD) 9
2 Receive Data (RXD) 4
8
3 Transmit Data (TXD) 3
4 Data Terminal Ready (DTR) 7
2
5 Ground (GND) 6
6 Data Set Ready (DSR) 1
7 Request to Send (RTS)
8 Clear to Send (CTS) DB9 CONNECTOR
9 Ring Indication (RING)

Figure 2.7: Pin Diagram of the DB9 Connector. The M20 Uses Only the Marked Signals

2.2.3 SMS with the M20

The GSM link will connect the NCN with the Central Office. The M20 provides us
with more than one way to establish the link (Voice, Data, Fax and SMS) and the SMS sounds
the best to use for the current system. SMS (Short Messaging System) is easy to implement and
is cheap. A standard message will contain 160 characters which is pretty enough for the type of
data we are planning the send and receive.
19
 The SMS in the M20 supports two modes: the Text mode and the PDU (Protocol Data
Unit) mode. The text mode is the normal mode used between people sending each other
messages using regular mobiles. The PDU mode is a special mode where data can be send
using certain protocols that comprises specific data and certain data coding schemes, PDU
messages can handle more data than a normal text mode message. Below is an example of a
PDU message. The normal Service Center may not handle PDU formatted messages.
DCS
MR
(7-bit
PDU- coding)
type PID

0011000781214365F70000AA05E8329BFD06
len VP
SCA
(four days) UD ("hello" in 7 bit
type of Destination Adress: UDL default alphabet)
number (Phone-number 1234567)

Figure 2.8: Example of a Message in the PDU Mode

We shall be using the SMS in the text mode because the data being sent and received is
simple. Security is handled through creating our own GSM protocol; this protocol helps
immune the system and handles errors. The GSM protocol will be mentioned with more details
when we talk about connecting the M20 with the PIC18C452.

SMS is handled using certain AT commands through which we can read and write
messages.

Remember
The Siemens M20 will be referred to throughout
the entire text as the symbol shown. M20

20
2.3 RF Modules
An important goal in our design is to make the system cost effective; we can't put an
M20 Terminal on every node in the system that we wish to get readings from, instead we shall
be using cheap RF modules to link the NCN with the surrounding CCUs. This will reduce cost
and will produce a nice challenge to stand for.

RF modules are specified by their operating frequency, modulation type, maximum data
rate and maximum distance of operation (in both open fields and in buildings). FM
transceivers are considered more stable in operation than AM transceivers, because FM
transceivers are more noise-immune than AM ones.

The frequency spectrum is divided between major systems in almost every country in
the world, e.g. the cellular system in Jordan operates in the frequency range around 900 MHz.
The frequency around 433 MHz can be used by the public (within some standard operating
ranges and conditions), and it is the frequency of operation of the RF modules in our hands.
The 433 MHz is not used in Jordan by any major system. According to the communication
theory, the antenna length of the modules should be (λ/4), that's about 17 cm. No special
antennas where used here because the transmitter and the receiver were too close to each other
for the demo purposes, but if they are to spaced a larger distance; we should consider changing
the antenna type.

The RF modules in our hands are the transmitter FM-RTFQ1-433 and the receiver
FM-RRFQ1-433; they are both made by the company (RF Solutions). They operate on the
433 MHz frequency and the transmitter range is about 75 meters between buildings (which
defines the maximum radius of our Region), this range depends mainly on the type of antenna
used and on the positioning of the module.

2.3.1 FM-RTFQ1-433
It is the transmitter module [6]; it uses a special purpose IC that does the job of
modulating the signal using FSK modulation. The maximum input frequency is 4.6 Kbps. The
transmitter operates at a voltage level of 3V; this was made to reduce power dissipation and to
let the module be used in handheld devices. (This requires some signal level modifications
because all of the other digital circuitry is devices that operate at the 5V voltage level).

In

XTAL PLL
XTAL Power Amp
Oscillator Synthesizer
Out

En

Pin Name Description

1 6 1 En Enable (active high)
Top View 2 IN Data input
(Components
3 GND Ground, Connect to RF earth return path
2 Underneath)
5 4 Vcc Supply Voltage
3 4 5 GND Ground, Connect to RF earth return path
6 EA External Antenna

Figure 2.9: The FM-RTFQ1-433 Transmitter, Operation and Pin Diagram

The FM transmitter uses an internal PLL circuit (Phase Locked Loop) to lock on the
carrier frequency of 433 MHz. This PLL requires time to lock to the frequency each time we
transmit, this problem is handled by software; we will mention it in more details when we put
the RF modules and the PIC together.

21
2.3.2 FM-RTFQ1-433
It is the receiver module [6]; it also uses a special purpose IC that does the job of
demodulating the received signal (it uses the TDA 5210, it is an ASK/FSK Single Conversion
Receiver). Unlike the transmitter, the receiver operates at the 5V voltage level; this makes it
ready for connection with the other digital circuitry.

Vcc Vcc
In

Pre Amp 10.7MHz FM

Mixer Demodulator
Comparator
Filter IF Filter
Out

PLL
GND Synthesizer GND GND

Pin Pin Name Pin Pin Name

Component Side 1 +Vcc 11 GND
2 GND 12,13 NC
1 3 7 11 15
3 Data In (Antenna) 14 Data Out
7 GND 15 Power Down

Figure 2.10: The FM-RRFQ1-433 Receiver, Operation and Pin Diagram

The FM receiver uses an internal intermediate frequency of 10.7 MHz, after that it
demodulates the signal and a comparator outputs either a "1" or a "0" at a CMOS/TTL voltage
level. The receiver operates only when there is a carrier signal detected, if there is no carrier
detected it will keep generating noise (this problem was faced because the manufacture didn’t
use the Carrier Detect signal provided on the TDA 5210 IC, this raised some problems when
implementing the modules in the final system, we will discuss this problem later on). For more
information about the RF modules, refer to the datasheets on the companion CD.

Remember
The RF module will be referred to throughout the
entire text as the symbol shown.

22
2.4 Analog ADE7756
From Analog Devices, the ADE7756 [4] is a high accuracy electrical power measuring
IC. Output data can be read either from the serial interface (using an MCU) or from the pulse
output (using a pulse counter). The ADE7756 incorporates two second order sigma-delta
ADCs, reference circuitry, and all the signal processing required to perform active power and
energy measurement.
AV DD RESET DVDD DGND

Reset 1 20 DIN
PGA
MULTIPLIER HPF1 MULTIPLIER LPF2 ADE7756 ZX
DV DD 2 19 DOUT V1P
ADC
V1N SAG
AV DD 3 18 SCLK

V1P 4 17 CS APGAIN[11:0]
TEMP APOS[11:0]
SENSOR
V1N 5 ADE7756 16 CLKOUT PHCAL[5:0]
TOP VIEW V2P
V2N 6 15 CLKIN
V2N
ADC O DFC

V2P 7 14 IRQ
AGND 8 13 4k ADE7756 REGISTERS CF
SAG 2.4V AND
REFERENCE SERIAL INTERFACE
REF IN/OUT 9 12 ZX
LPF1 CFDIV[11:0]
DGND 10 11 CF

AGND REFIN/OUT DIN DOUT SCLK CS IRQ CLKIN CLKOUT

Figure 2.11: Analog ADE7756, Pin Configuration & Functional Block Diagram

The ADE7756 uses two differential input Channels (voltage and current signals) to
calculate the energy consumed by a certain electrical load after performing certain digital steps.
The two channels are input both as voltage signals. The current signal is converted into a
voltage signal using traditional current sensing tools (shunt or current transformer).

2.4.1 Basic Operation

Simply, the main function of the ADE7756 is to multiply the voltage channel (CH2) by
the current channel (CH1) and extract the DC component of the product (LPF2). The result
of this DSP is the active (average) power. The ADE7756 accumulates this information and
provides the active energy in a register (AENERGY). A frequency output (CF) is proportional
to the active energy and can be used to drive a pulse counter or simply a LED.

0110010101
ADC
Currnet LPF Accumulate
Consumed
Instantaneous Average Energy
Power Power
0110010101
ADC
Voltage

Figure 2.12: ADE7756 Basic Operation

The input channels are sampled at a frequency up to CLKIN/128, each sample is a 20-
bit sample, this very high sampling rate helps reduce digitization errors and produce a more
correct output.

The ADE7756 is accessed through a data transfer protocol called SPI (refer to the
PIC18C452 section for details about this protocol). Through SPI we set the ADE7756's
different parameters to properly produce the desired output. Output on the other hand can be
read through two different ways: either using the SPI to read a 40-bit register (AENERGY) that
holds the accumulated energy (for a specific time), or using a hardware pulse counter to count
the pulse output (CF), this pulsing output is proportional to the accumulated energy. Since we
are using the PIC18C452, we shall be using the SPI method to configure the ADE7756 and
read the accumulated energy. A simple LED is connected to the (CF) signal to show the
frequency proportionality to the energy.

23
 This basic interpretation was made to simplify the ADE7756 operation. The ADE7756
goes in much more complex steps and process to produce the final output. For more detailed
information about the ADE7756 operation, refer to the datasheets on the companion CD.

2.4.2 Adapting the Input Signals

The load is connected to the ADE7756 by its
220V
voltage and current signals. The voltage channel LOAD
1
represents the line voltage. Referring back to the
datasheets; we see that there are conditions for the input Current
Sensor
voltage and current channels, the most important 220/12
condition is that the inputs don't exceed the max range of
20 1
1 volt; this condition is satisfied for both the current and 19
18
DIN RESET
DOUT DVDD
2
3
20K
SCLK AVDD
the voltage as follows: 17
16 CS V1P
4
5
15 CLKOUT V1N 6
14 CLKIN V2P 7 10K
Votage
13 IRQ V2N 8 Sensor
The voltage is dropped down from 220V to about 12 SAG
ZX
AGND
REF
9
11 10
1V in two steps; the 1st step is using a drop down CF DGND
ADE7756 100K
transformer that drop the voltage from 220V to 12V, the
2nd step is to use a voltage divider with large resistor
values to drop the voltage further to about 0.92V. Figure 2.13: Voltage and Current
Conditioning for the ADE7756
The current is not fed directly as it is to the ADE7756 but rather as a voltage
equivalent, this is made by allowing the current to flow in a small resistance (shunt) to produce
a voltage, this voltage is taken as to represent the current, and it should not exceed the 1 volt
barrier. This condition is simply made by putting a very small resistance (about 1Ω) in the way
of the current, making sure the voltage produced is less that 1 volt (for the present load). For
larger loads; smaller shunts are needed, some designs for the ADE7756 suggested a special type
shunt of 100µΩ. For the sake of demonstration; we are using a regular lamp bulb as a load.

2.4.3 Communicating with the ADE7756

Through SPI we talk to the ADE7756 using the PIC18C452, the ADE7756 is accessed
by simply writing to special registers and reading from certain register, it has 18 registers
different in length, most of them are writable but all are readable. All registers of the ADE7756
are on the datasheets.

Before reading or writing any register, we must provide the ADE7756 with the address
of that register, we use the Communication register to do that; by writing to this 8-bit register,
we say to the ADE7756 that we need to read a certain register (with the address A4A3A2A1A0) or
write to a one.

DB7 DB6 DB5 DB4 DB3 DB2 DB1 DB0

W/-R 0 0 A4 A3 A2 A1 A0
Figure 2.14: Communication Register in the ADE7756

Example: if we want to read the active energy register (AENERGY with address 02h),
we simply write 02h to the Communication register, after that we read from the SPI 5 times to
get the 40-bit AENERGY register. This register is a read only register; this means that writing
to this address (Communication register=82h) will result in nothing.

The most important register that determines the operation of the ADE7756 is the
MODE register, it is a 16-bit register that controls most of the features in the ADE7756;
features like enabling or disabling filters and choosing the sampling rate for the ADCs, we can
even choose what output we want (we can read the sampled waveforms of channel 1 and
channel2). Most of the features of the ADE7756 are listed in the next section.

24
 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 ADDR: 06H

TEST 1 DISHPF
(TEST MODE SELECTION SHOULD BE SET TO 0) (DISABLE HPF IN CHANNEL 1)
WAVSEL DISLPF2
(WAVE FORM SELECTION FOR SAMPLE MODE) (DISABLE LPF2 AFTER MULTIPLIER)
00 = LPF2 DISCF
01 = RESERVED
10 = CH1 (DISABLE FREQUENCY OUTPUT CF)
11 = CH2 DISSAG
(DISABLE SAG OUTPUT)
DTRT
(WAVE FORM SAMPLES OUTPUT DATA RATE) ASUSPEND
00 = 27.9kSPS (CLKIN/128) (SUSPEND CH1 AND CH2 ADCs)
01 = 14.4kSPS (CLKIN/256) TEMPSEL
10 = 7.2kSPS (CLKIN/512)
11 = 3.6kSPS (CLKIN/1024) (START TEMPERATURE SENSING)
SWRST
SWAP (SOFTWARE CHIP RESET)
(SWAP CH1 AND CH2 ADCs)
CMODE
DISCH2 (CALIBRATION MODE)
(SHORT THE ANALOG INPUTS ON CHANNEL 2)
DISCH1
(SHORT THE ANALOG INPUTS ON CHANNEL 1)

Figure 2.15: MODE Register in the ADE7756

2.4.4 Special Features

• Phase Compensation: While we are using a shunt as the current sensor, we can
use a current transformer (CT) to do the same job. A CT has the advantage of
isolating the primary signal from the secondary signal, but it also has the
disadvantage of producing phase errors between the two channels, this leads to
errors in the final reading. Phase errors could occur internally in the ADE7756
itself, we need a way to fix these errors if they happen. The ADE7756 can do phase
compensation in the range 0.1º-0.5º, phase compensation is made by setting a time
delay or a time advance in the channel 2 processing chain, the value of this time
compensation can bet set by writing to the register (PHCAL).

• DC Offset: We could face the problem of DC shifts in the channels after they are
sampled; such shifts add DC errors to the active power and this affects the final
output. We can get rid of this error by simply removing the DC offset from any
channel. The ADE7756 can remove the DC offset from channel 1 by filtering its
output using a high pass filter HPF1. DC offsets can happen either externally or
they can be due to the internal ADCs them self.

•Zero Crossing Indication: the ADE7756 has an output pin (ZX) that alternates
with the each zero crossing of the line voltage, the ZX signal will go logic high on a
positive going zero crossing and logic low on a negative going zero crossing on
Channel 2. Monitoring this pin will detect losses in the line voltage.

• Line Voltage SAG Detection: In addition to the detection of the loss of the line
voltage signal (zero crossing), the ADE7756 can also be programmed to detect
when the absolute value of the line voltage drops below a certain peak value
(specified by the SAGLVL register), for a number of half cycles of the voltage line
(specified by the SAGCYC register). If such condition happens; the ADE7756 will
set the SAG pin low. Monitoring the SAG pin can be used as a warning indication if
there is a fault on the voltage line. Using this feature; we can perform certain actions
as soon as possible (e.g. saving the data on an EEROM, sending a message
indicating a line failure, and so on…).

• Waveform Sampling: The default output from the ADE7756 is the active energy
read from the AENERGY register, yet we can get other outputs from the ADE7756
other than the energy; the ADE7756 can provide us with waveform sampling for
channel 1 and 2 (the current and the voltage signals), the availability of such digital
outputs widens the scope of the data we can provide for the electrical power system,
we can perform DSP on the waveforms and calculate different system parameters
like the active and reactive power, RMS values for the current and voltage, PF and
much more. Waveform samples can be taken by setting the WAVSEL two bits
MODE<14, 13> to select the desired waveform to sample and read at the output. 25
The default value is 00 which means that the output is the active energy, e.g. to read
the waveform samples of the current signal on channel 1; we set the WAVSEL bits
to 10.

•Temperature Sensor: The ADE7756 provides the user with a temperature

output, this output can also be used for performance monitoring too.

Remember
The ADE7756 will be referred to throughout the
entire text as the symbol shown. ADE7756

26
2.5 93C56, Serial EEPROM
Electricity meters currently used do not "Reset to Zero" when a power failure occurs.
Adding this feature to our digital system is a must.

Although the PIC18C452 is one of the most advanced MCUs available, it is still an
EPROM and therefore can't function as a non-volatile memory (NVR). We need the ability to
store data in the absence of the power supply feeding the digital circuitry, data such as the
accumulated energy for example. Adding this feature we can improve our system by making
sure no data is lost whatever the circumstances are. The 93C56 [7] is a Serial EEPROM that
helps accomplish that.

VCC
CS
INSTRUCTION
SK DECODER
CONTROL LOGIC
CS 1 8 VCC INSTRUCTION AND CLOCK
DI REGISTER GENERATORS
SK 2 7 NC
DI 3 6 NC HIGH VOLTAGE
ADDRESS GENERATOR
DO 4 5 GND REGISTER AND
PROGRAM
TIMER

DECODER EEPROM ARRAY

CS Chip Select 16
SK Serial Data Clock
DI Serial Data Input READ/WRITE AMPS
16 VSS
DO Serial Data Output
GND Ground
DATA IN/OUT REGISTER
NC No Connect 16 BITS

VCC Power Supply

DO DATA OUT BUFFER

Figure 2.16: 93C56, Connection and Functional Diagrams

The 93C56 is a 2048-bit serial CMOS non-volatile EEPROM; it is organized as an

array of 128X16. It can retain stored data up to 40 years with no power supply connected.
Fairchild Company manufactured the chip we're using.

The chip is interfaced through the serial peripheral interface (SPI), it is connected to the
standard 4 wires of the chip select (CS), clock (SK), data input (DI) and data output (DO)
signals. The SPI settings of the PIC should be set as to function properly with this chip.

2.5.1 Using the 93C56 EEPROM

All features of the 93C56 can be accessed through 7 instructions implemented on the
chip; these instructions are used for various Read, Write, Erase and Write Enable/Disable
operations. The instructions are shown in table 2.1.

Instruction Meaning
READ Read stored words in the 93C56. The address must be specified
Write Enable, this instruction must be issued at least once before any reading
WEN
process to take place
WRITE Store data as words in the 93C56. The address must be specified
WRALL Write All, used to fill all the 128 words of the 93C56 with the word in the data field
Write Disable, used to protect the data stored in the chip. This must be used if no
WDS
writing is to be made for a long period
ERASE Fills the specified addressed word with 1s
ERAL Erase All, used to fill all the 93C56 words with 1s
Table 2.1: Instructions of the 93C56
The 93C56 provides more than the basic Read and Write functions; other functions are
used to help maintain the system (like erasing all the data and disabling writing…). Any
function of these is accessed by sending the 93C56 the corresponding instruction data shown in
table 2.2.

27
 Since the 93C56 is a 2048-bit memory organized as 16-bit words; the address range for
the whole words stored is 0x00-0x7F. We will be using this address for accessing our data in
the 93C56.

Instruction Start Bit Opcode Field Address Field Data Field

READ 1 10 X A6 A5 A4 A3 A2 A1 A0
WEN 1 00 1 1 X X X X X X
WRITE 1 01 X A6 A5 A4 A3 A2 A1 A0 D15-D0
WRALL 1 00 0 1 X X X X X X D15-D0
WDS 1 00 0 0 X X X X X X
ERASE 1 11 X A6 A5 A4 A3 A2 A1 A0
ERAL 1 00 1 0 X X X X X X

Table 2.2: Instruction Set for the 93C56

2.5.2 Adapting Data to and from the 93C56

If noticed in table 2.2; an instruction can be done by sending the following to the chip:

• Start bit (“1”): should be issued to properly recognize the cycle

• 2-bit Opcode: to select the instruction to be issued
• 8-bit address: address for the read or write instruction, it could be a don't-care in
other instruction
• 16-bit data field: used only in the write instructions

Counting these bits will result in 11 bits or 27 bits (depending on the instruction). Such
numbers of bits don't constitute an integer number of bytes; we need them to constitute an
integer number of bytes because the SPI module of the PIC handles bytes only. This problem is
solved by introducing null zeros before issuing the start bit as shown in the figure. Such
argument is valid also when sending the data filed (not shown in figure).
Start
Bit Opcode Address Field

1 XXXXXXX XXX

Byte1 Byte2
Start
Null Zeros Bit Opcode Address Field

0 0 0 0 01 X X XXXXXXXX

Byte1 Byte2

Figure 2.17: Adapting the Instruction Bits sent to the 93C56 from the PIC18C452 to form an integer
number of bytes

Another issue that needs solving is the incoming data from the 93C56 (the addressed
word), the 93C56 sends a dummy zero bit before sending the 16-bit word to the PIC. This
additional added bit at the beginning of the data string forces the PIC to read 3 bytes from the
st
93C56 (because the total number of bits is 17), the 1 byte holds the dummy bit along with the
7 MSBs of the addressed word, and the reset of the addressed word is in the other 2 bytes. This
problem is handled in a similar manner, but further bits manipulation is involved.

Remember
The 93C56 EEPROM will be referred to throughout
the entire text as the symbol shown. NVR

28
2.6 Supplementary Hardware
We have introduced most of the major hardware modules we are going to use in the
system; however… the system still needs extra hardware to help put all these modules together.

2.6.1 Line Driver (MAX232)

Due to the voltage level difference between CMOS/TTL devices and devices using the
RS232 voltage level in the serial interface, we need a chip that can adapt voltage level between
such devices.

Devices that use the RS232 voltage levels are the PC and the Siemens M20. The PIC is
a CMOS chip that needs to be linked with these devices; the MAX232 [8] does the linking job.

Remember
The Line Driver will be referred to throughout the
entire text as the symbol shown. MAX232

2.6.2 Voltage Regulator (LM317)

While most digital devices are operating at the 5V power supply, the new trend is to
reduce power consumption by letting devices operate at a much lower voltage level; this is also
made for devices that are meant to be portable (and battery operated).

The RF transmitter FM-RTFQ1-433 used is one such device; it operates at a voltage

level of 3.3V, and all of the input signals should be adapted to this level too. We need either a
new 3.3V power supply (with its ground connected to the ground of the 5V power supply) or a
way to convert the 5V power supply to a 3.3V.
+5 V +3.25 V
Voltage regulation using the LM317 is very widely used; the
3 2
LM317 is used along with some resistors and capacitors to change the VIN VOUT
1
input voltage from 5V to 3.3V with a high output current, this is done ADJ R1
LM317 150
by following an equation that relates the connected resistors to give
the desired output. See the next equation that gives an output voltage
R2
of 3.25V using just two resistors. 240

VOUT = 1.25*(1 + R2/R1)

VOUT = 1.25*(1 + 240/150) = 3.25 V Figure 2.18: LM317

2.6.3 Open Collector Buffer

Solving the problem of devices operating at voltage levels of 3.3V, we are faced by the
problem of signals at a voltage level of 0V and 3.3V (like the data signal for the RF transmitter
FM-RTFQ1-433).

+3.25 V
We need a way to change the voltage level of
CMOS/TTL signals to any desired voltage level; this is
2.2K
done by using an open collector buffer (like the 74LS07). Signal U1A
Signal
(0 and 5 V) (0 and 3.25 V)
We just need to connect the open collector to the desired 1 2

voltage level through a resistor, put the desired signal we

74LS07
want to convert on the input of the buffer and take our
new adapted signal from the output of the buffer. Figure 2.19: 74LS07, OC Buffer

This configuration is used only for the input data driving the FM transmitter, we need
not do this for the output data coming from the FM receiver because the receiver outputs a
TTL signal be default. 29
GSM Link
3.1 The PIC and the M20
The basic idea in using GSM communications in NCNs is to receive order messages
from the utility in SMS form. The Master PIC [3] excutes the proper routine related to the
message content and passes its commands to the CCUs through the Slave PIC*. Reversly, the
system sends its replies, whether they are readings, reports or alerts, to the utility’s mainframe
again in SMS form via the GSM network. The Master PIC highly benefits from the M20
feature that it alerts for any action within the GSM network at the serial data terminal, so the
PIC can always monitor what is delivered to the M20 from network.

3.1.1 Hardware Interface

Since there should be a bidirectional communication between the Master PIC and the
M20 in the NCN, the transmitter pin (TX) of each device is to be connected to the receiver pin
(RX) of the other. However since data terminal equipment (DTE), the M20 [5] in our case,
mainly deals with TIA/EIA-232-E voltage level standards while data circuit-terminating
equipment, the PIC in this case, usually uses the CCITT V.28 standards; an extra circuitry
needs to be added to interface the PIC with the M20. The DS14C232 driver [8] realizes this
when the TX and RX pins of the USART module in the PIC are connected to those of the
M20 through the driver instead of being directly connected. Figure 3.1 shows a simplified
diagram of this connection. The full wiring of the NCN including the PIC and M20 connection
is described in Appendix A.

PIC MAX232 M20

Master PIC Driver M20

Mother Unit PIC
PIC18C425
1 40
2 MCLR/Vpp RB7 39
3 RA0/AN0 RB6 38
4 RA1/AN1 RB5 37
5 RA2/AN2/VREF- RB4 36
6 RA3/AN3/VREF+ RB3/CCP2 35
7 RA4/T0CKI RB2/INT2 34 13 12
8 RA5/AN4/SS/LVDIN RB1/INT1 33 8 R1IN R1OUT 9
9 RE0/RD/AN5 RB0/INT0 32 R2IN R2OUT
10 RE1/WR/AN6
RE2/CS/AN7
VDD
VSS
31 11
T1IN T1OUT
14 M20
11 30 10 7
12 VDD RD7/PSP7 29 T2IN T2OUT SIEMENSE
13 VSS RD6/PSP6 28 1
14 OSC1/CLKI RD5/PSP5 27 3 C1+
15 OSC2/CLKO/RA6 RD4/PSP4 26 4 C1- 1
16 RC0/T1OSO/T1CKI RC7/RX/DT 25 5 C2+ Tx
17 RC1/T1OSI/CCP2 RC6/TX/CK 24 2 C2- 2
18 RC2/CCP1 RC5/SDO 23 6 V+ Rx
19 RC3/SCK/SCL RC4/SDI/SDA 22 V- 3 4
20 RD0/PSP0 RD3/PSP3 21 16 GND Antenna
RD1/PSP1 RD2/PSP2 VCC
MAX232

Figure 3.1: M20-PIC Connection

3.1.2 AT Command Interface

As illustrated in Chapter 2, the language M20 uses to dialogue with other devices
through the serial-data terminal is the AT Hayes Command set. The criterion in which the PIC
and M20 communicate is provided here.

M20 Initialization
At the PIC startup, the PIC sends a byte representing the termination character of the
M20, which is the Carriage Return (CR) character, to cancel any previous pending command
in the M20 due to error. An ASCII character is output from the PIC as a hexadecimal code
representing that character.

Termination Character The character that must end every AT command line.
Default: CR

An optional feature of the M20 is that it can echo any character received through the
serial port. This is useful and user-friendly property when using a monitor-furnished system. 31
*The concept of Slave PIC is discussed in more details in Chapter 4
However in our system where the communication is fully machine communication the echoed
characters are redundant; thus the PIC disables this feature using the “ATE” command.

ATE Enable command echo

Command Format This command determines whether or not the M20 terminal
ATE [<value>] echoes characters received from serial port during command
state.

Parameter
<value> 0 Echo mode off
1 Echo mode on

Some GSM networks support more than one mode of SMS messages. Thus the M20
gives the user the option to select his/her mode using the “AT+CMGF” command. We only
need to use the common used one, TEXT mode.

AT+CMGF Select SMS message format

Command Format This command specifies the input and output mode of
AT+CMGF=[<mode>] messages to be used.

Parameter
<mode> 0 PDU mode
1 Text mode

The next step is to delete all messages in the memory of the SIM card installed in the
M20. This will ensure no “full-inbox’ case would take place during the M20 operation. The
delete command “AT+CMGD” should be repeated 19 times to empty all memory locations.

AT+CMGD Delete SMS message

Command Format M20 terminal deletes the SMS message stored in memory
AT+CMGD=<index> location <index>.

Parameter
<index> the index of the message storage location in the
memory. It could be in the range from 1 to 19.

That was the last step in the initializing of M20. Figure 3.2 shows the initialization
process in a glance. The M20 is now ready to receive messages and pass them to the PIC.

M20 PIC

M20 Master PIC

Startup Startup

Ready for a new Send termination

AT command character

M20 will not send Disable M20's

echoes command echo

SMS messages Set message Delete a message

are in TEXT mode mode to TEXT from M20

Message memory Yes Is it the 19th

Ready is empty message?

No

Figure 3.2: M20 Initialization

32
 New Message Reading
When an SMS message is received by the M20, a “new message” alert is sent to the
PIC’s USART receiver. The receiver interrupt flag (RCIF) is set due to this alert and the
interrupt subroutine is executed. It should be remembered that the new-message alert is not the
only string that could be received from M20, as there are many other important or unimportant
indications arriving the receiver terminal. In addition, we could be waiting another response
through the message reading or writing steps as will be shown later instead of waiting a new
message. For this reason a special response-type flag is introduced to mark for the waited string
in every step and a character-by-character scanning is to be performed to check the validity of
the received string of data referring to a pre-stored format for every step of message reading. In
the case when new message is received, the format looks like the following:

CMTI New SMS message indication

Indication Format This indication informs that the M20 terminal received a new
+CMTI: “SM”, <index> SMS message and stored it in the SIM card memory.

Parameter
<index> the index of the message storage location in the
memory. It could be in the range from 1 to 19.

The end of the alert is determined by the termination character. As mentioned before,
after loading the alert from the M20 its content is checked to ensure this is really a new-
message alert. If it passed the test the PIC sends an AT command ordering the message content
of the received index from the M20. the used command is “AT+CMGR”.

AT+CMGR Read SMS message

Command Format The M20 terminal returns SMS message with location value
AT+CMGR=<index> <index>.

Parameter
<index> the index of the message storage location in the
memory. It could be in the range from 1 to 19.

The response-type flag should be set to indicate that a message content is to be received
at the moment. The PIC is now ready to get the ordered message from the M20. The SMS
message is delivered from the M20 in the following format:

+CMGR: “REC READ”,”<number>”,”?????-????”,”<date>,<time>”<body>

Parameters
<number> The cellular phone number that sends the message.
<date> The message delivery data.
<time> The message delivery time.
<body> The message body.

In a similar manner, the content of the received string is scanned to check validity. The
body content should coincide with the protocol rules specially set for this project purposes.
This protocol is discussed in the next section of this chapter. Depending on the type of the
message, the PIC runs its proper subroutine.

In any step of the process, if an invalid character is detected the PIC starts special
invalidity subroutines depending on the type of invalidity and terminates the whole operation
by resetting the PIC.

In addition, if -for any reason- the PIC function got frozen for longer than 40 seconds it
automatically cancels the frozen task and resets itself.

The message reading process is described again briefly in figure 3.3. 33

 M20 PIC

M20 Master PIC

New
Main
Message

New message Other Check writing

interrupt Check other
alert sent process

USART
(M20)
No

Send message (i) No

New SMS? SMS Body?
content

Yes Yes

Store received Store received

Delete Message Reset parameters
data data

Valid SMS No
indication?

Yes

Assign the new Valid SMS No

Delete Message
message index (i) format?

Yes

Ready Read message of Run Protocol

Again index (i) Check

Figure 3.3: SMS Message Reading Process

New Message Sending

Whether a meter reading, quality report, or deficiency alert is to be sent to the utility, a
new SMS message needs to be constructed in the Master PIC of the NCN and turned to the
M20 which interfaces with the GSM network. Here is the way:

Unlike the reading process, the message writing process is activated by the Master PIC
itself not due to M20 alert interrupt. When the system intends to send a message to the Central
Office, first of all the message-type flag should indicate what type of message is to be sent. This
flag is set to the proper state through the PIC’s routine running. The real writing process starts
when the Master PIC sends the message-send command (AT+CMGS) to the M20.

AT+CMGS Send SMS message

Command Format The M20 terminal sends an SMS message with text in the
AT+CMGS=”<number>” <body>.
<CR><body><ctrl-Z>
Parameter
<number> the number to which the new message is sent.
<body> the body content of the sent message.

The number to which the message is sent is either picked from a received message from
the Central Office or either previously stored in the PIC as an emergency number. The
message-type flag determines the body content of the message, whether it is a meter reading
that has just been received or one of the message formats available in the PIC. The message
types and their formats will be clarified in more detail in the protocol section of this chapter.

34
 As the command format shows, after the telephone number is entered to the M20 a
termination character is send. The M20 will reply to the PIC with a stream of characters ended
by a “space” character that indicates the M20 readiness to receive the message content. For this
reason, the PIC will wait an interrupt from the M20 and check for the “space” reception before
it starts body writing. When this is done (i.e. the M20 response finishes), the PIC begins to
send the body content characters to the M20 terminal. A (ctrl-Z) character sent from the PIC
will inform the M20 that the body has ended and the PIC will send the load message to the
desired receiver. The message-sending process is reviewed again in figure 3.4.

M20 PIC

M20 Master PIC

Ready Main

Reply with a
Certain Process
“space”

A new SMS
This is the end of
message needs to Reset parameters
the message
be sent

Send a “send- Wait for the space

message” character
command to M20

No
USART
(M20) Writing Yes Is it a
interrupt response
“space”?
waited?
Other Yes

Check other

Send Send ctrl-Z

Send message Check Message-
body to Yes
Message M20 Type flag

Figure 3.4: SMS Message Sending Process

35
3.2 GSM Link Protocol

3.2.1 Protocol Main Rules

It has been mentioned in the previous section that a certain protocol was set to insure
the maximum and easiest understanding between the utility, exemplified by the Central Office,
and its NCNs spreading allover the network area. Moreover, such a protocol will make the
system more stable and secure against any distructive actions.

As a part of the protocol, any message sent from the Central Office to the NCN must be
in the following form

The Mobile Number to Reply to. 2-digit ID for the CCU Extra Optional Data Needed
It Can be of Any Length Requested to do a Task in some Tasks

Space Space

@<Reply Number> <NCN ID> <CCU ID><Task Index><Extra Data>@

6-digit Unique Security 1-digit Index for the Task

Starting Character Ending Character
Code for the NCN to Perform

of all the above-format parts, the 1-digit Task Index should be presented in more detail.
The possibilities of (Task Index) part of the protocol are shown in table 3.1, where the (NCN)
column indicates an NCN-related task and (CCU) column indicates a CCU-related task. An
“Extra” column tells if the task needs any additional values to be attached within the (Extra
Data) part of the string.

Task Index Task Description NCN CCU Extra

0 Check Operation (Check Link Status) 3 3
1 Read Meter’s Accumulated Energy 3
2 Read Power Factor 3
3 Read Switch Status 3
4 Connect/Disconnect Line 3 3
5 Reset PIC 3 3
6 Reset Slave PIC 3
7 Change Security Code 3 3
8 Change Emergency Number 3 3
Table 3.1: Task Index Description

As the table shows, three tasks would need extra data. The shapes of these information
are descriped in table 3.2.

Task Index Extra Data Description Size

4 Switch On or Off Task Indicator 1 digit
7 New Security Code for the NCN 6 digits
8 New Emergency Mobile Telephone Number 12 digits
Table 3.2: Extra Data Description

For tasks where data readings are aquired, the reply for successful tasks from the NCN
to the Central Office should be in this form

@<CCU ID><Task Index> <ExtraData>@

36
Space
 This format is applicable for Accumulated Energy , Power Factor and Switch Status
reading tasks (Task Index 1, 2 and 3). For the other tasks which include specific-action jobs,
the response of a successful job should be as follows

@<CCU ID><Task Index>@

If -for any reason- the aquired task failed to be done within a certain time, the NCN will
reply to the utility through an SMS message indicating the failure. The reply format is

@<CCU ID>F@

The “F” character means that a task related to the assigned CCU failed to be finished
due to some problem. It is then a matter of the Central Office to take the proper action in order
to solve that problem.

It should be mentioned that to access the NCN itself, the "CCU ID" field is set to "00".
In the same way, when the NCN replies on its tasks it sends the same previous ID.

3.2.2 Invalidity Reports

All our speech on GSM protocol discussed only valid code formats which is not always
the case. The NCN may receive an invalid ID, invalid task or any other type of format
invalidity. The NCN should be able to take action in any of these situations. These actions are
summarized in table. Please note the error place is marked with an undelined bold font.

Invalidity Example Action

Tel. Number @+962771F345 ID1234 151@ Delete Message (No Number)
NCN ID @+9627712345 IM1234 151@ Send “Wrong Code” Message
Task Index @+9627712345 ID1234 15A@ Send “Wrong Task” Message
Extra Data @+9627712345 ID1234 004A3@ (Not 6) Send “Wrong Task” Message
No Space @+9627712345 ID1234151@ (No space) Send “Wrong Format” Message
All This is an invalid SMS for test! Delete Message (No Number)
Table 3.3: Invalidity Action Overview

In the “Send Message” actions of the table, the bold words are the content of the
invalidity message which should be sent to the utility. It can be also noted that when we don’t
receive a valid mobile telephone number, such as the case when we receive invalid characters in
the number or the whole string is invalid, we can’t respond to the sender at all. Not only this,
but also we can’t even check wither the received message has been sent by an authorized system
user or not.

3.2.3 An Example
Let us assume this scenario. A utility’s Central Office wanted to perform these tasks:

1. Read accumulated energy of the meter connected to CCU of ID 12 in the region of

the NCN with security code ID1234. This NCN has a mobile telephone number
+96277910803.
2. Check the status of the link with CCU 15 in the same previous region.
3. Reset the Slave PIC of the same previous NCN.

The utility was able to get the meter reading successfully. However, due to a failure in
CCU 15, it can’t respond to the NCN calls. And due to a personal error, the system user
inserted the security code as (CD1234) instead of (ID1234). The conversation between the
Central Office and the NCN according to this scenario is illustrated in figure 3.5.
37
1 Send Task to
CCU 12
Read Energy in CCU 12
@+96277910803 ID1234 121@

PIC
M20 M20
CCU 12

PIC
PIC
NCN
CCU 15 Central Office
(ID 1234)

Send Reading Ok, Send to Utility

to NCN @121 00000000004F6A32672B@
Process Task
and Respond

PIC
M20 M20
CCU 12

PIC
PIC
NCN
CCU 15 Central Office
(ID 1234)

2 Send Task to
CCU 15
Check Status of CCU 15 Link
@+96277910803 ID1234 150@

PIC
M20 M20
CCU 12

PIC
PIC
NCN
CCU 15 Central Office
(ID 1234)

?
Not OK, Send Failed Task Report
@15F@

Something
PIC Wrong M20 M20
CCU 12

PIC
PIC
Can’t Receive NCN
Central Office
CCU 15 the Task! (ID 1234)

3 Reset the Slave PIC in the NCN

@+96277910803 CD1234 006@

M20 M20

PIC
NCN
Central Office
(ID 1234)

Not Ok, Send Invalidity Report

Invalid Security Code! Wrong Code

M20 M20

PIC
NCN
Central Office
(ID 1234)

Figure 3.5: GSM Link Protocol Example

38
RF Link
 When discussing the basic design to our problem’s system, the RF link was introduced
as the economic solution to interface NCNs with the relevant CCUs in their regions. This
chapter explores the RF link in more detail and the connection implementation between an
NCN and its CCUs is also described.

4.1 Need Another USART Module

4.1.1 The Slave PIC

The RF modules used; FM-RTFQ1-433 and FM-RRFQ1-433 [6], use serial
transmission to get data to be broadcasted and deilver received data, respectively. This is true
for both pairs of RF modules at the NCN and CCU sides. This fact axiomatically lead us to
state that the component which should be connected to these modules must have a serial
communication module. In our case this component is the PIC integrated in the CCU and
NCN units, which already has the serial USART module. For the CCU part this is perfect;
however, for the NCN we found a new problem since the lonely USART module of the NCN’s
PIC is already in use to interface with the M20 terminal [5]. The problem is sybolized in figure
4.1.

M20

Serial
? PIC
Serial

NCN’s PIC
Figure 4.1: The Missing Connection Mystery

More than one solution can be realized to solve this problem. The solution we selected
is simply to use another “Slave PIC” as a transponder between the “Master” PIC and the RF
modules utilizing its free USART port in this job. This is surely not the optimal solution since
there are special-purpose chips that can do this job in a relatively very cheaper cost. But we
preferred not to give this point a lot of caring since we already know that there are other types
of the PIC having two USART ports. This would completely eliminate our problem if we
didn’t plan our design on the PIC18C452 basis from the beginning.

4.1.2 The Master PIC and the Slave PIC

We now understand the need of two PICs at the NCN instead of one. But shouldn’t we
ask ourselves, if the Slave PIC is connected to the RF module through the USART, then what
is there between the Master and Slave?! This is a reasonable question, and if we go back to
Chapter 2 when mentioning some of the PIC features [3], we can conclude that the best to use
in this position is the SPI (Serial Peripheral Interface) which is available –of course– in both
PICs. As an entry to this connection details, it is worth to go over the basics of SPI one more
time. The SPI uses mainly four pins of the PIC, these are:

• Serial Data In (SDI)

• Serial Data Out (SDO)
• Serial Clock (SCK)
• Chip Select (CS)

The first two pins are the data input and output pins of the PIC. It is clear that the
output of the Master should be connected to the input of the Slave and vice versa. The Serial
Clock is the synchronization clock generated in the Master PIC and harmonizes the Slave’s
work proceeding with that of the Master. The Chip Select is a very useful feature for us, since 40
we can drive many Slaves by one Master. This pin shows its power when connecting an
ADE7756 [4] and an NVR to the NCN’s Master PIC through the SPI in addition to the Slave
PIC.

The previous four pins are not the only used to interface the Master PIC with the Slave
PIC, since there are still four more pins used in each PIC in order to maximize the control of
the Master on the system and consequently increase the system stability. These Pins are:

• New Action Interrupt In (INT1)

• New Action Interrupter
• Rx-Data Blocking
• Slave Reset

The reason behind the two interrupt pins can be clarified easily if we fully digest the
statement that the Slave PIC renders its “Slavery” role adroitly. The Master PIC can always
wake up the Slave and send to it; however, the Slave PIC can never tell the Master anything
unless it receives a license from the Master to do so. In formal words, the Master PIC selects
the desired Slave through the CS pin; this directly starts to feed the Slave with Master’s clock
through the SCK pin. Then the Master can transfer data to the Slave at any time. When this
happen, the Slave replies to the Master with the data stored in a special SPI register (SSPBUF).
For more details on SPI operation in general you can refer to the PIC’s section in Chapter Two
or to its datasheet in the attached Companion CD. The use of SPI in this part of the system is
illustrated in the coming subsection of this Chapter “How It Works”.

As mentioned, a third Rx-Data Blocking pin is used. The function of this pin is to stop
the Slave PIC from passing the data received at the RF Rx to the Master PIC. This blocking is
highly desired when data is sent through the NCN’s RF Tx to avoid “hearing echo” case which
may disturb the Master.

The last Pin connection between Master and Slave PICs is the “Slave Reset” pin. This
pin carries a command to the Slave PIC to reset itself when required. The ability to do this
gives the system a bonus feature as a fully freezing-immune system.

Figure 4.2 shows the basic connections between the Master and Slave PIC. The full
schematics are provided in Appendix A.

Interrupt
Interrupt
Clock
SDO

PIC CS
SDO
PIC
Give Data

Slave PIC Master PIC

M20
RF Module Tx Tx Rx

Rx
Slav e Master
PIC18C425 PIC18C425
1 40 Stop Data 1 40
2 MCLR/Vpp RB7 39 Reset 2 MCLR/Vpp RB7 39
3 RA0/AN0 RB6 38 3 RA0/AN0 RB6 38
4 RA1/AN1 RB5 37 4 RA1/AN1 RB5 37
5 RA2/AN2/VREF- RB4 36 5 RA2/AN2/VREF- RB4 36
6 RA3/AN3/VREF+ RB3/CCP2 35 6 RA3/AN3/VREF+ RB3/CCP2 35
CS 7 RA4/T0CKI RB2/INT2 34 INT1 7 RA4/T0CKI RB2/INT2 34 INT1
8 RA5/AN4/SS/LVDIN RB1/INT1 33 8 RA5/AN4/SS/LVDIN RB1/INT1 33
9 RE0/RD/AN5 RB0/INT0 32 9 RE0/RD/AN5 RB0/INT0 32
10 RE1/WR/AN6 VDD 31 10 RE1/WR/AN6 VDD 31
11 RE2/CS/AN7 VSS 30 11 RE2/CS/AN7 VSS 30
12 VDD RD7/PSP7 29 12 VDD RD7/PSP7 29
13 VSS RD6/PSP6 28 13 VSS RD6/PSP6 28
14 OSC1/CLKI RD5/PSP5 27 14 OSC1/CLKI RD5/PSP5 27
15 OSC2/CLKO/RA6 RD4/PSP4 26 15 OSC2/CLKO/RA6 RD4/PSP4 26
16 RC0/T1OSO/T1CKI RC7/RX/DT 25 16 RC0/T1OSO/T1CKI RC7/RX/DT 25
17 RC1/T1OSI/CCP2 RC6/TX/CK 24 SDO 17 RC1/T1OSI/CCP2 RC6/TX/CK 24 SDO
SCK 18 RC2/CCP1 RC5/SDO 23 SDI SCK 18 RC2/CCP1 RC5/SDO 23 SDI
19 RC3/SCK/SCL RC4/SDI/SDA 22 CS 19 RC3/SCK/SCL RC4/SDI/SDA 22
20 RD0/PSP0 RD3/PSP3 21 CS 20 RD0/PSP0 RD3/PSP3 21
RD1/PSP1 RD2/PSP2 RD1/PSP1 RD2/PSP2

Figure 4.2: Master-Slave PICs Connection

41
4.1.3 How It Works
The communication between the Master and Slave PIC at the NCN unit is not as
simple as it may seem for the first glance. When the Master wants to pass a command to a
CCU, it first sends an interrupt signal to the Slave PIC; this will make it ready to receive data
from the Master. A byte received from the Master is sent directly to the RF Tx. After sending
data to the CCU, the Master disables the NCN’s RF Tx to overcome the destructive effect of
two transmitters working on the same RF carrier frequency. This point will be studied in more
detail in the coming section of this chapter.

When the RF Rx of the NCN detects a signal, the Slave checks the Rx-Data Blocking
flag to know if it should alert the Master or not. If yes, the Slave sends its own interrupt signal
which will wakeup the Master’s SPI operation again. Thus, the Master PIC calls the data
received by the Slave. Once this finishes, the Master enables the RF Tx again to be ready for
the next action. The Master-Slave communication process is simplified in the flow chart below,
figure 4.3.

PIC PIC

Slave PIC Master PIC

Main Main

Other Send Interrupt to Need to send data

Check other interrupt the Slave to the RF

INT1

No Does SPI No Yes

Send data to Finish all
transfer
complete? Slave through SPI data?

Yes
Send data to
USART (RF)

Figure 4.3: (a) Master-to-Slave Writing Process

PIC PIC

Slave PIC Master PIC

Other
Check other interrupt Main Main
USART

Can alert the Yes Send Interrupt to Other

Master? the Master
interrupt Check other

No INT1

No Received data Send dummy data

via SPI? to read the Slave

Yes

Return to Send data through Receive Slave’s Process

Main SPI data received data

(b) Slave-to-Master Writing Process

42
4.2 Adapting the RF Transmitter
In their original situation, the chosen RF modules can not said to be “ready” to install
in our system, as there are some technical parameters that should be matched for the RF
modules and the rest of the system, as well as there are some technical limitations that should
be overcome. This section will be dedicated to talk about and solve specific RF transmitter
problems, while the next section will describe the solution of some problems arise due to the
real operation of the RF-module pairs within the system.

4.2.1 Power Supply

While the RF Rx (FM-RRFQ1-433) is driven by a 5-volt source (Vcc = 5) and delivers
received data to the system in 0 or 5-volt level mode, which is fully compatible with whole
system theme, the RF Tx somehow contradicts this theme. The RF (FM-RTFQ1-433)
Transmitter module needs a 3.3-volt source to supply power to its circuit, and due to this it is
also limited to receive a shrinked voltage range of (0 V – 3.3 V) only from the system which
outputs up to 5-volt signals.

For this reason, and to solve the 3.3-V power supply problem, an additional regulator
IC (LM317) is used. This regulator, with the aid of a simple resistor-capacitor circuitry
connected to it, can output a fixed voltage-level of DC signal that satisfies our purpose needs.
Figure 4.4 shows the connection between the RF Tx and the regulator and the additional
circuit components.

+5 V +3.25 V
LM317

3 2
VIN VOUT
3.25 Volt 5 Volt 1
ADJ
R1
150 C1 1 FM-RTFQ1 6
En EA
LM317 100 uF
2
3 In GND
5
4
GND Vcc

R2
240
RF Tx Regulator

Figure 4.4: RF Transmitter-Regulator Connection

The voltage level supplied by the regulator is determined using the formula

VOUT = 1.25 [ 1 + ( R2/R1 ) ]

Then using an R2 = 240 Ω and R1 = 150 Ω will lead to a VOUT value of 3.25 volts, which is
acceptable.

4.2.2 Input Data Level

As mentioned above, another problem regarding the RF Tx is that it is limited to stand
not more than 3.3-V signal level as a maximum, so it is obvious that an extra stage should be
inserted between the TTL-level pin of the PIC and the data-input pin of the RF Tx to drop
down the voltage level from 5 to 3.3 volts. We can introduce at least 3 simple approaches to
perform this stage’s function, the Open-Collector Buffer, a transistor circuit and finally the very
basic series of diodes.

Open-Collector Buffer
Using the design shown in figure 4.5 of the Open-Collector Buffer (47LS07), and
benefiting from the already produced 3.25 voltages of the regulator circuit, the maximum
output of the buffer will never exceed the supply voltage level (3.25 V).

43
 +3.25 V

R3
2.2K

U1A 1 FM-RTFQ1 6
En EA
PIC Tx (0 and 5 V)
1 2 2 5
3 In GND 4
GND Vcc

74LS07

Figure 4.5: Open-Collector Voltage Adapter

Transistor Circuit
In a similar manner of the previous approach, the transistor circuit shown in fig 4.5 and
supplied by a Vcc= 3.25 volts will always bring out a maximum of less than 3.25 V-level signal.

VCC = +3 V

9.7 k 9.7 k

V out
2.7 k
Q2
2.7 k
V in Q1

Figure 4.5: Transistor-Based Voltage Adapter

Diode Series
Although it seems to be a trivial solution when compared to the other two solutions, it
can prove itself as well-doing approach. Every diode in the series simply “cuts” about 0.7 volts
from the PIC’s signal. As much we need to reduce the signal level, as much we add extra diodes
to the series. Figure 4.6 shows this simple circuit. As our goal is not to complicate the system
structure, we used the last solution in our system implementation. However the other two
solutions were tested and worked properly.

From the PIC

1 FM-RTFQ1 6
En EA
2 5
3 In GND 4
GND Vcc

Figure 4.6: Diode-Based Voltage Adapter

4.2.3 Transmitter's Phased Locked Loop (PLL)

Reading the documentation of the RF transmitter, we notice the Initial Frequency
Accuracy parameter (-25 to 25 KHz). This parameter was not understood until the actual
implementation of the module with the system.

The internal PLL of the transmitter module needs time to lock at the right carrier
frequency (433 MHz), this needed time is not measured in seconds but in the number of
successive bytes sent to the transmitter in a row (i.e. each time we need to send data through
the transmitter, it needs to start its PLL synchronization process, thus loosing a significant
number of bytes from the first bytes sent to the transmitter).

This problem was solved by sending a row of specific dump bytes to the transmitter
before sending our own data. This dump data was tested over and over to make sure PLL
synchronization is made as soon as possible. We have found that the best dump data to send
was:

0000FFFFh 44
4.3 The PIC and the RF Modules

4.3.1 Hardware Interface

Whether it is at the CCU or NCN side, the RF-module pair is connected to a PIC
which controls its operation. As expected, the Data-Input pin of the RF transmitter should be
connected to the TX pin of the USART port of the PIC; however this does not mean a direct
connection as we have just described the level voltage conflict between both pins. In the other
side, the Data-Out pin of the RF receiver should be connected directly to the RX pin of the
PIC’s USART. Of course, for the CCU this description is clear since there is only a unique
PIC. But when talking about the NCN, we should notice that some pins are connected to the
Slave PIC while others are connected to the Master PIC. Figure 4.7 provides the part of
schematic diagram related to the PIC-RF module connection. The full schematics are available
in Appendix A.

FM1
1 FM-RTFQ1 6
En EA
2 5
3 In GND 4
GND Vcc
PIC18C425
1 40
2 MCLR/Vpp RB7 39
3 RA0/AN0 RB6 38
RA1/AN1 RB5
RF Tx 4
5 RA2/AN2/VREF- RB4
37
36
TX 6 RA3/AN3/VREF+ RB3/CCP2 35 FM2
7 RA4/T0CKI RB2/INT2 34 FM-RRFQ1
RX 8 RA5/AN4/SS/LVDIN RB1/INT1 33
PIC

GND

GND

GND

Dout
RE0/RD/AN5 RB0/INT0

Vcc
9 32

Din

NC
NC

PD
10 RE1/WR/AN6 VDD 31
11 RE2/CS/AN7 VSS 30
VDD RD7/PSP7

11
12
13
14
15
12 29

1
2
3

7
13 VSS RD6/PSP6 28
14 OSC1/CLKI RD5/PSP5 27
RF Rx 15 OSC2/CLKO/RA6 RD4/PSP4 26
16 RC0/T1OSO/T1CKI RC7/RX/DT 25
17 RC1/T1OSI/CCP2 RC6/TX/CK 24
18 RC2/CCP1 RC5/SDO 23
19 RC3/SCK/SCL RC4/SDI/SDA 22
20 RD0/PSP0 RD3/PSP3 21
RD1/PSP1 RD2/PSP2

Figure 4.7: PIC–RF Modules Connection

The Data In and Out pins of the RF Tx and Rx, respectively, are connected to the Slave
PIC of the NCN. In addition, the CCU’s PIC and the NCN’s Master PIC have control on
their relevant RF transmitters’ Enable signals. This control is a must to solve two contrary
problems; the first is that one transmitter at least must be enabled at any time; otherwise the RF
receiver will keep detecting noise and delivering it to the system as a sort of data. This is a
design limitation of the RF receiver, because a special-purpose IC called (TDA 5210) used in
the receiver has the “Carrier Detect” feature which would enable the receiver to stay at sleep
mode when no signal is broadcast from any transmitter. This is why we should always keep one
transmitter, the NCN’s transmitter, enabled to broadcast its carrier without data.

As the RF receiver can never understand from a transmitter when another one, even if
null, is enabled, the NCN must disable its transmitter when it waits data from a CCU, and only
the transmitter of that desired CCU should be enabled.

4.3.2 How It Works

When the NCN wants to communicate to a CCU, it runs the SPI routine to pass the
data from the Master to the Slave PIC. The Master sets the Rx Data-Blocking pin to disable
echo. The Slave will send every bit of data to the RF Tx. When this finishes, the Master
disables the NCN’s Tx to give the CCU to transmit its reply. At the CCU side, the Rx is always
enabled to receive data, so when data is detected it starts forwarding the received data to the
CCU’s PIC, which processes the data and outputs the proper reply to the NCN which is sent
through the RF Tx. After receiving the reply at the NCN side, the CCU’s Tx is disabled and
that of the NCN is enabled again.

45
4.4 RF Link Protocol
After the Master PIC of the NCN extracts the SMS message, it should reshape the task
included in this message to a form that can be understood at the CCU side. For this reason, a
simple RF-Link Protocol was developed to be respected in every conversation between the
NCN and any of its CCUs through the RF channel. Any task alert sent from the NCN to the
CCU must take the following format

2-digit ID for the CCU Extra Optional Data Needed

Requested to do a Task in some Tasks

@<CCU ID><Task Index><Extra Data>@

Starting Character 1-digit Index for the Task Ending Character

to Perform

The information included in the format is exactly that received from the utility, but only
those related to the CCU. When the called CCU receives the alert, it should reply to the NCN
with the same received message added to it the acquired reading if any. For example if the
meter at CCU number 12 is to be read, the NCN should send

@121@

The CCU will get the meter reading and send a message similar to the following to the
NCN.

@121 000000000010C1F368B2@

A good indication of the system’s stability is that if the CCU did not reply to the NCN’s
request within 40 second, this is considered a failure and the NCN replies to the Central Office
with a failed-task report.

@12F@

46
 Data Gathering
& Process Controlling
 After we have established a rigid base for data to be easily interchanged between the
Central Office and any CCU in the system, we now move on to the process of gathering data
for the Central Office and controlling different processes at the CCU; this is done with the help
of some special ICs and circuitry that are interfaced with the PIC.

Data collected for an electricity system ranges from the simple energy consumed by
people to the 3rd harmonics of a 3-phase power line going into a factory. Here we introduce an
example of collecting the consumed energy by a load, try to read its power factor and read the
status of a switch at the CCU. Chapter 8 should give further ideas for more data to be collected
(either using the same hardware available or by new means).

Processes to control wirelessly are endless. It ranges from controlling a simple switch to
flying an airplane. As an example, we will try to control a switch connected to a small light bulb
on and off at the CCU site, the switch represents the power line that we need to control
(connect/disconnect). This should simulate process controlling as a proof of concept for the
credibility of the system.

5.1 Data Gathering

We will try to collect the energy consumed by an electrical load at the customer side.
We will also try to gather other additional data (like the power factor and the status of a
switch). The Analog ADE7756 plays the key role in providing most of the data at the CCU.

5.1.1 Calculate Digitally with the ADE7756

Almost all old electricity meters use coils and magnets to calculate the energy, and spin
gears with numbers printed on them to give the reading. Such meters are still used until this
day here in Jordan and in most of the Arab countries. Getting a digital microcontroller to
interface such meters would require the need to convert mechanical energy into a digital stream
of 1s and 0s (theoretically it is possible), but we need not to because there are new electricity
meters that give us the ability to interface them digitally (ABB and other companies
manufacture such meters); this way we can simply get the data at the CCU and deliver it to the
Central Office.

Since we don’t have a meter that can be digitally interfaced, we will work with a very
famous chip in this field: the ADE7756 from Analog Company [4]. This chip takes an electrical
load (via its voltage and current) and calculates the consumed energy by that load. Refer back
to Chapter 2 for brief description and information about the ADE7756 or refer back to the
datasheets on the companion CD.

As a reminder, the ADE7556 uses both the voltage and current inputs to digitally
calculate the consumed energy by the load. This is simply done by first converting the voltage
and current signals into digital signals. The average power is calculated by multiplying both
signals and passing the output through a digital LPF to extract the DC content (average
power); the output is then accumulated periodically at a very high speed (integrated) to
calculate the energy.

0110010101
ADC
Currnet LPF Accumulate
Consumed
Instantaneous Average Energy
Power Power
0110010101
ADC
Voltage

Figure 5.1: Basic Operation of the ADE7756

48
 The ADE7756 is considered a very cheap solution for the AMR system. The ADE7756
represents only a data source; this data source could be anything else (e.g. a data from a system
other than the electricity system itself). The ADE7756 was used as a proof of concept.

5.1.2 The PIC18C452 and the ADE7756

The PIC18C452 [3] is the device that shall be connected to the ADE7756. The PIC
interfaces the ADE7756 through the Serial Peripheral Interface (SPI). The specifications for
the SPI used are extracted from the datasheet of the ADE7756, and the PIC should configure
its SPI settings as so (please note that the PICs in the NCN and the CUU also interface the
93C56 EEPROM [7] via SPI, and the SPI settings are different from those used with the
ADE7756, this forces the PIC to change its SPI settings each time it talks to either one of
them).

Communication between the PIC and the ADE7756 is made by requesting something
then taking the response after that; e.g. when the PIC needs to know the consumed energy by
the load; it sends 02h (Read AENEGY register) to the Communication register in the
ADE7756 to indicate so. The ADE7756 then lets the PIC read (one byte at a time) the 40-bit
consumed energy register AENERGY. Refer back to Chapter 2 or to the ADE7756 datasheets
on the companion CD.
Master
PIC18C425
1 40 220V
2 MCLR/Vpp RB7 39
3 RA0/AN0 RB6 38 1
4 RA1/AN1 RB5 37
5 RA2/AN2/VREF- RB4 36
6 RA3/AN3/VREF+ RB3/CCP2 35
7 RA4/T0CKI RB2/INT2 34
8 RA5/AN4/SS/LVDIN RB1/INT1 33 220/12
9 RE0/RD/AN5 RB0/INT0 32
10 RE1/WR/AN6 VDD 31
11 RE2/CS/AN7 VSS 30 SDI 20 1
12 VDD RD7/PSP7 29 SDO 19 DIN RESET 2 20K
13 VSS RD6/PSP6 28 SCK 18 DOUT DVDD 3
14 OSC1/CLKI RD5/PSP5 27 CS 17 SCLK AVDD 4
15 OSC2/CLKO/RA6 RD4/PSP4 26 16 CS V1P 5
SPI 16
17
RC0/T1OSO/T1CKI RC7/RX/DT
RC1/T1OSI/CCP2 RC6/TX/CK
25
24 SDO
15
14
CLKOUT V1N
CLKIN V2P
6
7 10K
SCK 18 RC2/CCP1 RC5/SDO 23 SDI 13 IRQ V2N 8
RC3/SCK/SCL RC4/SDI/SDA SAG AGND
PIC
CS 19 22 12 9
ADE7756 CS 20 RD0/PSP0
RD1/PSP1
RD3/PSP3
RD2/PSP2
21 11 ZX
CF
REF
DGND
10

ADE7756 100K

Figure 5.2: The PIC is interfaced with the ADE7756 through SPI

One of the SPI settings that the PIC must properly set so that valid SPI communication
is to be established is the Idle Clock Polarity; the datasheet of the ADE7756 specifies that the
idle state of the clock of the serial communication (SCK) should be low (0V). This condition is
met with the PIC by setting the SSPCON1<CKP> bit to 0. Other settings are set using this
register and the SSPSTAT register.

The ADE7756 is interfaced in both the NCN and the CCU. The chip-select signal is
used in the SPI connection; this is because the chip select resets the ADE7756 serial interface
and puts it in the communication mode. The chip-select signal for the ADE7756 is active low
(while it is active high for the 93C56). Appendix A contains full schematics for the NCN and
CCU.
93C56
4 3
DO DI
2
CLK M20
RF Module Tx CS
1 Tx Rx

Rx
220V
Slav e Master

PIC18C425 PIC18C425 1
1 40 1 40
2 MCLR/Vpp RB7 39 2 MCLR/Vpp RB7 39
3 RA0/AN0 RB6 38 3 RA0/AN0 RB6 38
4 RA1/AN1 RB5 37 4 RA1/AN1 RB5 37
5 RA2/AN2/VREF- RB4 36 5 RA2/AN2/VREF- RB4 36
6 RA3/AN3/VREF+ RB3/CCP2 35 6 RA3/AN3/VREF+ RB3/CCP2 35
CS 7 RA4/T0CKI RB2/INT2 34 7 RA4/T0CKI RB2/INT2 34
8 RA5/AN4/SS/LVDIN RB1/INT1 33 8 RA5/AN4/SS/LVDIN RB1/INT1 33 220/12
9 RE0/RD/AN5 RB0/INT0 32 9 RE0/RD/AN5 RB0/INT0 32
10 RE1/WR/AN6 VDD 31 10 RE1/WR/AN6 VDD 31
11 RE2/CS/AN7 VSS 30 11 RE2/CS/AN7 VSS 30 CS SDI 20 1
12 VDD RD7/PSP7 29 12 VDD RD7/PSP7 29 SDO 19 DIN RESET 2 20K
13 VSS RD6/PSP6 28 13 VSS RD6/PSP6 28 SCK 18 DOUT DVDD 3
14 OSC1/CLKI RD5/PSP5 27 14 OSC1/CLKI RD5/PSP5 27 CS 17 SCLK AVDD 4
15 OSC2/CLKO/RA6 RD4/PSP4 26 15 OSC2/CLKO/RA6 RD4/PSP4 26 16 CS V1P 5
16 RC0/T1OSO/T1CKI RC7/RX/DT 25 16 RC0/T1OSO/T1CKI RC7/RX/DT 25 15 CLKOUT V1N 6
17 RC1/T1OSI/CCP2 RC6/TX/CK 24 SDO 17 RC1/T1OSI/CCP2 RC6/TX/CK 24 SDO 14 CLKIN V2P 7 10K
SCK 18 RC2/CCP1 RC5/SDO 23 SDI SCK 18 RC2/CCP1 RC5/SDO 23 SDI 13 IRQ V2N 8
19 RC3/SCK/SCL RC4/SDI/SDA 22 CS 19 RC3/SCK/SCL RC4/SDI/SDA 22 12 SAG AGND 9
20 RD0/PSP0 RD3/PSP3 21 CS 20 RD0/PSP0 RD3/PSP3 21 11 ZX REF 10
RD1/PSP1 RD2/PSP2 RD1/PSP1 RD2/PSP2 CF DGND
ADE7756 100K

Figure 5.3: Master PIC in the NCN Interfaces the ADE7756, the Slave PIC and the 93C56 Using SPI. Each device
has its Own Chip Select Signal. There are Other Control Signals between the Master and the Slave not Shown
49
 The NCN also uses SPI to interface the Slave PIC (to use its USART to communicate
with the CCUs as explained in Chapter 2) alongside with the ADE7756 and the 93C56. A
separate Chip Select signal is given to each device.

5.1.3 Calculating the Consumed Energy

The steps to go through for calculating the consumed energy using the ADE7756 are:

A. Setup the ADE7756

B. With the PIC, keep reading the RSTENERGY register every fixed period of
time and accumulate the reading in a larger register
C. Convert the reading to energy, This step needs the ADE7756 to be calibrated

The final accumulated reading is just a null number that must be converted into a watt-
hour reading. Tthis is done by multiplying it by a constant figure (We will name this constant
the Energy Constant. It has the units of watt-hour/LSB).

This energy constant is the output of the Calibrating process of the ADE7756; it is a
process that determines what does a LSB in the AENERGY register equals as watt-hour
energy. We will introduce the calibration process soon.

A. Setting the ADE7756

The ADE7756's basic operation mentioned in the beginning is BASIC. There are still
more steps and operations to go through to yield the final output. These steps need
configurations and settings to be made, this is done through writing to some registers in the
ADE7756 (mainly the MODE register). We will mention some of these registers and the values
assigned to each.

• MODE (16-bit): This register controls most of the settings (refer to chapter 2 for
details); we will set it to zero (0000h). This setting does the following: select the output
to be the active energy, the sampling rate to the maximum (CLKIN/128), enable HPF
in channel 1 (to remove any DC offset errors), enable LPF2 after the multiplier (the
result is the average power), and enable frequency output at CF. See the figure for more
information.
15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ADDR: 06H

TEST 1 DISHPF
(TEST MODE SELECTION SHOULD BE SET TO 0) (ENABLE HPF IN CHANNEL 1)
WAVSEL 00 = LPF2 DISLPF2
(ENABLE LPF2 AFTER MULTIPLIER)
DTRT (WAVE FORM SAMPLES OUTPUT DATA RATE) DISCF
00 = 27.9kSPS (CLKIN/128) (ENABLE FREQUENCY OUTPUT CF)
SWAP DISSAG
(DON'T SWAP CH1 AND CH2 ADCs) (ENABLE SAG OUTPUT)
DISCH2 ASUSPEND
(DON'T SUSPEND CH1 AND CH2 ADCs)
(DON'T SHORT THE ANALOG INPUTS ON CHANNEL 2)
TEMPSEL
DISCH1 (STOP TEMPERATURE SENSING)
(DON'T SHORT THE ANALOG INPUTS ON CHANNEL 1)
SWRST
(USED FOR SOFTWARE RESET)
CMODE
(USED IN CALIBRATION MODE)

Figure 5.4: Setting the MODE Register in the ADE7756

• SAGLVL (7-bit): Sag Voltage Level. This register determines at what peak signal level
on channel2 (line voltage) the SAG pin will become active (set to 0). The signal must
remain low for the number of cycles specified in the SAGCYC register before the SAG
pin is pulled low. We set this register to zero or to any small value; this means that the
SAG pin will only go low if the voltage level is zero or very small (i.e. no voltage or line
failure).
50
 • SAGCYC (7-bit): Sag Line Cycle Register. This register specifies the number of half
line cycles the signal on channel2 must be below SAGLVL before the SAG output is
activated. We set this register to 06h, indicating that 6 half line cycles must happen
before setting the SAG pin low at a SAG event.

• APOS (12-bit): Active Power Offset Register. This register contains a constant energy
offset that will always be added to the accumulated energy in the AENERGY register,
this register helps in correcting errors that cannot be corrected by any of the other error-
correction registers (e.g. phase compensation, channel1 and 2 offsets). To calculate the
value in this register is simple and direct; sixteen LSBs (APOS = 010h) written to the
APOS register are equivalent to 1 LSB in the Waveform Sample register.

11 APOS [11:0] 0

CURRENT
CHANNEL LPF2 WAVEFORM [23:0]
23 0

VOLTAGE 39 AENERGY[39:0]
0
CHANNEL ACTIVE POWER
SIGNAL P

4 WAVEFOR
T
CLKIN REGISTER
OUTPUT LPF2

VALUES WAVEFORM REGISTER VALUES ARE

ACCUMULATED (INTEGRATED) IN
THE ACTIVE ENERGY REGISTER

TIME nT

Figure 5.5: APOS register added to the Total Accumulated Energy

There are still other registers to set. Examples are the CH1OS (channel 1 offset),
CH2OS (channel 2 offset), APGAIN (active power gain) and other registers. Such registers
depend on the calibration process mentioned ahead, this means that the values of such registers
could change from one ADE7756 to another.

All these registers are set by the connected PIC. Again, setting the registers is made by
writing to the Communication register the address of the desired register to write to, then
sending the value to that register. The following listing is part of an assembly code that shows
how to set the MODE register to 0040h (software-reset the ADE7756).

; Setting the MODE register to 1808

movlw 0x86 ;
Assembly Code

movwf CommReg ; Save 86h to the CommReg (it is a temp register in the PIC)
call ChipSelectADE ; Chip select the ADE7756
call SendComm ; Call the "Send Communication Register" subroutine
movlw 0x00 ; Send 0040h to the ADE (2 bytes)
call SendADE ;
movlw 0x40 ;
call SendADE ;
call ChipDeselectADE; Chip deselect the ADE7756

Listing 5.1: Assembly Code for Setting the MODE register to 0040h

B. Accumulating the Consumed Energy with the PIC

Although the accumulated energy in the ADE7756 is put in the AENERGY register, we
are reading the RSTENERGY register instead; this register is a copy of the AENERGY
register. The difference is that the RSTENERGY will reset the accumulated energy reading to
zero after each reading process.

The RSTENERGY register can be simply read by writing 03h to the Communication
register. We then keep accumulating this reading in a larger 80-bit register in the PIC. We will
call this new 80-bit register the Total Accumulated Energy register (TAE).

51
 At full load (inputs are at max range = 1V); the ADE7756 could take up to 5 seconds to
fill the RSTENERGY 40-bit register (depending on the ADE7756 settings), this means that we
should read the energy in this register before it is full and data is lost; this is done periodically
by the connected PIC every predefined time elapse.

We have set the PIC to interrupt every 2 seconds (using the Timer0 module). On each
interrupt; the PIC will read the energy from the ADE7756 and accumulate the reading into the
TAE register. The following workflow diagram shows part of the PICs operation when
connected to the ADE7756.

ADE7756 PIC

Calculate
Main
Energy

No Comm. Other
Interrupt Check Other
Reg?

Timer0
Yes
Interrupt

Check Comm. Send 0x03 to

Register Comm. Reg.

Read 40-bit
Ready Data for Add to 80-bit
Active Energy
the PIC TAE Register
(RSTENERGY)

Figure 5.6: the PIC Accumulating the Energy from the ADE7756

Every 40 seconds, the PIC saves the TAE register in the 93C56 EEPROM (along with
other data); this is done to insure no data is lost at a power down situation. At startup, the PIC
loads this stored energy in the 93C56 and keeps accumulating the ADE7756 energy over this
reading.

NVR PIC

Main

Other
Interrupt Check Other

Timer0
No Interrupt

Store TAE Yes 40 Sec Accumulate

Register in
93C56 EEPROM Passed? ADE7756 Energy

Figure 5.7: the PIC Stores the ATE Register in the 93C56 EEPROM Every 40 Seconds

The workflows shown here add to the whole software implementing the system in both
the NCN and the CCU. Chapter 7 should contain workflows for the entire system. All software
files used to program the PICs can be found on the companion CD.

C. Calibrating the ADE7756

The last step in calculating the energy from the ADE7756 is to convert the value
received from the TAE register into a more meaningful reading of watt-hour energy. This is 52
done by multiplying the output by the energy constant; it has the units of (watt-hour/LSB). The
process of calibrating the ADE7756 will yield this energy constant.

The process of calibrating the ADE7756 is simple; we apply a fixed load (known watt-
hour load) at the inputs of the ADE7756 for a fixed period of time, then we read the
accumulated energy from the PIC. This reading represents the watt-hour consumption of the
load for that period. An example will illustrate.

Example: If we put a lamp bulb (energy = 60 watt-hour) at the inputs of the ADE7756, and after an hour
we read the TAE register from the PIC and found it to be 000000000013B4BF8E26h, then the Energy
Constant for that particular ADE7756 will be:
Example

Energy Constant = (Energy/TAE) = 60d/13B4BF8E26h = 7.0891122812e-10 watt-hour/LSB

• If (at normal operation) we find that the TAE register reads 0000000004B4ACE7AF32h, then the
consumed energy can be found like this:

Energy = TAE*(Energy Constant) = 4B4ACE7AF32h*7.0891122812e-10d= 3667.9361064661 watt-hour

Listing 5.2: Calculating the Energy Constant, an Example

Although this calibration process seems easy, it still needs a programmed PIC with the
human interface software to do it. This is why we have developed a calibration program that
helps the user calibrate his ADE7756.

The calibration software is very simple; it lets the user write to and read any register in
the ADE7756 using a normal PC. The PC is connected to the PIC through the serial port
(USART module in the PIC).The program also accumulates the energy from the ADE7756
every 2 seconds, relieving the user from manually doing so. The user only has to set the
different registers of the ADE7756, and the final output will help calculate the energy constant.

HyperTerminal can be used to calibrate the ADE7756. Just connect the PC to the PIC
at a baud rate of 19200 bps. The process of writing any value to the ADE775d is like this:

• Enter the 2 bytes you want to write to the ADE7756 register

• Enter W (Write)
• Enter the address of the register you want to write to, follow the table below

Add. Registers Read after an (R) # bits Registers Written after a (W)
1 WAVEFORM 24 bit Invalid (Read Only)
2 AENERGY 40 bit Invalid (Read Only)
3 RSTENERGY 40 bit Invalid (Read Only)
4 STATUS 8 bit Invalid (Read Only)
5 RSTATUS 8 bit Invalid (Read Only)
6 MODE 16 bit MODE
7 CFDIV 12 bit CFDIV
8 CH1OS 6 bit CH1OS
9 CH2OS 6 bit CH2OS
A GAIN 8 bit GAIN
B APGAIN 12 bit APGAIN
C PHCAL 6 bit PHCAL
D APOS 12 bit APOS
E ZXTOUT 12 bit ZXTOUT
F SAGCYC 8 bit SAGCYC
G IRQEN 8 bit IRQEN
H SAGLVL 8 bit SAGLVL
I TEMP 8 bit Invalid (Read Only)
J Users stored 2 bytes 16 bit Reset user bytes to 0000h
K TAE 80-bit Reset the TAE register
Table 5.1: Calibration Program, Registers Addresses
53
 Addresses J and K are not registers on the ADE7756, they are registers on the PIC. J is
the 2 byte register the user stores his data in, and K is the 10 bytes TAE register. Any address
entered outside the above range will be ignored. Please note that the calibration program is case
sensitive.

Example: if you want to set the MODE register to 1808h, just type 1808, W, 6.
Example: if you want to read the SAGCYC register, just type R, F.

The data going to the ADE7756 is right justified; this means that if you enter 1506, W,
F. this will fill the SAGCYC register with the value 06h. The schematic diagram for the
calibration hardware is shown in appendix A.

5.1.4 Calculating the Power Factor

The power factor of the load is another type of data we will try to deliver to the Central
Office. Power factor can be calculated in many different ways, one way is by using DSP. Since
we can read the waveforms of the current and voltage channels from the ADE7756; we can do
any mathematical process with this discreet data and find our power factor (we can also find the
RMS values of the current and voltage, the reactive power and much more information about
the line), but as this DSP seems easy to do, the PIC18C452 can't do it.

The PIC18XXXX family is the newest family of MCUs from Microchip, and support
for this family is not yet fully complete. Older MCU families (like the PIC17XXXX) have full
support from Microchip. Support includes publishing libraries to be used with the MCU (like
the math libraries we need). Our PIC18C452 doesn't have any library from any kind, and
trying to build a toolbox such as the math library from scratch would be tedious and very hard.

Yet, the power factor can still be found using the PIC and the ADE7756 with no real
DSP involved. This is done by finding the phase difference between the current and voltage
signals going into the ADE7756.

We shall use a simple algorithm that scans for the zero crossing of both the current and
the voltage. When the zero crossing of one channel is found; we start a counter that stops at the
zero crossing of the other channel. The value in this counter represents the phase difference
between the two channels. This counter is converted to a phase difference in degrees, and the
cosine of this phase difference is calculated to be the power factor.

Start Counter

Stop Counter

Current Voltage
(Channel1) (Channe2)

Figure 5.7: Power Factor Calculation

We can read the current and voltage signals from the ADE7756 using WAVEFORM
register. The data source sampled in this register is selected by data bits 14 and 13 in the Mode
Register (WAVSEL bits).

• To read the current (channel1) waveform; set the WAVSEL bits to 10

• To read the voltage (channel2) waveform; set the WAVSEL bits to 11

54
 The WAVEFORM register is a 24-bit two's complement register; this means that the
24th bit determines whether the total value in the register is positive (0) or negative (1). This is
how we scan for zero crossings in each signal; we just keep monitoring the last bit in the
waveform.

The scanning process here is not made as the scanning process for a new incoming
message from the M20 (i.e. to store the whole data in a memory bank then scan it); this is
because the number of samples to store here is very large and there is no enough memory space
to save them all. Instead, we will do real time scanning for the samples; each sample (from each
channel) will be scanned immediately after it is read.

5.1.5 Reading the Status of a Switch

This last data we are delivering is only for proving that the system can deliver any data
available at the CCU. It is a simple user-controlled switch tied to the input of an I/O port of the
PIC. We will read the switch status for the CCU units only.

A similar switch is connected to the PIC in the NCN, but it functions as an alarm
system that will automatically send a message using the Emergency Number stored in the
93C56. This switch represents any major parameter change the Central Office would like to
know about ASAP (e.g. power down, low power factor, system manipulation, theft…). The
NCN will alert the Central Office once this switch is changed from off (logic 0) to on (logic 1).

As for the connection of the switch with the PIC; we can't just use a simple Single-Pole
Double-Throat (SPDT) switch, this is because the transient period will disturb the normal
operation of the PIC. Instead, we will use a very simple way to gain a very stable switch. The
following schematic explains.

PIC18C425
1 40
2 MCLR/Vpp RB7 39
3 RA0/AN0 RB6 38
4 RA1/AN1 RB5 37
5 RA2/AN2/VREF- RB4 36
6 RA3/AN3/VREF+ RB3/CCP2 35
7 RA4/T0CKI RB2/INT2 34
8 RA5/AN4/SS/LVDIN RB1/INT1 33
9 RE0/RD/AN5 RB0/INT0 32
10 RE1/WR/AN6 VDD 31
RE2/CS/AN7 VSS
11
VDD RD7/PSP7
30 A
12 29 4 3 2 1
13 VSS RD6/PSP6 28
OSC1/CLKI RD5/PSP5
14
OSC2/CLKO/RA6 RD4/PSP4
27 B
15 26
16 RC0/T1OSO/T1CKI RC7/RX/DT 25
17 RC1/T1OSI/CCP2 RC6/TX/CK 24
RC2/CCP1 RC5/SDO
18
RC3/SCK/SCL RC4/SDI/SDA
23 C
19 22
20 RD0/PSP0 RD3/PSP3 21
RD1/PSP1 RD2/PSP2

Figure 5.8: Simple Implementation of a Switch

Proving the validity of this switch is very simple, first assume the switch is at point (A),
thus making the final input to the PIC a 0. If the switch is to be moved to (B); then at the
transition point (C) the 0 signal at the PIC side will still be latched due to the feedback of one
inverter to the other. By reaching point (B), the circuit will force the new signal to be a 1. This
simple solution solves the problem of transients due to wires not connected to anything.

5.2 Process Controlling

An AMR system could be the part responsible for the physical disconnection of the
power line if someone didn't pay his bills. This is a simple example for a process the Central
Office should be able to control. We can think of other processes like sending some sort of an
indication to the customer that he/she is absorbing to much current or a bill is overdue, the
number of ideas is endless.

For the sake of demonstration; we will wirelessly control a switch connected to a regular
DC lamp bulb. The switch is a transistor-based switch were a CMOS signal from an I/O port of
the PIC is going to control the switching. See the figure.
55
 +5V

PIC18C425
1 40
2 MCLR/Vpp RB7 39
3 RA0/AN0 RB6 38
4 RA1/AN1 RB5 37
5 RA2/AN2/VREF- RB4 36
6 RA3/AN3/VREF+ RB3/CCP2 35
7 RA4/T0CKI RB2/INT2 34
8 RA5/AN4/SS/LVDIN RB1/INT1 33
9 RE0/RD/AN5 RB0/INT0 32
10 RE1/WR/AN6 VDD 31 LAMP 1
11 RE2/CS/AN7 VSS 30
12 VDD RD7/PSP7 29
13 VSS RD6/PSP6 28
14 OSC1/CLKI RD5/PSP5 27
15 OSC2/CLKO/RA6 RD4/PSP4 26
16 RC0/T1OSO/T1CKI RC7/RX/DT 25 C945
17 RC1/T1OSI/CCP2 RC6/TX/CK 24 2.7K
18 RC2/CCP1 RC5/SDO 23
19 RC3/SCK/SCL RC4/SDI/SDA 22 2N3053
20 RD0/PSP0 RD3/PSP3 21
RD1/PSP1 RD2/PSP2

Figure 5.9: DC Lamp Switch

This DC lamp needs about 300 mA of current to lighten, thus a special transistor
(2N3053) is used. This transistor handles large currents than the usual ones. Another regular
BJT transistor is used as the driving base for the above transistor, thus forming a Darlington
pair. The diodes are used because the lamp operates at a lower voltage level than the 5V.

The Central Office controls the switch wirelessly (as everything else), but when a power
failure happens and no power supply is feeding the digital circuits, the status of this switch is
saved in the 93C56, and in the next startup the switch will be set to the last state the Central
Office desired.

As simple this switch-control process seems; it forms the basic idea for controlling a
state of an AC line at the customer side.

56
Central Office
6.1 What is a Central Office?
After a whole communication system was established and data was gathered, we need to
use this communication system in order to get the data and to control the processes. This is the
job of the Central Office.

The Central Office is the main harvesting node that collects data from all NCNs (and
CCUs) in the system and manages process controlling too. The automation of the system
should mainly be implemented in the Central Office.

So, in its simplest form, it could be a software program installed in the electricity utility.
It should be connected to a database to store various data and transactions from different nodes
in the system. The computer should interface a GSM device to enable sending/receiving data
from/to NCNs in the AMR system.

6.2 System Overview

We have developed a very simple software
program to simulate the operation of the Central
Office; no database connection was made because this
was developed for the sake of demonstration only.
The development of the software was made with
Visual Studio.NET and the targeted platform was
Windows XP, this is why all screen shots in this
chapter are taken from a Windows XP environment.

Figure 6.1: Central Office

The software has the basic features of sending and receiving all kinds of tasks, it also
helps to setup the different parameters needed for the system to operate correctly (like the
COM port settings and the mobile settings), plus debugging errors faced (especially when
trying to link with the GSM device). Below is a snapshot of the main page in the Central
Office, we will explain every section (feature) of the software ahead.

Figure 6.2: Central Office Main Form

58
 Since we have only one M20 device; we needed another GSM device to be interfaced
with the Central Office in order to access the GSM network and establish a connection with the
NCN units. The Nokia 8210 mobile phone was used to accomplish this.

Nokia 8210 is not a GSM modem (it cannot be accessed directly with regular AT
commands). An interpreting medium was needed to establish the connection with the mobile.
This problem was solved using the Nokia Data Suite; this utility installs a virtual modem (with
a new COM Port number) on the Host Computer. This virtual modem is the interpreting
medium used to communicate with the Nokia 8210 (Connected to a physical port).

Virtual Physical
COM COM

Nokia Specific
Central AT Commands
Nokia Data Commands
Office Suite
Nokia 8210

Figure 6.2: Nokia Data Suite is used to Communicate with the Nokia 8210

With the Data Suite; the 8210 can be accessed using special AT commands (which are
very similar to those used with the M20, this is because AT commands are standardized).

6.3 System Walkthrough

The following figure explores the main areas of the IDE for the Central Office and gives
comments on them.

Menu Bar, Use it to exit the User Control Area, here the user
program or to see the Help About performs actions and sees results

Navigation Tabs, use these tabs to

navigate other pages in the Central
Office, these pages contain most of
the system features

Port Status, this shows the

number of port opened, or
indicates if no port is opened

Figure 6.3: Overview of the Central Office IDE

We will introduce the main features of the Central Office system as walkthrough steps.
We shall introduce the following features:

• Initializing the Software

• Sending Tasks
• Receiving Tasks
• Debugging and Troubleshooting
59
6.3.1 Initializing the Software
At startup, most of the software controls are disabled; this is because the Central Office
hasn't been initialized yet. Initializing the system is about setting up the system parameters
properly in order to establish a healthy communication with the Nokia 8210. The figure below
shows the Settings tab; here we initialize the system by the steps mentioned below.

Figure 6.4: Initializing the Central Office from the Settings Tab

1. Connect the Nokia 8210 to any COM port on the computer

2. Launch the Central Office

3. Select the tab

4. in the COM Port Settings, select the COM port number (which is the number of the
virtual COM port for the Nokia Data Suite)

5. Click the button (it will be enabled once a COM port number is selected)
*If the COM port is opened (the status bar indicates so), all controls in the system will be
enabled. Otherwise, a message pops showing that the desired port can't be opened.

6. To initialize the Nokia 8210; in the Mobile Settings click the button, this
will software-initialize the Nokia 8210 (and the Nokia Data Suite) by sending specific
AT commands in a row to the mobile (this is similar to initializing the M20 with the
PIC). Please note that not clicking this button will result in invalid messages read from
the mobile. The system is now partially initialized, but there are still more steps to go
7. The Task Reply Number field must be filled with a valid mobile number (this is the
number to which the NCN will reply)
8. the Default Message Index combo box determines the index of the incoming message to
read (the Nokia 8210 doesn't give indication for new messages arrival like the M20, so
the user has to manually click a "Read Task" button when a message arrives. This
scenario was faced because of the use of the Nokia 8210 and it will not happen if an
M20 was used)

Now the Central Office is fully initialized and ready to Send Tasks (our next section)
60
6.3.2 Sending Tasks
After we have initialized the Central Office; we move on to sending tasks to CCUs in
the system, this is done in the tab as shown below.

Figure 6.5: Sending Tasks from the Central Office Tab

The Send Task group box holds all task information that must be sent to an NCN
(these information makeup the SMS that will be sent to the NCN). All tasks to send are sent
from here.

1. Specify the NCN Security Code in its respective field. It is a 6-digit code
2. Specify the NCN Mobile Number in its respective field; it is the number of the M20 at
the NCN side. It is a numeric field of any length
3. Specify the CCU ID in its respective field; it is the ID for the specific CCU the user
wishes to access. It is a 2-digit ID
4. From the Required Task combo box, select the required task to perform. Each selected
task opens its own specific task box for the user (this task box contains either notes
about the task or further information needed to perform the task). figure 6.6 shows all
tasks available along with their task boxes
5. After selecting the required task and specifying any further information needed by that
task (if any); the task is now ready to be sent to the targeted NCN. Sending the task is
made by simply clicking the Send Task button (you won't miss it)

Please note that all controls in the Central Office have validation criteria assigned to
them, e.g. the user can't input 07788G4717 in the NCN Mobile Number field, the software
will not allow him to send the task unless the number field is corrected and filled with numbers
only. This applies for most of the fields in the system.

61
 Figure 6.6: Available Tasks and Their Task Boxes

As an example; the task information shown in figure 6.5 will send the following SMS to
the M20 with mobile number 077884717:

Reply Number Specified

2-digit CCU ID
in the Settings Tab
Connect Line

@ 077910803 ID1234 5541 @

6-digit NCN Task Index

Security Code (Connect/Disconnect
Line)

To view this actual sent message; the software allows the user to see all transactions
made between the Central Office software and the Nokia 8210. All these transactions can be
viewed in the tab under the Data Sent to the Mobile group box (we will see this
feature in the Debugging and Troubleshooting section).

6.3.3 Receiving Tasks

When the CCU or NCN performs the required task, the NCN should reply to the
Central Office with a message showing the status of the performed task. Here we will read this
message and analyze it to extract the information sent from the NCN.

As mentioned before; the Nokia 8210 doesn’t give indication about the arrival of new
messages. This leads us (for the sake of demonstration) to supply the user with a "Read Task"
button along with the "Default Message Index to read". Such situation -of course- should not
happen in a real implemented system.

When the Nokia 8210 indicates the arrival of a new message (i.e. to ring and the user
hears it, not to send an AT statement); the user clicks the "Read Task" button. The Central
Office sends the appropriate AT command to read the message (with the Default Message
Index specified by the user). If a message is successfully read, the Central Office begins
62
scanning the message. The output of this scanning process is displayed in the Receive Tasks
group box. The figure below shows a valid response for the task sent in the previous section.

Figure 6.6: Receiving Tasks with the Central Office

The Received Task Details shows the details of the message received; it shows the CCU
ID, Task Index and the Extra Data (if any) received from the NCN unit. The example shown
in the figure above indicates that CCU 55 has received the task of switching on the line and it
did perform this task.

The Received Task Details also indicates an unrecognized received message or a no-
message-received situation. This may happen if either the specified Default Message Index is
wrong or if the mobile is not properly connected to the COM port. The following will be
displayed to the user:

Figure 6.7: Unrecognized Messages and No Message Read conditions

The message analysis procedure is simple and straightforward. In a real system scenario;
the output of this message analysis process should drive other data-based subroutines and
procedures to maintain the CCUs' data and transactions.

63
 All valid analyzed tasks are stored in a log file for demonstration only. This log file can
be viewed from the tab.

Figure 6.9: Logfile Holds All Valid Tasks Received

6.3.4 Debugging and Troubleshooting

The basic operation of the Central Office is complete as explained in the previous
sections. This section presents some further useful features implemented in the Central Office
that help the user debug errors and troubleshoot them.

All transactions made between the Central Office and the Nokia 8210 mobile can be
viewed by the user, these transactions help determine errors affecting the system operation. All
data sent from the Central Office and data received from the Nokia 8210 can be viewed from
the tab.

Figure 6.10: The Debug View Helps Determine Errors in the System
64
 If it helps to overtake the communication between the Central Office and the mobile to
solve errors, then the user can use the "Send Custom AT Command" feature in the Mobile
Setting group box. This lets the user to force specific AT commands to be sent to the mobile.

Figure 6.11: The Send Custom AT Command.

The button is used to empty unwanted messages from the mobile. It

uses the message index specified by the combo box to delete the specified message.

Emptying unwanted messages avoids errors due to messages waiting in queues to arrive
to the mobile.

65
Total System Implementation
 As we have just finished exploring the AMR management software and its cool user
interface in the previous chapter, we are now ready to reorganize our knowledge on the
different stages of the system studied through this report chapters as individual subsystems, and
start looking to these subsystems as a one full system that functions to achieve the goal it is
designed for. The first part of the chapter combines the three stages of the system, the Central
Office, the Network Control Node (NCN) and the Cell Control Unit (CCU) together and
presents a brief review of the main features and functions of every stage, but this time from the
whole system point of view. The last section simulates a real case of the system operation in an
example that would make the reader more familiar with the Fixed-Network AMR technology in
general, and with our own implementation of this technology-using system in particular.

7.1 System Startup

The whole system startup is a fundamental part of its operation as a step that if made
carefully and accurately will ensure a stable behaviour of the system all the time. The startup
procedure should take place at the first time the system is brought to job. Moreover, every
system stage must be ready to perform its own startup procedure on any component that seems
to fail or restart itself completely when needed but also ensure that it could resynchronize its
operation with the whole system after every startup.

7.1.1 NCN Startup

We do not need to repeat that the PIC is the “Heart” and “Brain” of every hardware
stage in our system. This statement will obviously lead us to understand why it is the device
responsible for the startup of the stage components in the NCN. Actually the Master PIC not
only plays the main role in setting up the other components of the NCN, but can also
command the Slave PIC to restart itself at any time. The Master PIC startup procedure
includes the following steps:

• Variables and constants defining

• Indication flags resetting
• I/O ports setup
• SPI setup
• USART setup
• Timer setup and resetting
• Interrupts setup
• Read previous state from the 93C56
• ADE7756 setup and resetting (if applicable)
• M20 setup

The full description of every step was fulfilled in earlier chapters and is not part of our
discussion as we only summarize the main ideas in this chapter.

7.1.2 CCU Startup

The PIC of the CCU operates the basic startup procedure, as well as it resets all the
ADE meters connected to it, setups them and makes them ready to calculate and accumulate
energy.

7.1.3 Central Office Startup

To provide full communication between the utility and the system’s network
components, the AMR management system software should be properly installed in the utility’s
computer terminals. The software should be carefully tested and all setting should be made
before logging in to the network. This includes communication port setting, default reply
67
number and many other parameters.
7.2 System Operation
After startup; The NCNs, CCUs and the Central Office software are all waiting the
system-user command to perform the task in request. Not only this, but also the system may be
brought to work by an emergency alert sent by the NCNs.

Regardless of the task type, the system user should start with inserting the mobile
telephone number and the Security Code of the NCN to be called and the CCU number
within the chosen NCN region. Then the desired task is selected and any extra data needed to
accomplish that task is fed to the system. A “Send Task” button click will build the SMS text
message that contains all the information above.

Only one NCN’s M20 terminal can receive the utility’s sent message. This will wakeup
the Master PIC to read the message, extract it, check its validity and determine whether the
task held by the massage is an NCN or CCU-based task; therefore there are two routes to
follow here.

• When an NCN-based task is received, the Master PIC performs it by itself. This
includes the NCN Security Code updating, Emergency Contact Number updating,
resetting the Master or Slave PICs. In addition, the Master PIC may be controlling its
own meters cell, so in this case if any command concerns any of these meters, the
Master PIC will do the job as if it is a CCU PIC. After the task is done, the Master PIC
enforces the M20 to reply to the Central Office either with the data acquired in the
system-user message if any, or just with a message confirming that the task has been
successfully finished.

• When a CCU-based task is received, the Master PIC converts the task from the GSM-
Link protocol format to the RF-Link protocol format. This format, as previously
illustrated, contains only the CCU number and the task index with any additional
information needed by the CCU to perform the task. The rebuilt task is then passed to
the Slave PIC of the NCN which will broadcast the new-task alert through its RF
transmitter to all CCUs in the range. Activating the RF transmission from the NCN
automatically disables the Slave-PIC response to any data received at the RF receiver,
and once transmission is completed the NCN’s RF transmitter is disabled. The
technical reasoning of these two steps is fully discussed in Chapter 4. Although all
CCUs’ RF receivers will detect the new-task alert sent by their mother NCN; only one
CCU, whose number is carried in the alert, is able to execute the requested task. This
task may be a CCU-link check, Total Accumulated Energy reading, power-factor
reading, power-line status check and power-line switching on or off. Similar to the
NCN operation, if a data-reading task is performed; the CCU number, the task index
and the read data are sent to the NCN through the enabled-on-request RF transmitter
of the CCU. Otherwise only a confirmation of task accomplishment is sent to the NCN.
The NCN’s RF receiver will forward the CCU message to the Slave PIC which passes it
again to the Master PIC. The latter reshapes the message to be consistent with the
GSM-Link protocol format, and finally the M20 terminal will send the message to the
number received within the utility’s task message content. At the utility, the Central
Office is sensitive for new messages. It reads and extracts these messages and runs the
proper operation depending on the data received from the NCNs.

In emergency cases at the NCN, such as power-down, the NCN itself alerts the utility
on this emergency. This is done through an SMS message sent by the M20 terminal.

Figure 7.1 summarizes the different system operation cases. The example presented in
the next section is to clarify this operation.

68
 To Main
PIC

Master PIC
NVR
Main
No

PIC Store TAE Yes 40 Sec

Register in
INT1 Timer0 93C56 EEPROM Passed?
Slave PIC interrupt

INT1 USART U SART

interrupt A ccumulate
Send Dummy ADE7756
Data to R ead Energy
Slave Data
New Data to
New Data
Send
from RF
Through RF Receive Slave’s
Data
No No Write SMS
New SMS? SMS Body?
Indication?
Discard Data No Give Data to M20 M20
and Return Master? Yes Yes
Yes
Process
Received New Ne ed to send
Yes Data Store Received Store Received Send Message to m essage
Message Data Data M20

Send interrupt
Receive Data from
INT1 to Master to
the Mater PIC
Read this data
Send Message (i) To Main

Figure 7.1: Total NCN Operation Flowchart

Body Valid SMS No
Reset Parameters
Indication? Receive message
body from the
Yes Master PIC
No
SPI Data?

Delete Message Valid SMS No

Delete Message
Yes Format?

Yes
Send Data to Give Data
USA RT (RF) Through SPI Ready Send
Read New Run Protoc ol
Again Message(i) Check Message

69
 Reply to NCN
through RF
Transmitter

Transmitter
Disable RF

Enable RF
Receive Message
New Task

Perform the
from the RF

acquired
Task
USART
PIC

CCU PIC

Interrupt
Main

Timer0

93C56 EEPROM

Add to 80-bit
TAE Register
Register in
Store TAE
NVR

Passed?
Yes

40 Sec
No
To Main

Active Energy
(RSTENERGY)
Send 0x03 to

Read 40-bit
Comm. Reg.

Ready Data for

Calculate

Check Comm.
Energy

Yes

Register
Comm.

the PIC
Reg?
ADE7756

No

Figure 7.2: Total CCU Operation Flowchart

Due to any kind of error, the desired task may not be achieved. It could be a Central-
Office user error or a hardware failure at any system stage. In such cases, the NCN will respond
to the Utility with an error report that indicates the failed task or the invalidity type in the
Central-Office message.

In addition, if any error causes the NCN or the CCU to freeze its operation for more
than 40 seconds, it automatically resets the system and starts from the beginning again. The
NCN or CCU may be forced to shut down due to low-power situations and restart themselves
when power is supplied again. In either cases of startup, no fear of losing data should take
place, since the EEPROM chip connected to every PIC of the system is capable of storing the
PIC’s memory content every 40 seconds. This feature enables the PIC to reload its memory at
startup.
70
7.3 The System Through an Example
Our scenario can be presented as follows:

A consumer was fully disappointed of her last electricity-consumption bill. She

complained that the meter installed at her house door was “crazy”. As she said, her family
consumption was no more than half the value shown on the bill. The Customer-Service officer
calmed the lady down and asked her:

- May I have your bill, please?

He picked the Consumer ID number from the bill and inserted it to the Central Office
Consumer Information screen. One click and all information about the consumer and her
consumption were on the screen. From the data provided, the officer realized that the
consumer meter was covered by NCN of Security Code (ID1234) and it is directly connected
to the CCU numbered (15) within the mentioned NCN’s region. The screen also showed that
the mobile telephone number (077910803) was used to communicate with the desired NCN.

The first thing the officer thought of was to get the present meter reading and compare it
with the last reading, which was just the day before, since the reading process became a daily
habit of the system’s NCNs. He selected “Read Energy” task from the tasks list of the Central
Office, filled the needed fields and clicked on “Send Task”… and that’s it!

Few seconds and the Central Office alerted for the reply reception. As the data got from
the meter and after the automatic analysis of the consumer load-line history, the consumer had
an unusual high-consumption period for one week. When the officer described the situation to
the lady, she shouted:

- Oh, sorry! I have just remembered. It was my son wedding and we kept celebrating
for a week. No lights, flashes, sound systems and other staff were shut down for total
seven days! Sorry sir, but I still wonder, how could you read my house-meter in a
glance?!

Here is the Answer…

71
 @151@

AT+CMGS=”077910803"
@077884717 ID1234 151@
Cellular Base
Station
PIC PIC PIC ADE7756
M20
RF Module RF Module ADE7756 Electrical Load
Nokia 8210 M20 Master Slave PIC

NCN ID1234 CCU 15

AT+CMGS=”077884717" Process Requested Task :

@151 00000037F2ED24B56B54@ (Send TAE to Central Office)

@151 00000037F2ED24B56B54@

@151 00000037F2ED24B56B54@ Cellular Base

Station
PIC PIC PIC ADE7756
M20
RF Module RF Module ADE7756 Electrical Load
Nokia 8210 M20 Master Slave PIC

NCN ID1234 CCU 15

72
Conclusion
 We have chosen to implement the Fixed-Network AMR system because it seemed the
best solution for the problem of wireless electricity metering. We tried to give as much as
possible to open an eye on the endless capabilities of such system.

8.1 Work Done

Most of what was planned in Chapter 1 has been done. An automatic meter reading
system utilizing the GSM network and a custom RF solution along with the data acquiring
tools, storage devices, process-controlling equipment and software was designed and tested for
the simple reason of demonstration.

Data is interchanged between different sections of the system; a Central Office

wirelessly communicates with a network of NCNs. These NCNs extend the capacity of the
system by controlling other sub-networks of CCUs.

Each CCU is equipped with data acquiring tools used to gather the customer's
information (e.g. consumption, line status…) and other process-controlling tools (e.g. line
switching). Thus the CCU is the hand of the electrical utility at the customer side.

The NCN interfaces the Central Office with the CCUs; this is done by reshaping the
SMS-formatted data from the Central Office to a new format sent through the RF to the CCU
and vice versa.

8.2 Problems Faced

As much the previous chapters gave an idea of the simplicity of the system, as many
major problems were faced (and solved).

These problems were:

• Need another USART Module: The Master PIC in the NCN unit needed to interface
with two devices (the M20 and the RF module) through the USART. However, the
PIC18C452 used has only one USART module. This lead us to use another PIC
(through SPI) and use its USART as the second port to which the RF modules where
connected. This usage wasn't easy to make as it sounds to be, since too many control
lines and interrupts were used. This issue was covered in Chapter 3.

• RF Synchronization: The RF receiver used had a problem of that it can't operate

properly unless a carrier signal (from the transmitter) is detected. But if no carrier is
detected; the receiver will take noise entering its circuit as valid data and output it to the
PIC. This was due to bad design in the RF receiver by not implementing a Carrier-
Detect signal. We have overcome this problem enforcing a single transmitter to always
enable its carrier signal, and because there were many transmitters in the system;
software-controlled synchronization was made to ensure both data exchangeability and
receiver stability.

• Software Management: The work flowchart for the Master PIC in the NCN unit was
full of interrupts and interconnected branches of software modules. Developing such a
complex structure was hard especially because Assembly language was used. However,
even if the development language was a higher-level language (e.g. C) and hence easier
to manage; the size of the final hex dump file could be larger and could not fit the PIC's
memory space (we are already using 2/3 the PIC's memory).

74
8.3 Further Suggestions
We have introduced a basic design of a Fixed-Network AMR system. Yet there are
more ideas and suggestions that can be integrated within this design.

Power Supply Backup System

The reader could have the thought that this system needs a power supply independent
from the line voltage feeding the load. This helps the system to be always online and report
different urgent problems at the site (e.g. a power down situation).

The figure below shows one simple design using the 7805 voltage regulator and a
regular 5V rechargeable battery.
+12V 7805 +5V Power Supply
1 3 Feeding the
VIN VOUT
System
GND

R
2

RECHARGEABLE BATTERY

Figure 8.1: Rechargeable Power Supply Backup System

Other Ways for Gathering Data

There could be other ways than the GSM network to gather the customer's data.
Telephone lines are one example; a V22 chipset can be used with the PIC to interface the PIC
with the telephone line. The dial tones can be used as the data form. Customers' information
can be gathered at times when people are not using the phone (mainly at night).

There is another new method (or idea) to collect data; it is by using the power lines
themselves. The power lines mainly contain the 50 Hz frequency. Higher frequencies can be
used to interchange data between the system parts. This method is used in high-voltage power
lines (to quickly control the circuit breakers switching).

Web-Enabled Customer-Service Solution

As all information and statistics about the customers' consumption are available at a
higher resolution than the traditional way; a web-enabled customer-service solution can be
introduced. Such solution would enable the customer to monitor e.g. his consumption rate and
view his overdue bills online.

Other Systems Benefiting from this System

As the PIC is interfaced with the M20; a data channel is always available from the
customer's side, therefore enabling any other ideas for new systems to gather whatever data
from the customer's side.

Examples are many; water and gas consumption, real time security alarm systems,
wireless control systems…

75
References
References:

[1] Automatic Meter Reading Association, www.amra.org

[2] Itron Company, www.itron.com, and other resources
[3] Microchip PIC18C452 datasheets on the companion CD, www.microchip.com
[4] Analog ADE7756 datasheets on the companion CD, www.analog.com
[5] Siemens M20 Terminal datasheets on the companion CD, www.siemens.com
[6] RF Solutions RTFQ1/RRFQ1 datasheets on the companion CD, www.rfsolutions.co.uk
[7] Fairchild 93C56 datasheets on the companion CD, www.fairchildsemi.com
[8] DS14C232 datasheets on the companion CD, www.national.com
[9] Automatic Remote Meter Reading, graduation project (2002) by Camelia A. Najjar and
Salam M. Zalloum

77
Appendix A
 Tx Rx

+5 V

+3.25 V 05 05 05 +5 V
LM317

3 2 FM4
VIN VOUT FM-RRFQ1
1
ADJ 100 uF

Vcc
GND
Din
GND
GND
NC
NC
Dout
PD
93C56
4 3 150 1 FM-RTFQ1 6
DO DI En EA

1
2
3
7
11
12
13
14
15
2 2 5
CLK 3 In GND 4
1 GND Vcc
CS
240 100 uF
100 uF

MasterPIC Slav ePIC

300 LED

Enable Tx
PIC18C425 PIC18C425
2.7K 1 40 1 40 300 LED
2 MCLR/Vpp RB7 39 2 MCLR/Vpp RB7 39
3 RA0/AN0 RB6 38 3 RA0/AN0 RB6 38
4 RA1/AN1 RB5 37 100 4 RA1/AN1 RB5 37
5 RA2/AN2/VREF- RB4 36 5 RA2/AN2/VREF- RB4 36
C945 100 6 RA3/AN3/VREF+ RB3/CCP2 35 6 RA3/AN3/VREF+ RB3/CCP2 35
7 RA4/T0CKI RB2/INT2 34 7 RA4/T0CKI RB2/INT2 34
+5V RA5/AN4/SS/LVDIN RB1/INT1 RA5/AN4/SS/LVDIN RB1/INT1
8 33 8 33
2N3053 9 RE0/RD/AN5 RB0/INT0 32 VCC 9 RE0/RD/AN5 RB0/INT0 32 VCC
10 RE1/WR/AN6 VDD 31 10 RE1/WR/AN6 VDD 31
11 RE2/CS/AN7 VSS 30 11 RE2/CS/AN7 VSS 30
12 VDD RD7/PSP7 29 CS 12 VDD RD7/PSP7 29
VSS RD6/PSP6 93C56 VSS RD6/PSP6

0.1 uF
0.1 uF

100 uF
100 uF

LAMP 13 28 13 28
14 OSC1/CLKI RD5/PSP5 27 14 OSC1/CLKI RD5/PSP5 27
15 OSC2/CLKO/RA6 RD4/PSP4 26 15 OSC2/CLKO/RA6 RD4/PSP4 26
3.6864 Mhz 16 RC0/T1OSO/T1CKI RC7/RX/DT 25 3.6864 Mhz 16 RC0/T1OSO/T1CKI RC7/RX/DT 25
17 RC1/T1OSI/CCP2 RC6/TX/CK 24 SDO Reset Slave 17 RC1/T1OSI/CCP2 RC6/TX/CK 24
18 RC2/CCP1 RC5/SDO 23 SDI 18 RC2/CCP1 RC5/SDO 23
OFF RC3/SCK/SCL RC4/SDI/SDA RC3/SCK/SCL RC4/SDI/SDA
1 2 3 4 19 22 Disable Receive 19 22
20 RD0/PSP0 RD3/PSP3 21 to Salve Interrupt1 20 RD0/PSP0 RD3/PSP3 21

20 pF
20 pF
20 pF
20 pF

RD1/PSP1 RD2/PSP2 RD1/PSP1 RD2/PSP2

CS Slave
CS ADE7756
SCK
ON
to MasterInterrupt1

U5
13 12
8 R1IN R1OUT 9
R2IN R2OUT 220V
M20 11 14
10 T1IN T1OUT 7 1
SIEMENSE 1 uF T2IN T2OUT
1
C1+

+
1 uF 3 4.7 uF
1 4 C1-
Tx C2+

+
5
2 2 C2- 10K 220/12
Rx V+ U2

+
6
3 4 1 uF V- ADE7756
GND Antenna 16 20 1 20K
VCC VCC DIN RESET VCC
15 19 2
+ TC232 18 DOUT DVDD 3
1 uF 17 SCLK AVDD 4
CS V1P
16
15 CLKOUT V1N
5
6
NCN Schematic
1 uF 14 CLKIN V2P 7
+
3.6864 Mhz 13 IRQ V2N 8
12 SAG AGND 9
ZX REF
33 pF

11
CF DGND
10
10K
Diagram
33 pF

100K
300

VCC

79
 Tx Rx

+5 V

+3.25 V 05 05 05 +5 V
LM317

3 2 FM4
VIN VOUT FM-RRFQ1
1

GND

GND

GND

Dout
ADJ

Vcc
100 uF

Din

NC
NC

PD
150 1 FM-RTFQ1 6
En EA

11
12
13
14
15
1
2
3

7
2 5
3 In GND 4
+5V GND Vcc

240 100 uF
100 uF
Enable Tx

ChildPIC
LAMP 1 300 LED
PIC18C425
1 40
2 MCLR/Vpp RB7 39
3 RA0/AN0 RB6 38
4 RA1/AN1 RB5 37
RA2/AN2/VREF- RB4
5
RA3/AN3/VREF+ RB3/CCP2
36 A
C945 100 6 35 4 3 2 1
7 RA4/T0CKI RB2/INT2 34
RA5/AN4/SS/LVDIN RB1/INT1
2N3053 2.7K 8
RE0/RD/AN5 RB0/INT0
33 B
9 32 VCC
10 RE1/WR/AN6 VDD 31
11 RE2/CS/AN7 VSS 30
0.1 uF

100 uF

VDD RD7/PSP7
12
VSS RD6/PSP6
29 3
DI DO
4 C
13 28
14 OSC1/CLKI RD5/PSP5 27 2
15 OSC2/CLKO/RA6 RD4/PSP4 26 CLK
3.6864 Mhz 16 RC0/T1OSO/T1CKI RC7/RX/DT 25 1
17 RC1/T1OSI/CCP2 RC6/TX/CK 24 CS
SCK 18 RC2/CCP1 RC5/SDO 23
19 RC3/SCK/SCL RC4/SDI/SDA 22 93C56
20 pF

20 pF

20 RD0/PSP0 RD3/PSP3 21
RD1/PSP1 RD2/PSP2

CS 93C56

CS ADE7756 220V

4.7 uF

10K
U2 220/12
ADE7756
20 1
DIN RESET VCC
19 2
18 DOUT DVDD 3 20K
17 SCLK AVDD 4
16 CS V1P 5

CCU Schematic
15 CLKOUT V1N 6
14 CLKIN V2P 7
3.6864 Mhz 13 IRQ V2N 8
12 SAG AGND 9
ZX REF

Diagram
11 10
CF DGND
33 pF

33 pF

10K

300
100K

VCC

80
 U4
300 LED
PIC18C425
1 40
2 MCLR/Vpp RB7 39 220V
3 RA0/AN0 RB6 38
4 RA1/AN1 RB5 37 1
5 RA2/AN2/VREF- RB4 36
100 6 RA3/AN3/VREF+ RB3/CCP2 35
7 RA4/T0CKI RB2/INT2 34
8 RA5/AN4/SS/LVDIN RB1/INT1 33
9 RE0/RD/AN5 RB0/INT0 32 VCC
10 RE1/WR/AN6 VDD 31
11 RE2/CS/AN7 VSS 30 4.7 uF
12 VDD RD7/PSP7 29
VSS RD6/PSP6

0.1 uF
100 uF
13 28
14 OSC1/CLKI RD5/PSP5 27 10K
15 OSC2/CLKO/RA6 RD4/PSP4 26 U2
3.6864 Mhz 16 RC0/T1OSO/T1CKI RC7/RX/DT 25 ADE7756 220/12
17 RC1/T1OSI/CCP2 RC6/TX/CK 24 20 1
RC2/CCP1 RC5/SDO DIN RESET VCC
18 23 19 2
19 RC3/SCK/SCL RC4/SDI/SDA 22 18 DOUT DVDD 3
20 RD0/PSP0 RD3/PSP3 21 17 SCLK AVDD 4 20K

20 pF
20 pF
RD1/PSP1 RD2/PSP2 16 CS V1P 5
15 CLKOUT V1N 6
14 CLKIN V2P 7
3.6864 Mhz 13 IRQ V2N 8
12 SAG AGND 9 10K
11 ZX REF 10
CF DGND

33 pF
33 pF
100K
U5
13 12
8 R1IN R1OUT 9
R2IN R2OUT 300
11 14
10 T1IN T1OUT 7
1 uF T2IN T2OUT
1
C1+

+
1 uF 3 VCC
4 C1-
C2+

+
5 1
2 C2- 6
V+

+
6 2 PC Rx
1 uF V- 7
16 3 PC Tx
VCC VCC
15 8
ADE7756 Calibration
+ TC232 4
1 uF 9
5 GND Schematic Diagram
+ 1 uF
PC Serial Connector DB9

81