Full Version: Power Theft Detector
SANATH MCX
Oct 1 2008, 11:29 AM
Hi to all,

I want to design the Power Theft Detector. Can anybody help me on this.

Regards
Sanath
Pune

SANATH MCX
Oct 3 2008, 5:19 AM
I think using TDR which is known as Time Domain Reflectometry we can detect the power theft and also locate the position of power theft. But the problem is how to design the TDR System.

Ajay
Oct 3 2008, 7:28 PM
i have no idea about TDR

SANATH MCX
Oct 4 2008, 10:03 AM
Thanx Ajay for the Reply :-) :-) :-)

Ok ..... First let me tell about the TDR. it is very simple

( Time domain Reflectometry) is a widely used technique to find the fault occurrence in cable. and the instrument is called as Time Domain Reflectometer or Cable Radar or Echo Meter.In this
1. Sending a well defined voltage pulse down to the Testing line and waiting for the echo. 2. Displaying the sending pulse and echo on Fast oscilloscope screen
3. Analyzing the characteristics of Echo such as Shape of the waveform, Direction, Impedance, Magnitude of the pulse.
4. Measuring the Echo time to find the cable length or distance to fault.

and following are the parameters included this
1. Rise Time of the Pulse
2. Pulse width
3. The amplitude of the Pulse
4. Impedance over the cable
5. PVF ( Velocity of Pulse)

I hope the above details have given clear idea about the TDR

[lightbox=image_url|title|group|float]http://rapidshare.com/../files/150797434/01_Outstanding_TDRs.pdf[/lightbox]

Now my problem is How we can design the TDR System and How we can make use of this TDR Technique in Embedded system for the Power Theft Detection and Location Over the Transformer Lines which I feel one of the great challenging tasks in the Present scenorio.

Can anyone help me on this.

Regards
Sanath
Pune

Email: [email protected]

SANATH EMBEDDED
Oct 6 2008, 12:08 PM
Hi to all,

I think I need to put the end for the TDR ( Time Domain Reflectometry)

let me have one more idea for this
we know DTMF ( Dual Tone Multi-frequency ) which can be sent over the electric power lines for the communication i.e switching of device ( which is called as Device switching via DTMF) by making use of of DTMF Encoder and DTMF decoder like HT9170.
So can we make of the above technique with some other modifications for the power theft detection. please help me on this.

Regards
Sanath
Pune

Email: [email protected]
ExperimenterUK
Oct 6 2008, 8:02 PM

Hi to all,

I think I need to put the end for the TDR ( Time Domain Reflectometry)

let me have one more idea for this
we know DTMF ( Dual Tone Multi-frequency ) which can be sent over the electric power lines for the communication i.e switching of device ( which is called as Device switching via DTMF) by making use of of DTMF Encoder and DTMF decoder like HT9170.
So can we make of the above technique with some other modifications for the power theft detection. please help me on this.

SANATH EMBEDDED

I think TDR *is* the way to go, but it seems you are the expert around here
Any method that works is likely to specialised and complex, I think you will just have to
invest in some expensive test gear and good books,
sorry

SANATH EMBEDDED
Oct 7 2008, 5:55 AM

While going through the TDR concepts I got stuck in below condition.

For the TDR to work the Cable should be Metallic and atleast two conductors insulated from each other.

Sanath
Pune

Email: [email protected]
ExperimenterUK
Oct 7 2008, 6:13 AM
I think it just means metallic conductors.. ie copper.
You need at least two conductors to look for unusual
interactions between them.

SANATH EMBEDDED
Oct 7 2008, 7:22 AM
Hi Experiment UK,

Ok .... Let me take the example of Electric transformer power lines.
it has got two lines ( Metallic conductors) one is Phase and another one is Neutral.
which are insulated from each other. Is it right or not....

Regards
Sanath
Pune

Email: [email protected]
ExperimenterUK
Oct 7 2008, 7:35 AM

Hi Experiment UK,

Ok .... Let me take the example of Electric transformer power lines.
it has got two lines ( Metallic conductors) one is Phase and another one is Neutral.
which are insulated from each other. Is it right or not....

Regards
Sanath
Pune

Email: [email protected]

SANATH EMBEDDED

Yes that would be correct, as long as the two lines are together in a cable,
and not six feet apart on big pot insulators

I think working with high power systems may make a difficult job
even harder, and rather dangerous.

SANATH EMBEDDED
Oct 7 2008, 7:45 AM

Experiment Uk Wrote ....

I think it just means metallic conductors.. ie copper.
You need at least two conductors to look for unusual
interactions between them.

ya u r right .... the unusual interaction between them is nothing but the Chage in IMPEDANCE..So how to know about impedance of perticular Cable. please suggest me.

Regards
Sanath
Pune

Email: [email protected]

ExperimenterUK
Oct 7 2008, 7:52 AM

Experiment Uk Wrote ....

I think it just means metallic conductors.. ie copper.
You need at least two conductors to look for unusual
interactions between them.

ya u r right .... the unusual interaction between them is nothing but the Chage in IMPEDANCE..So how to know about impedance of perticular Cable. please suggest me.

SANATH EMBEDDED

LOL.. my expertise comes from reading the pdf you recommended.

But it did say that if you don't know the impedance of a cable
you can measure a sample length.

SANATH EMBEDDED
Oct 7 2008, 8:04 AM
may be by measuring sample length...

I think The CHANGE IN IMPEDANCE is the key detection for the Fault detection.
Ajay
Oct 7 2008, 1:05 PM
well the o/p side of transformer, you can do some work.

On running cable you cannot insert or have a conductor there, right?

The lines going out from transformer you can easily have two conductors placed there with proper insulation.. not sure if this is correct method but can say one of the method.

not to forget its all dangerous
shyam
Oct 10 2008, 5:24 AM
hi SANATH,
sorry if i sound foolish..
but what is the najor hurdle? detection of power theft/fault or the transmission of data?

http://www.epanorama.net/circuits/tdr.html
copy paste the address in the browser if it doesnt work...
SANATH EMBEDDED
Oct 10 2008, 1:06 PM
Hi Shyam and Ajay
thanx so much for the reply

The major hurdle is to detect the power theft i.e nothng but the detection of tap over the transformer lines.even I have gt that refernce circuit from epanorama.
But I don't think so it will work out for this .Shall I proceed with that.

Regards
Sanath Kumar
Pune

Email: [email protected]
SANATH EMBEDDED
Oct 11 2008, 5:35 AM
Hi guyzzz

Let me tell u wat exactly I am planning to design whose details I attached in Rapidshare. Please find it.

http://rapidshare.com/../files/152866550/T631.pdf

Using this T631 any power-line compatible TDR will be able to identify where there is a tap. For example, this T631 will be able to find taps on live power lines with up to 415V (ie 3-phase mains).

If u beleive it or not this T631 High Specification Coaxial cable fault locator costs around GBP 3,108.( 1 British pound = 1.6887 U.S. dollars).

3108 British Pounds = 5248.4796 U.S. Dollars= ( 5248.4796 x 48) = 251927 INR (Indian Rupee)

I really wonder what makes it that much costlier.

nw let me know whether this can be designed or nt by making it cost effective by our Embedded system .

Regards
Sanath Kumar
Mangalore/Bangalore/Pune

Email: [email protected]

Arun Kumar V
Oct 11 2008, 1:34 PM

Hi sanath,

you had raised this topic of TDR some time back in our forum, this time the topic has resurfaced again.

none of us (i guess ) have any experience with the TDR concept nor had any need for it, since you have digged deep into the concept, you can keep the forum updated about project's progress.

Arun
SANATH EMBEDDED
Oct 11 2008, 6:37 PM

ya even I too agree, none of us know about TDR and working with TDR is very complex since lot of parameters, analog circuitary and analyzing the waveforms are involved.But for me it is must and I feel it is one of the challenging task .
Hope u guyz are with me.... :-) :-) :-)
and sure I will keep informing the forum regarding the project's progress and it's my pleasure. :-|

Regards
Sanath Kumar
Mangalore/Bangalore/Pune

Email: [email protected]
SANATH EMBEDDED
Oct 20 2008, 10:52 AM
Hi guys,

This project is still alive guys....don't make it dead:-) & will be completed at any cost

ok let me tell you one more alternative for this. other way for the power theft detection where the embedded based system will be kept in a distribution transformer which calculates the amount of power delivered to the consumer and also reads the actuall amount of power consumed by the user by making use of wireless network like RF from the meter end. So whenever there is a difference in the amount of power which will be considered as power theft over the line and easily we can have GSM interface for the communication to the power distribution system .

Arun Kumar V
Oct 20 2008, 3:21 PM

So whenever there is a difference in the amount of power which will be considered as power theft over the line

this is not always true, there are transmission losses, frequency dips, and energy is lost when trees are grounding the line wires ( that's why you would have seen the electricity dept. staff cutting down branches touching the line wires)

so, this method cannot guarantee accuracy.

Arun
SANATH EMBEDDED
Oct 21 2008, 5:03 AM

Even I too agree this method cannot guarnatee acuuracy. Looking art the positive & negatives, the positives of this
1.Power theft can be sensed but not with accuracy
2. Automatic Meter Reading - no need of assessors to go to the house to collect the electrical power reading from the meter.
3. Can be used to detect the other power tampering like Magnetic tampering, Current reversal tampering,Neutral tampering
and the negatives are
1. ya it can be used to sense the power theft....bt is it possible to get the exact location of power theft by saying okkk boss, theft is happening at 20 feet from here....
2. For this system to install, u should change almost current infrastructure of distribution system which is bit complex with the present scenario

3. Is this system is Real Time system ? In the sense (can it perform the task in specific time period given) for eg: People wen do the power theft by putting the metal tap, with in specific time that exact information has to be stored in database for the proof and sent to the power distribution system and that bugger should be caught red hand with proof in no time......is this possible with this system?
4. The system should be general one with high efficiency,easy to install anywhere with the small footprint and of course cost effective... u guyz may or may not be knowing today most of the R&D professionals are not working for creating the new inventions...but working on with the present invention & how to make it cost effective by making use of cheaper technologies.
and I think if we get the solutions for the above said problems it can be procedded ..B-)

SANATH EMBEDDED
Oct 23 2008, 5:02 AM
Hi Guyz .

I think there is no other way out other than going with TDR for this project. But I am more concerned about the following issues rather than the design problems.

1. TDR technique works on transmission lines with a controlled constant impedance.By its very nature the power distribution system is generally considered to have an uncontrolled variable impedance.

2. For the TDR to work effeciently there should not be much of discontinuities in the transmission lines but in the power distribution lines discontinuities caused by the connectors,joints, Legal taps and EMi caused by the capacitor & inductance fields.

So first we need to have a better solution to overcome these problems