Looking for info on how Current Transformers work

Peter_V · December 2010

I'm thinking of building my own power monitoring setup like "The Energy Detective" using a pair of CTs around both legs of the power lines coming from the grid and from my array.

I can understand the basic idea of how a CT works (just like a regular transformer) but what I can't figure out is how to tell which direction the power is flowing, i.e. from the grid or to the grid.

Not a big deal on the output from the array since it will mostly flow in one direction, but a huge deal on the grid side of the breaker panel.

I'd appreciate any help.

Thanks, Pete.

NEOH · December 2010

Re: Looking for info on how Current Transformers work

I think, the Amperage waveform will be "In-Phase" (approx 0 Degrees Phase Shift) with the Voltage Waveform when current is flowing IN (ie: when NET BUYING Power) and then the Amperage Waveform will be approx 180 Degrees Out-of-Phase with the Voltage Waveform when current is flowing OUT (ie: when NET SELLING power).

The Waveforms you observe may be exactly the opposite relationship as stated depending upon how YOU wind your CT's - CW vs CCW.

Also, for extremely Inductive loads the voltage waveform will slightly lead the amperage waveform and for extremely Capacitive loads the amperage waveform will slighly lead the voltage waveform.

You always want a load on the secondary windings of a CT otherwise a potentially "shocking" voltage can appear.

OK ?

anthem · December 2010

Re: Looking for info on how Current Transformers work

also watch out for reactive power loads. Looks like some inverters like to have large reactive power loads (no real power, but CT's read them high) using CT's and meters. . .

Peter_V · December 2010

Re: Looking for info on how Current Transformers work

NEOH wrote: »

I think, the Amperage waveform will be "In-Phase" (approx 0 Degrees Phase Shift) with the Voltage Waveform

Thanks, I didn't even think about comparing the current waveform to the voltage waveform, seems pretty obvious now.

I should even be able to estimate power factor by comparing the shapes and phase difference (though coding for that is going to be a pain)

Thanks again.

NEOH · December 2010

Re: Looking for info on how Current Transformers work

Peter,
How do you plan to compute the True RMS Amps value from the almost Sine Wave shaped and potentially very noisy signal returned by the CT?

Are you using a Hardware solution, Software or a combination?

Do you think you can get +/-1% accuracy for your computed Real Power (RMS Volts x RMS Amps x Power Factor Correction) ?

Peter_V · December 2010

Re: Looking for info on how Current Transformers work

NEOH wrote: »

Peter,
How do you plan to compute the True RMS Amps value from the almost Sine Wave shaped and potentially very noisy signal returned by the CT?

Are you using a Hardware solution, Software or a combination?

Do you think you can get +/-1% accuracy for your computed Real Power (RMS Volts x RMS Amps x Power Factor Correction) ?

I'm in the early stages of researching this, but I'm planning on using a combined hardware/software solution.
Right now I'm thinking a 1000:1 CT with a precision 50ohm load, that should provide a 0-5V output for 0-100 amps. Run the output through a low pass filter to get rid of most of the noise.

I expect there will be quite a bit noise, I use a lot of X10 stuff and the power company using powerline transmissions for reporting from the meter to their office.
I was originally thinking that I might be able to figure out how to monitor the power meters communications, but I talked to a guy at the power company and he said it takes over a day for the meter to send out one packet of data. That's a bit to slow to give me the real time monitoring that I'm interested in.

Anyway run the filtered voltage into a ADC, possibly use a Arduino for this, though from what I can gather they only have 10 bit ADCs. Maybe use a PIC, have to look and see what they are capable of these days, probably don't have enough power though.

I doubt I'll get anywhere near 1% accuracy, I'd be happy with 5% but might have to settle for 10%, at least at low power loading.

The CTs I'm looking at only have 1% accuracy so it can only get worse than that.

If I use a 10 bit ADC them the precision will only be to about 12 watts. To get the precision down to 1 watt would require at least a 14 bit ADC. Not sure it's worth that much effort though.

Most of the folks that I've come across on the web that are doing similar projects just take a swag at the power factor.
However, if I can get enough samples of the current waveform then it should be possible to calculate the power factor based on it's shape and phase difference from the voltage waveform.

I have a fair amount of programming experience, but I haven't done much work with micro controllers and I might be expecting to much from them (and from me)

I'd be happy to listen to any advice or suggestions.

n3qik · December 2010

Re: Looking for info on how Current Transformers work

I am using this for my data logging:

http://www.advantech.com/products/ADAM-5000-Series/sub_1-368QIL.aspx

Does need a computer, so using this mounted in the hallway:

http://us.shuttle.com/barebone/Models/X50v2.html

Peter_V · January 2011

Re: Looking for info on how Current Transformers work

n3qik wrote: »

I am using this for my data logging:

http://www.advantech.com/products/ADAM-5000-Series/sub_1-368QIL.aspx

Hmm, how much do those cost?

Thanks.

n3qik · January 2011

Re: Looking for info on how Current Transformers work

Here is where I got them from:

http://www.bb-elec.com/product_multi_family.asp?MultiFamilyId=54

I have $ 1700 in the current setup. But using only 1/2 of the I/Os, left room for expansion.

Looking for info on how Current Transformers work

Comments

Categories