Archived Forum

I am testing with one appliance now just to get things going.  My burdern resistor was sized to allow 15 A RMS for now so 160 ohm is the value I got and I am using.

The problem is that when I hook the CT to the scope, without any load on the CT, I still get a sinusoidal waveform that fluctuates and has noise in it.

I am leaning towards having a faulty efergy CT.

My testing load is an iron which draws 9.22 A RMS (measured by the Kill-a-watt) constant until the sensor/switch kick in and disconnect it.

Therefore, during the time the iron is on, the peak value that the arduino should read is 9.22A RMS x root(2) / 1350 turns x 160 ohm + 2.5 V = 4.04V

For some reason, the arduino is only reading a sinusoid with a peak of 3.1V instead of 4.04.

The burden resistor was sized to 160 ohm so that it accommodates a max load of 15A RMS (for testing purposes only). The power factor read by the Kill-a-watt is 1 so it's a purely resistive load.

My testing load is an iron which draws 9.22 A RMS (measured by the Kill-a-watt) constant until the sensor/switch kick in and disconnect it.

Therefore, during the time the iron is on, the peak value that the arduino should read is 9.22A RMS x root(2) / 1350 turns x 160 ohm + 2.5 V = 4.04V

For some reason, the arduino is only reading a sinusoid with a peak of 3.1V instead of 4.04.

The burden resistor was sized to 160 ohm so that it accommodates a max load of 15A RMS (for testing purposes only). The power factor read by the Kill-a-watt is 1 so it's a purely resistive load.

You mentioned earlier that you are using the Efergy current transformer. There are two types of Efergy current transformer:

The old type is white with a DC power plug and has no burden resistor nor any other protection built in. Consequently, extreme care must be taken to ensure that it is not unplugged whilst current is flowing in the primary, otherwise possibly dangerous and damaging voltages may appear.

The new type is black, with a 2.5mm stereo jack plug. Inside there are a number of components (see http://openenergymonitor.org/emon/node/476#comment-3495) that mean this device is in general unsuitable for use with the emonTx. So what you are seeing is probably correct for the new Efergy CT, but not good for the emonTx. A couple of months ago I wrote a detailed report on this, which Glyn & Trystan have; and this is the reason why the Efergy CT is no longer recommended. Why Efergy have included these components we cannot know, but remember that the Efergy monitor does not measure power, it measures current only and estimates power from the nominal supply voltage (which you set via a software menu), and an assumed power factor. So my best guess is the diode-resistor network forms part of the input signal conditioning circuit, and Marc_Eberhard's theory that it improves sensitivity at low powers is probably correct. 

The emonTx calculates true power by multiplying instantaneous values of voltage and current and then averaging the result, so for a number of reasons it would be difficult to 'correct' the voltage output of the new black Efergy CT.

I have the new type since it is black and it came with the stereo plug which I cut in order to have access to both terminals.

I totally understand that it's going to be measuring current only and that the power calculations are done via software. I am an Electrical Engineer so you can assume I know all the basic stuff.

Since the Efergy model is pretty much junk for our use with the emonTx, which CT do you recommend I use then from your experience?

Thank you very much for the help and looking forward to your recommendation.

This one: http://shop.openenergymonitor.com/

(There is no need to cut the plug off this one - it is correct for the emonTx kit as supplied).

"I totally understand that it's going to be measuring current only and that the power calculations are done via software. I am an Electrical Engineer so you can assume I know all the basic stuff."

The point I was trying to make was that I have a suspicion that the Efergy energy monitor (as a system) is not reliant on accurately monitoring the shape of the current waveform in the same way that the emonTx is, so it probably uses the average of the positive half wave, or maybe the peak, to compute its idea of rms current that it multiplies by what the user has told it for the rms voltage to get the indicated (estimated!) power (and ignores too any variation in power factor).

Thanks for the info Robert.

I am not starting with emonTx now so will I need to cut the stereo plug so that I can use it in my circuit I have above? I need the CT's output to be connected across the burden (unless the SCT-013-030 has a burden resistor built-in).

I think all you could possibly need to know about the yhdc CT is now here: http://openenergymonitor.org/emon/buildingblocks/report-yhdc-sct-013-000...