Archived Forum

You can put all three cables through your CT, but it won't measure what you want and need to know! What it will measure is the vector sum of the three currents which, if your connected load is perfectly balanced as it should be (but rarely is), will be zero.

So I need more details to be able to help:

What do you mean by a Raspberry Pi system? What components do you have?

Is your supply a 3-phase, three-wire or a three-phase, four-wire system; i.e. do you have a neutral connection?

What is the line-line voltage (if a 3-wire system), or the line-neutral voltage (if a 4-wire system)?

(If you don't know the answers, knowing your country might help!)

Hi Robert,

  First of all thanks for your fast reply. I bought the raspberry pi 2 with a current and temperature sensor adapter compatible with emoncms (http://www.ebay.co.uk/itm/Raspberrypi-Current-Temperature-Sensor-Adaptor...), and I have the clamp for the wires. I'm in Portugal so we use 220v current. I'm sending you the pictures of the board and wires :

What you say, and the writing on the PCB: "2 CT, 1 TEMP" and the fact that you have a 4-wire supply (Brown, Grey, Black, Blue) makes me certain that it is not suitable. You must have 3 CTs to be able to measure your total power. You need three current inputs and 3 voltage inputs to accurately measure 3-phase real power, you need three current inputs and 1 voltage input to have a reasonably accurate measurement of 3-phase real power. If you are content with only an estimate for apparent power (VA, based on an assumed supply voltage and power factor) then you do not need the voltage input(s) but you still needed to buy the version with 3 current inputs. The Ebay seller does not appear to have a version that can measure real power (W).

(With your CT clipped on the 3 cables as your picture, you are measuring the out-of-balance, i.e. the neutral, current! You will get exactly the same reading by clipping the CT onto the blue wire. That is not reading the current you need to measure.)

If you want measurements of real power, I think the cheapest and safest way forward for you is to purchase from our shop an emonTx V3 without a radio module, a programmer (and lead if required), a Euro ac adapter and 2 more CTs. You can then re-program the emonTx with a 3-phase sketch (from Github, slightly modified to give the correct serial output) and connect it serially to your Pi, on which you load the emonHub software.

There are plenty of people here who know about the Shop products. You will find there is less support for products from other places, simply because fewer people here are likely to have them.

Hi Robert,

 Thanks for your answer. I bought the version on ebay because It was cheap and I always buy the cheaps option and I really don't know what I'm doing :). The problem what the option that you are telling me is not the price but the fact that I'm going to need some electronic knowledge and I'm more an IT guy. Is there any other option available that I can buy with everything already setup?  

Unfortunately, it was the wrong thing, so not at all cheap.

There is a page in Building Blocks ("AC Power Theory - An introduction") that explains how we measure power (and it continues to explain the Arduino maths etc).

Briefly, the adapter that you have measures only current. If you multiply that current by the voltage you think your supply is, you will get an estimate of Apparent Power. How accurate that is depends on the actual voltage, and a thing called 'power factor' that depends on the type of load that you have. If you buy the "3 current" version, that is what you will read. You will add the three apparent powers for your three phases to get the total apparent power. (And you will need to multiply by 0.917 in emoncms because the adapter is set to use the UK voltage of 240 V, so it will read 9% high in any case.)

If you measure the voltage separately from the current, you can improve that estimate because you know the voltage, but you still have the power factor (which you don't know and varies according to the appliances in use) causing inaccuracies. But if you measure voltage and current samples at the same instant, multiply and then average the result, you get real power, and that is what your electricity supplier charges you for. Real power is never greater than apparent power. The two are equal for a load like a water heater, but real power may be only half of apparent power for a load like a pool pump.

If you don't want to do emonTx - serial - Pi (and the only electronics you need is 3 wires for a serial connection), then you could buy the emonTx with the radio installed, and buy a RFM69Pi receiver board to go on your Pi. There's no electronics there that you need to do, it's all software (reloading a sketch - programmer needed - and installing emonHub on your Pi).

If you don't like that, then I think your only choice is to buy the "3 current" version of your adapter and accept that there will be a difference between the power you read and the power you are charged for. You should be able to add a calibration factor that will make the average over a long time accurate, but over a short time as the appliances you use change by the hour or the minute, there will always be an inaccuracy.

Hi Robert,

  I was out on vacations that's why I didn't replied to your comment. I made a mistake. My adapter has 3 CT's link above.

http://www.ebay.co.uk/itm/Raspberrypi-3x-Current-Sensor-Adaptor-emoncms-...

So if I buy just 3 clamps I can monitor every phase am I correct?

Thanks in advance,

Indeed. I think I've covered that in the third and last paragraphs of mine above.