Wrong power reporting on some channels

[vespaman]

So, I purchased an extra GEM about two years ago, as I wanted to log my new electric car. This spring, I did some renovation at home, with updated wiring and outlets, so I needed to use a few new channels on my second GEM.
Problem is, it does not work well - the new circuits give errors on some GEM channels.
I have all micro-40 ct's and three phase 230V system. Both GEM are the black board version with ethernet only set-up. 4.33/1.49 fw.

Screenshot_20190716_142920.png (69.91 KiB) Viewed 6375 times

Channels 123 are my EV, I have previously assumed it was working, but haven't verified.
4 is a circulation pump (disconnected currently) I have previously assumed it was working, but haven't verified.
5 is verified working fine, an outlet.
6 is verified working fine, an outlet.
7 is showing 34W when I connect my reference 40W incandescent bulb.
(8 is not connected)
9 is showing 34W when I connect my reference 40W incandescent bulb.
10,11 are not tested yet.

Screenshot_20190716_143215.png (69.73 KiB) Viewed 6375 times

On my old GEM, I had to connect pole 3 to 4 on the CT input, so I tried this also now on channel 7, but that did not change anything. The micro 40 are connected between 2 and 3 on the terminal block.

What can be the cause? I have verified the phase-connections and set-up, If I move e.g. 7 to phase B or C, the error is much higher, so I think I can say, that the phase is correct.

[ben]

Is the amperage reporting correctly? VA should be equal to Watts in terms of incandescent light bulbs as they're purely resistive loads.

I wonder if your voltage is reading a bit low. Can you try cleaning the PT contact to see if that improves things? (assuming it's a stereo jack and not USB connector).

[vespaman]

So, the "good" circuit gives 0.18A @233V and 40W in the readouts.
The bad gives 0.14A @233V and 35W
(The 233V is the same, since the one PT is the same)

The nominal Voltage over here is 230V, so 233 should be fine, and also exactly what my other GEM reports.
Also my DVM gives the same. Unfortunately, I don't know off hand which phase the PT is connected to, but I think it is phase 'A'. (Note that I have both working and non-working on phase 'B' (5 & 9), so I don't think different phase Voltage is the reason here).

But:- I previously have had issues with the stereo connector, (much to my surprise), so I understand where you are coming from.

[ben]

So, the "good" circuit gives 0.18A @233V and 40W in the readouts.
The bad gives 0.14A @233V and 35W
(The 233V is the same, since the one PT is the same)

The nominal Voltage over here is 230V, so 233 should be fine, and also exactly what my other GEM reports.
Also my DVM gives the same. Unfortunately, I don't know off hand which phase the PT is connected to, but I think it is phase 'A'. (Note that I have both working and non-working on phase 'B' (5 & 9), so I don't think different phase Voltage is the reason here).

But:- I previously have had issues with the stereo connector, (much to my surprise), so I understand where you are coming from.

Are you using same bulb for for all the tests?

The PT is always on Phase A, correct.

[vespaman]

Yes, I move it around plugging it into different outlets. I even considered that the heat build up of the bulb might give some very small variation as well, so I tried short tests.

When I move it to a circuit of the other (first) GEM, I get slightly higher, about 42W, but that I can understand, since it is a different GEM, with different PT etc. Ideally I would like to see them similar, but that might be wishing for too much. Anyway, the 5 missing W on some of the GEM 2 inputs is rather too much, and it does not make sense to me.

Could it be some reference voltage that is bad on inputs above 6? Is there any such grouping internally?
Or could there be some issues with some of the CT's?

I guess one thing to test, is to switch the working and non working inputs on the GEM over to see if the problem follows the input? I'll try this tomorrow if you don't have any other suggestion (it getting late here).

[ben]

Yes, I move it around plugging it into different outlets. I even considered that the heat build up of the bulb might give some very small variation as well, so I tried short tests.

When I move it to a circuit of the other (first) GEM, I get slightly higher, about 42W, but that I can understand, since it is a different GEM, with different PT etc. Ideally I would like to see them similar, but that might be wishing for too much. Anyway, the 5 missing W on some of the GEM 2 inputs is rather too much, and it does not make sense to me.

Could it be some reference voltage that is bad on inputs above 6? Is there any such grouping internally?
Or could there be some issues with some of the CT's?

I guess one thing to test, is to switch the working and non working inputs on the GEM over to see if the problem follows the input? I'll try this tomorrow if you don't have any other suggestion (it getting late here).

The difference across GEMs could just need a voltage tweak. Different outlets may read different voltages so it's important that if you're calibrating the voltage against the outlet the PT is plugged into.

The input test would be interesting. You'll have to switch the phases on the channels also so they match the reference CT. A bit more annoying but could try swapping CTs also to see if the Micro-40 connected to a bad channel reads the same on a good channel (leaving the GEM config the same).

[vespaman]

Hi Ben, I was just about to post this, when I saw your last post..

So today, I switched 6 and 7, and the problem still exists on the same circuits, i.e. the error follows the CT's. Changing CT is a much larger/time consuming thing, so I have not tried to do this. Is this something you have had issues with before? The Micro 40's was of the last batch I bought from you in 2017.

Or could it be something else being the issue here, that you can think of?

Here's some measurements I did;
First off on circuit on phase C, input 6 on GEM. This is pretty on spot.

20190717_183640.jpg (440.2 KiB) Viewed 6354 times

Second: on circuit on phase B, input 9 on GEM. A bit off, but fluctuating.

20190717_184108.jpg (453.87 KiB) Viewed 6354 times

Third (the worst) on circuit on phase A, input 7 on GEM.

20190717_185332.jpg (534.49 KiB) Viewed 6354 times

[ben]

So today, I switched 6 and 7, and the problem still exists on the same circuits, i.e. the error follows the CT's. Changing CT is a much larger/time consuming thing, so I have not tried to do this. Is this something you have had issues with before? The Micro 40's was of the last batch I bought from you in 2017.

We haven't had any that we're aware of but it's always possible. Got a spare you can slide on the same circuit as 7 and see if it reads differently?