I have an emonpi2 with 6 available inputs;
2x 200A on the main North American split phase lines from the meter.
1x 50A on a not in use 50A circuit
1x 20a on the circuit powering the EmonVS and thus powering the emonpi (a bit redundant but it was covered by a non OEM energy monitoring system and simplifing the panel was best)
The P1 first 200A CT is reading what I assume is roughly half the power draw of the building
Since the get go the second 200A CT was reading 0 or 1 or -1 I assumed it was bad, I reseated it, verified no visible cracks in the exposed magnet. Fliped the direction in case some insane variable of inverted casing was at play(even though it would have just read the number in the negative) Nothing.
Iâve tested the other channels, re-configuring the serial config to the CT rating. Doing the whole sequence is getting too familiar. The numbers never stray from the -1-0-1.
To just double check that the one Iâm replacing is or isnât bad I did a resistance test on my Fluke 101. It shows 5.7 and sometimes goes up to 10 and down to 4, but I think that is just the variance of the probes on the jack.
So I ordered another CT.
Interestingly, the new one out of box shows the same results. All I have found in the forum here is they should be 100. I havenât found a post specifically about the 200A CT and resistance readings. . I tried to find the solution before reaching out further.
I have a network of Shelly EMâs that I used for high powered devices that have a frequent use case. Some even on whole circuits. The power and energy reading they provide show the emonpi is definitey not reading all the power on the one functioning CT and missing all the sensor information on the other.
That being said, I love the emonCMS/hub/lib/etc⌠The platform is great, forwarding things over to my MQTT network was easy as ever to setup. It feels like a (hardware) product that was cared for and most likely hand assembled and verified. Really glad I got one. Now to get it working.
Did you buy your emonPi2 set up for use in N.America, and did you buy a N.America or 3-phase emonVs with it?
My feeling is your âbadâ c.t. is happily (and properly) measuring current but it hasnât got a voltage to multiply by to get power, and as we all know anything multiplied by zero is zero - or in your case anything multiplied by the tiniest sniff of voltage (or electrical noise) is 1 or -1.
You should be able to prove this very easily by unplugging the âgoodâ c.t. and plugging the âbadâ one in its place. Anything close to a sensible number says the c.t. and its installation is OK. (If itâs one from our shop, or it says 333 mV on the side, the c.tâs output is only about 330 mV maximum, no need to isolate anything when you swap it.)
The software in the âemonTxâ part of your emonPi is capable of being set up for any configuration of inputs, both voltage and current, but if you have the default UK settings, itâs likely the pairing of voltage and current is wrong for your load centre. And the serial config doesnât allow you to change the channel pairings - why this has never been included I donât know, maybe âtoo complicatedâ.
If you look in emonCMS â Admin â Update and the options under âFirmware onlyâ youâll see an option for N.American Split Phase 6-channelâŚ, but I canât see that in Github and as Trystan wrote it (I only wrote the library and the demonstration sketches - but I know it does work correctly and give the right answers on your split supply when configured correctly), and without knowing whatâs in your emonPi or which emonVs you have, I canât say exactly whether what Iâve suggested above is indeed your problem. Unfortunately, the âemonTxâ software doesnât tell us what configuration itâs set up for, so unless you know, all you can do is reload something else and see what happens.
I got the North American Emonpi from the OEM shop, with a NA 3 phase VS, whereas I am only measuring the single line voltage with it. I have the NA split phase firmware installed 2.1.0
Are you saying that it is reading the CT and comparing it to a predesignated voltage measurement, to dertmine power draw? I didnât catch that in the documentation, at least not different channels for different voltages.
If that is so I just need to tell it to use the single line voltage reading, or just adjust the math to use the same V1 measurement.
If it is indeed still a OK, I have future project in the workshop I can use it on, itâs essentially guranteed to get used eventually.
I think you need to connect the âBâ leg voltage to the second voltage input of your emonVs - this is going to be the simplest and best solution. Ignore the 3rd voltage input.
If you mean what I think you mean, yes.
On your split-phase system, you have 120 - 0 - 120 V, so you have loads on the âAâ leg running on the voltage from one half of the transformer winding, and you have loads on the âBâ leg running on the voltage from the other half. Those voltages will be close but rarely the same. I canât find any documentation for your âNA split phase firmware installed 2.1.0â, Iâve a feeling that Trystan has written it so that channels 1, 3 & 5 are using the âAâ leg voltage to calculate power, and channels 2, 4 & 6 are using the âBâ leg.
The pairing can be adjusted very simply, but it means recompiling the âsketchâ (Arduino-speak for software) and transferring it and uploading it into the emonTx board inside your emonPi2. What youâve got to change if you do go down that route is fully detailed in the emonLibDB documentation:
void EmonLibDB_set_pInput(uint8_t _cInput, uint8_t _vInput1 [, uint8_t _vInput2])
Sets the relationship between voltage and current inputs to enable the calculation of real
power. For single-phase, one leg of a split-phase and 3-phase 4-wire measurements, voltage and current sensor must be on the same phase/leg. The second voltage is needed only for the line-line power calculation in split-phase and 3-phase systems; i.e. when the current measured is line-line.
For example, to link CT3 to V1: EmonLibDB_set_pInput(3, 1);
and to link CT4 to V2: EmonLibDB_set_pInput(4, 2);
If you canât or donât want to do this, then I think you can reload a single-phase front-end software, but your powers wonât be as accurate as they might be because the loads (CTs) on the âBâ leg will be using the wrong voltage (unless all the loads youâre interested in are all on the âAâ leg - very unlikely, or the two legs are always perfectly balanced - equally unlikely), and there was no point in having the 3-phase emonVs.
If that is the case the simplest way to fix the problem would be to use channel 3 for the second phase (I swear I read a community post saying to forget measuring the voltage at each phase, oneâll do)
I do appreciate you identifiying the part to recompile with the (my)correct settings, I just donât think Iâll have time to tackle that today. But I have learned more. Need to read those other docs more thoroughly.
This would be one personâs opinion. Iâve pointed out the reason for measuring both, if that person decided one voltage was adequate for their needs and the resulting inaccuracy acceptable, and you agree it will be good enough for you, thatâs fine. But as you have already bought the hardware to do it properly, and it involves one extra wire (and a breaker to protect it, unless you are already using a 3-pole-plus-ground plug - maybe NEMA 14-20P ? though 20 A is serious overkill - in the UK it would be no more than a 6 A MCB), it seems silly not to use it.
I setup this over the winter holiday and havenât had a lot of time to work on it till now. Last week I started tinkering with it and ended up re-reading the docs. I have a dedicated breaker for the EmonVS and I have a seperate phase breaker available to get the other voltage from. In a few months Iâll have the time to dig further into the panels and software. I can confirm that with the CTâs plugged into the 1 3 and 5 ports I get reading from the 3 CTâs in use. And the math for the other phase suddenly is much more realistic.
Thanks again, Iâll update once I get to rewiring the EMonVS and then re-adjusting the CTâs
Not trying to be nit-picky, but what youâre referring to is actually one of two legs
of the North American split phase system.
I make the distinction so as to keep the spilt-phase system from being confused with an actual 2-phase system that (believe it or not!) is still in use today.
