Hello 
We took delivery of an Enyaq last week and stuck it on charge yesterday at home for the first time with our PodPoint charger.
I opened up Solar dashboard to have a look at the current draw and noticed that we had â460Wâ of Solar being produced in pitch black. Obviously this is an error within the system but can anyone explain this and possibly a way to rectify it at all ?
Hello @BioHzrd can you provide a bit more information on your monitoring setup? Is this an emonTx or emonPi? how many CTâs and what circuits are they connected too? Are you using an ACAC adapter for voltage sensing? I assume all single-phase?
Sorry for the delay in reply.
Itâs is an EmonPi set up with two CT clamps, I am using the provided adaptors to to ascertain voltage readings. The CT clamps are located on the homes tails to read Solar Input & Home Draw.
The above is what appears while the vehicle is plugged in.
Thatâs a little ambiguous. Can you clarify âon the homes tailsâ - do you mean the meter tails between the supplierâs fuse and the meter, or between the meter and any Henley blocks or the consumer unit? Where does the PV infeed from the inverter connect into the system? Maybe a single line diagram or a photo will help.
ClarificationâŚthe clamps are positioned on the 25mm tails on the consumer side of the home electricity meter and one on the live tail exiting the solar voltage regulator that then interiors I to the home consumer unit.
OK, thatâs clear now. In the top picture, the right-hand c.t. is on the incoming grid supply, the left-hand is on the feed to the house and downstream of the Growatt âOptimiserâ and the PV infeed.
That doesnât appear to be what you first wrote.
The immediate problem is a âfeatureâ of the Growatt - âOutput voltage range: 0.925*Vin (approximately input voltage less 18V)â which means that it has automatically introduced an error of around 7.5% because the two places you are measuring current are not seeing the same voltage.
With the c.tâs as they are, you have neither a âType 1â nor a âType 2â set-up, so youâll need to derive the PV contribution from the quantities you are measuring - having allowed for the 7½% discrepancy the Growatt has introduced.
I think you need to look at the processing youâre doing to derive the values youâre seeing. I donât think the EV has created the problem, it might have made it bigger so that youâve noticed it.
Incidentally, the advertising claim âto reduce power consumption of electrical appliances and save on electricity bills.â is not universally true. Some types of load obey a âconstant powerâ law (e.g. computer power supplies) and these will automatically increase the current to compensate for the reduced voltage. Some motors may run at reduced efficiency and actually increase the power consumed because of the lower voltage, while heating loads (kettles) will just take longer to reach the required temperature and while the power is reduced, the energy used will be the same.
It kindaâ makes you wonder what loads they had in mind - incandescent bulbs maybe, but does anyone still use them?
As well as constant-P loads and constant-R loads, EVs bring a new type of load (well, new for my household) in the form of constant-I loads:
It sits there sucking 31.5A almost oblivious to what V is doing. Like your heating load example, a lower V just means it takes longer to charge, but in this case Power is proportional to V rather than V2.
As tungsten lamp efficacy â V² (nearly), it isnât hard to envisage the situation where SWMBO decides itâs too dark and turns a second lamp on as well, so rather than saving 15%, the power increases by 70%. 
So recomendation would be to remove this voltage regulator ?
Was in when I bought the house.
I would check to see if thereâs a good reason for it being there. Looking at the data sheet, it also contains a (your only?) generation meter, so I think if itâs the only one (your supplier hasnât provided one), you need to replace that, plus youâll need the appropriate isolator, Henley blocks, protection, etc.
I suggest you log the voltage downstream of the regulator and check the range of voltages youâre seeing - first making sure your voltage measurement is accurately calibrated. Bear in mind that even though the UK voltage is a nominal 230 V -6% +10%, that (as far as I know) is a fudge and the âcentreâ voltage remains at 240 V.
If youâre regularly going over 235 V ( = 254 V Ă 0.925) then Iâd retain it, because your supply is out of tolerance on the high side.
If youâre regularly going below 216 V (= 230 V - 6%), then Iâd definitely remove it, because its output is out of tolerance on the low side.
Otherwise, what does it cost to run (there must be some losses - itâs got ventilation slots to let the heat out) and is it worth the cost of removing it?
Alternatively, if you know what youâre doing, what I suspect is in there apart from the stuff youâd need to replace, is an auto-transformer, which would be easy enough to take out of circuit. That way, youâd keep the box and replace the auto-transformer by a suitable wire link.)
My personal view is the previous owner of your house was taken in by the advertising and most likely hadnât evaluated its veracity under all circumstances.
I remember having a look when I moved in it is a transformers about the size a side plate.
Iâm inclined to use the internal Din Rail connections and just remove transformer and link it with 25mm tails.
Did you check the actual voltages you get on the output side?
If you do want to disable it, I think that would be the way to go.
No I didnât but next chance I get to have it open Iâll check voltage on both sides of it.
Youâre really interested in the output side voltage, which you can check anywhere in the house. We know the input side voltage is 1.081 times that. If your emonCMS is not logging the voltage, do start logging it now (ideally, check the calibration first, you can tweak the voltage reported and logged by adding a calibration on the Inputs page).
Input voltage is 248.2 output is 238 between live & Neutral.
Above it was said I had it set up in neither type 1 or type 2 system. How should I have the clamps positioned to get the correct readings ?
The black armoured cable is the supply from the PV investor within the attic.
The emon pi is logging voltage through the supplied adapters obviously on the âoptimisedâ side of the system.
Typo?
The question still remains - what does your voltage look like over time? If itâs consistently 238 on the output, Iâd suggest you should probably leave the auto-transformer in circuit. But if it never goes above but goes below 220, then Iâd take it out.
I hope you donât mind me calling your current transformers c.tâs - thatâs how Iâve known them for the last 50+ years, âclampsâ to me are things you use for woodworking.
Youâve no option but to leave your c.tâs where they are, but you need to fiddle the maths on the Inputs page of emonCMS to get the correct values.
The one on the C.U. tail is âcorrectâ in that itâs measuring the current to the House and using the correct voltage with it to calculate the power/energy.
The other one on the Grid side will be reading the power low, because while the current it is reading is correct, itâs lower than the current would be on the consumer side of the auto-transformer because of the transformation ratio of the transformer. And itâs using the voltage on to consumer side to measure the power. So that power should be multiplied by 1/0.925 (according to the data) or 248.2 / 238 according to your measurement.
Explanation: the auto-transformer lowers the voltage on the house side by a factor of 0.925, therefore the current on the house side increases by 1/0.925
Or if you prefer, looking backwards through the auto-transformer, the voltage is stepped up, therefore the current must be stepped down.
This is because, barring losses, the power on each side of the transformer is the same.
What you are reading then is house power/energy (correctly), and grid power (incorrectly, but it can be put right with a fiddle factor).
Our convention is grid import and PV are both positive:
PV generated = House Power - ( Grid power Ă 1.08) (or 1.043 - their figure or yours)
You can do the multiplication on the Inputs page in emonCMS for the Grid Input with the calibration process, then log Grid power to a Feed; then for the House Input you log that to a Feed for the House Power, then subtract the Grid Feed and log the result to the PV Power Feed.
The fly I can see in that ointment is youâre subtracting two nearly equal quantities to arrive at the PV power, so very accurate calibration will be needed. Thatâs probably best done with minor tweaks to the â1.08â constant in emonCMS to get that to agree long-term with your Grid meter, then add another calibration value to the house c.t. so that the PV infeed power comes close to zero (but ideally a tiny negative value) in the dark - because the inverter will actually be consuming a few watts then.
Isnât that transformer likely to introduce quite a phase error? It seems youâd want your VT and all your CTs on one side of it, or the other.
If you could find a way to move the grid CT over to the PV feed youâd achieve that. But youâd need to find a way to split the inverterâs Active and Neutral out, perhaps inside the bottom of the Volt Dr.? Then you could just subtract one CT from the other to know how much the grid is supplying (or being fed).
An auto-transformer with a ratio of 0.925? I didnât think it would be very much at all, and now having done Fourier on my Variac, Iâm afraid your fears were unfounded.
Over a range of 16 V above the input voltage (around 236 V at the time) to 40 V below, I see the phase shift vary over the range +0.01° to +0.03° . I think we can disregard that. It is within Âą0.06° all the way down to 50 V out, itâs only below that the phase shift leaps up, to +0.32° at 15 V out.
That would be the best idea. We donât (yet) know just how the PV is connected in relation to the Volt Dr, I canât see it being anywhere other than on the primary side - but thatâs only from thinking about losses.
That looks very promising then, for your suggested scaling solution.
Good point. Another argument for it being on the primary side: most jurisdictions place requirements on inverter behaviour as the voltage approaches limits. To that end, youâd hope the inverter is monitoring actual grid voltage rather than stepped down grid voltage.
As you say above, a long term (at least a week) log of calibrated voltage readings is needed to decide if the transformer is doing anything helpful. If that does show the grid is running high, how easy is it over there to get them to come out and adjust the local transformer tap? Theyâve retrofitted remote voltage and current monitoring on the final distribution transformers around these parts. Hereâs what mine looks like over a week (as measured by me, not them):
As it stands, the OPâs auto-transformer is making the EV charge take 8% longer than it otherwise would.
Itâs possible, Iâve never needed to do it, but people here have reported that they take some convincing before theyâll even put a monitor on.
Yes, it looks as if that wonât be forthcoming, yet it would be exceedingly helpful in deciding whether to keep the Volt Dr or not.
Not knowing how the on-board charger controls the charge rate, is that necessarily true? If itâs controlling the d.c. current into the battery (which will be constant power for any given battery voltage), then surely it will be constant power on the a.c. side too, so 8% more current to compensate for the 7.5% lower voltage. And 8% more IR loss on the cables feeding the car, thus a negative saving on the electricity bill. 
If itâs regulating on a.c. current, then yes, 8% longer, so no saving on the bill. I guess whoever wrote that hadnât heard of Q = I Ă t.
