Brand new. Granny charger trips RCD when charging. Replace how?

[WannabeOwner]

My Model-3 Gen 2 UMC tripped the inlaws house (both plugged into the garage (separate distribution board) and also into a couple of different 13 AMP sockets in the main house). No such problem at their neighbours (needed a charge, any charge!), back at home, nor at a couple of other friends' houses

All plugged in to 13 AMP. Not had opportunity to try Commando at any of those locations.

Have there been threads of this happening in other countries, or might it relate to how we do our wiring / earth-leakage in UK?

there are the persistent reports of charging at half speed, which seems like a software bug

I too think that is something different. I don't have that problem charging on Tesla Wall Charger at home.

 

[Bluebell]

Same problem here. I asked the SC to check out my UMC when my car was in for delivery issues to be fixed and they were able to charge using the 3 pin and commando adaptors. The problem therefore lies with my house electrics but I haven't been able to identify the culprit. A pain at the moment as it means that I have to use public chargers and superchargers until I get an EVSE fitted next week.

 

[Glan gluaisne]

I'm inclined to do some leakage current measurements when using the UMC when I get my car next week. I've a couple of spare current transformers, so could have a go at looking at the current waveform as well, to see if there's anything odd going on. It'd only be a single data point, but it might give some clues as to what may be happening.

 

[ianto999]

1) I tried umc on an old car park external 3 pin and it didn’t trip at the start of charge.
2) It does trip immediately car starts to charge, every time in house, so I think it is house//umc/car combo. I have PV. It doesn’t trip when plugged into car and socket, only when charge tries to start.
3) It also always immediately trips when I use a leaf Esve 3 pins and a type 2 adaptor.
4) It doesn’t trip immediately if I use the old leafs tethered rolec with the type 2 adaptor, but it does trip sometimes even with this setup during the Night charge, or when I turn on hearing remotely in the morning.
5) The house never trips otherwise.

I’m having zappi installed soon, so going to ask the electrician then, and will report back.

I should add that my leaf esve 3 pin never trips the house rcd.
The only common factors I have are the m3, umc 3 pin/(or leaf esve with adaptor), and house. If I use all three together - it trips, if I take one of these three out of the system, it doesn’t.

 

[ianto999]

I think talking about this must have jinxed the system.
Tonight I had my first trip in ages using my old leaf rolec type1 cable plus an adaptor to type 2
(kids on xboxes not happy...)
Went into house - reset rcd, tried again Immediately and it was fine!

 

[Mr Miserable]

I'm inclined to do some leakage current measurements when using the UMC when I get my car next week. I've a couple of spare current transformers, so could have a go at looking at the current waveform as well, to see if there's anything odd going on. It'd only be a single data point, but it might give some clues as to what may be happening.

Good idea. Must get out my spare current transformers...

 

[rotor2k]

After all this talk of RCDs tripping and extension cords melting, I decided I'd best test my granny charger in case it also needed replacing. But I'm happy to report it seems to be working well, saying it's going to take my LRAWD from 78% to 90% in *just* 7 hours! I can't wait for my 32A charger to be installed... Also kind of terrifying to see this amount of power being consumed in my house:

 

[WannabeOwner]

kind of terrifying to see this amount of power being consumed in my house

Might well be as much as 5x cheaper than petrol ...

EDIT: You need to do something about that 14.71p/kWh for Leccy, Off peak rates should be around 5p-8p

 

[rotor2k]

Might well be as much as 5x cheaper than petrol ...

EDIT: You need to do something about that 14.71p/kWh for Leccy, Off peak rates should be around 5p-8p

Haha that screenshot is from 11am. I just wanted to test the granny charger to make sure it worked. I’m not keen on an orange extension cord hanging out my letterbox overnight.

 

[MrBadger]

It could just as easily be something common plugged into both houses. I got ours working by turning off two white appliances, both made by the same manufacturer.

This is probably a red herring and just luck first time around. Thanks to @NorfolkMustard I was able to try things on a different car (but same UMC) but even fully unplugging the offending white goods, we still tripped the RCD. We know his car is good, so just the uncertainty of our UMC. I'm hoping that I can convince Tesla to borrow a UMC.

I've also asked a sparky about changing our RCD - maybe he will find something in testing that.

 

[arg]

I've also asked a sparky about changing our RCD - maybe he will find something in testing that.

It would not be at all surprising to change the RCD for another one of nominally the same type and the problem just goes away, unexplained.
There's quite a range of acceptable tripping current, and if you add in to the mix leakage currents that aren't pure 50Hz then variation between RCDs can be significant.

One experiment it would be nice to do would be to try UMCv1 vs UMCv2 - the combination of Model3 + UMCv2 seems to cause more problems than Model S + UMCv1, but it's not clear how the blame divides between the UMC and the car.

 

[WannabeOwner]

This is probably a red herring and just luck first time around.

My Model-3 UMC tripped In laws house (repeatably : garage [separate distribution board] and a couple of sockets connected to main distribution board] , but not neighbours, nor home.

"Neighbours" was using a very long extension (car wasn't moved ) ... dunno if that helped )

 

[MrBadger]

Maybe I will see if I can borrow a gen 1 UMC. I'm having snagging done tomorrow so maybe they will have one to try - hopefully pickup from Brooklands which is closer to return to.

 

[Glan gluaisne]

FWIW, I turned everything on in the house this morning, and without the charge point turned on the total house leakage current is about 17.5 mA. By isolating loads one by one I concluded there was no single culprit, each load circuit seemed to have between 1 and 2 mA of leakage. The PV inverter seems to have a bit more, measuring that (it's on it's own circuit) gave an earth leakage of about 7 mA. I expected our ASHP to be around the same, as it's inverter controlled, but it seems to only around 2 mA.

I'll have a go at measuring the earth leakage from the UMC/car onboard charger this weekend, to see how much that adds. The max allowable for a class 1 appliance of this category is (I think) 3.5 mA, so if the UMC/onboard charger leakage exceeds this I think it may not be compliant, as any equipment with a leakage current in excess of 3.5 mA cannot be used with a 13 A plug.

It's hard to see how an additional leakage current of just 3.5 mA could be enough to take the cumulative leakage up to 30 mA, as I suspect our house has a higher than typical total earth leakage so may be close to a worst case. We have loads of LED lights with switched mode supplies, MVHR with switched mode motor controls, ASHP with inverter control, an electric water heating system and I had every device charger plugged in for this test, as they all have switched mode supplies.

 

[MrBadger]

so if the UMC/onboard charger leakage exceeds this I think it may not be compliant, as any equipment with a leakage current in excess of 3.5 mA cannot be used with a 13 A plug.

What is the implications of this? Not tripping wise but non compliance wise?

 

[Glan gluaisne]

What is the implications of this? Not tripping wise but non compliance wise?

The UMC and onboard charger should be compliant with Section 543.7.1.201 of BS7671:2018:

543.7.1.201 Equipment having a protective conductor current exceeding 3.5 mA but not exceeding 10 mA, shall
be either permanently connected to the fixed wiring of the installation without the use of a plug and socket-outlet or
connected by means of a plug and socket-outlet complying with BS EN 60309-2.

BS EN 60309-2 refers to a commando plug. Anything with a 13 A plug must have an earth leakage current that's less than 3.5 mA.

 

[RealEvil_UK]

FWIW, I turned everything on in the house this morning, and without the charge point turned on the total house leakage current is about 17.5 mA. By isolating loads one by one I concluded there was no single culprit, each load circuit seemed to have between 1 and 2 mA of leakage. The PV inverter seems to have a bit more, measuring that (it's on it's own circuit) gave an earth leakage of about 7 mA. I expected our ASHP to be around the same, as it's inverter controlled, but it seems to only around 2 mA.

I'll have a go at measuring the earth leakage from the UMC/car onboard charger this weekend, to see how much that adds. The max allowable for a class 1 appliance of this category is (I think) 3.5 mA, so if the UMC/onboard charger leakage exceeds this I think it may not be compliant, as any equipment with a leakage current in excess of 3.5 mA cannot be used with a 13 A plug.

It's hard to see how an additional leakage current of just 3.5 mA could be enough to take the cumulative leakage up to 30 mA, as I suspect our house has a higher than typical total earth leakage so may be close to a worst case. We have loads of LED lights with switched mode supplies, MVHR with switched mode motor controls, ASHP with inverter control, an electric water heating system and I had every device charger plugged in for this test, as they all have switched mode supplies.

Thanks so much for this - fascinating. I tried in another house and it worked so it looks like a cumulative leak. But what are chances of 2 houses being so close to 30ma limit... All seems unlikely to me that it's only leaking 3.5ma but who knows ?

 

[Glan gluaisne]

Thanks so much for this - fascinating. I tried in another house and it worked so it looks like a cumulative leak. But what are chances of 2 houses being so close to 30ma limit... All seems unlikely to me that it's only leaking 3.5ma but who knows ?

It may well be as @arg suggested, and be a combination of leakage current, plus maybe some artefacts from the car onboard charger, maybe relating to odd frequency related effects. As well as just looking at the AC leakage current I also tried looking at the leakage current waveform with a 'scope connected to a clip-on current transformer. On the house CPC all I could see was a pretty clean 50Hz leakage current. On the PV inverter there was a fair bit of high frequency noise current, presumably being coupled to earth via line filters, I'd guess.

What we don't know is exactly how RCDs react to high frequency noise. I do know that some can trip very easily when running switched mode inverter motor drives, and also know that swapping out an RCD prone to nuisance trips on kit like this for one of the Schneider SI series resolved the problem completely. This suggests that some RCDs may be more prone to tripping when there are some high frequency artefacts on the supply than others.

 

[WannabeOwner]

What we don't know is exactly how RCDs react to high frequency noise

Ignorant question : are RCDs available which are "more tolerant" ?

 

[MP3Mike]

Ignorant question : are RCDs available which are "more tolerant" ?

Did you read the next sentence after what you quoted?