Why did my Tesla throttle charge to 24/32 Amps

[MrBadger]

As you have TeslaFi, it may be worth investigating whether you can bump the charge rate at various points during the evening. This might workaround an artificial charge rate drop if it becomes an issue and you are sure the rate drop is not for safety reasons. It may not work at all if it’s intentional but no different to using charge rate in app, just automated.

I do a timed TeslaFi rate boost back to 10A at start of a charge at 00:31 (start of Go) to history to get around the issue where charge limit use to reset to 9A or I had been charging from solar at lower rate and forgot to revert. Will need to revisit that when I swap to IO.

 

[Astrape]

I agree, but its the sort of thing where OTA can fix it once the problem is know ... and having 3rd party Elastoplast solution does at least allow working around the problem. Back in the dark ages a problem like this would have required "upgrading to latest model" ... bringing with it a new set of bugs/compromises!

Not heard of it before on the airwaves, so might be a physical problem that needs fixing?

It’s a thing my search has found mentioned in quite a few threads where owners have local voltage that gets dragged low and also triggers intermittent current restriction. The car is trying to detect high resistance (hot / overloaded) connections by looking for a particular pattern of voltage droop. This post is a good summary.

Tesla charger - any known issues with occasional low supply voltage?

So my car may have restricted the current because
1) It spotted a poor connection in my installation
2) There was a voltage fluctuation that it mistakes as a poor connection
3) The voltage went to a too low level

3 seems unlikely since the car can be charged at 208V nominal in the US, so must be capable of accepting voltages down to the 180s. It’s permitted to being US cars to Europe and vice versa so its the same circuit.

1 also seems unlikely since its a new professional install which had no issues so far. However its worth keep an eye on.

2 seems the most likely since the supply was at its amps limit and so at its biggest voltage drop, and other posts indicate that the car seems to mistake low voltage dips for poor connections. As the load / current changes the voltage drops varies, and the car may be mistaking the voltage drop as bad connection droop.

Physical problem? Yes it may be there is a real physical problem.
In the supply? I will check for hot joints.
In the car? Maybe since it threw a hardware error. No reply from service on that yet.

Elastoplast workaround.... Yes you are quite right its amazing that we have third party automation provision to impose our own solution. As an engineer the one things that makes me cautious is “fixing” a reduced current that is intended as a safety feature. But yes a single return to 32A is not going to make much difference. Probably I can program that at 1am after the time I am likely to be causing the extra going to bed water heating that caused the event.

 

[Astrape]

No. Once it throttles it stays there. Stopping and restarting the charger will get it back to 32A, and sometimes setting the charge to 5A for a minute and then back to 32A will also fix it.

When the car is reducing the current does it show 24/24A or 24/32A in the app?

 

[Astrape]

We have patchy quality mains supply and I've certainly had instances in the past where the charge has needed to back off but then returns to 32A later in the charge. I've also had occasional instances where the amps have been lowered and then doesn't return later in the charging session. Whether this is "another case of poor Tesla coding" is impossible to know unless you have access to the Tesla coding!

Fair comment. I was following the logic that multiple posts report that in installations where there is high voltage drop there is also current limiting, which appears to be the car misinterpreting voltage drop as a high resistance connection in the supply. So it would appear that the coding is causing unnecessary reduction in current.

But to give Tesla the benefit of the doubt it could be that in all those cases there is actually poor joint in the supply that the car is protecting, and as you say without deeper investigation we cant be sure.

 

[Astrape]

I read all the posts I could find on slow/limited charging. In the US there are a lot of cases of this happening due to plugging into 40A wall sockets being allowed! So in those cases the car is detecting real poor connection issues in the supply. It may be that this code to detect hot connection is focussed on a problem common in the US due to their not requiring dedicated wall chargers.

UMC Charger AMP went down from 32 to 16

“[Can be due to hot garage / Mine went away when colder]... However I also see this happen to others for loose outlet / plug, not dedicated circuit and other items also drawing a load, undersized panel getting close to main breaker tripping, damaged charging cable due to being twisted back and forth too many time or stepped on.”

 

[WannabeOwner]

also seems unlikely since its a new professional install which had no issues so far.

I agree, although after we had whole house rewired we had one subsequent failure. I've forgotten details now but I think it might have been when we (i.e. some time later) plugged the Xmas tree lights in. Turned out one strand of copper had missed the ring&clamp in the 13AMP socket. I have a big house so trades are large outfits, not spotty youth! so all sorts of fancy testing equipment didn't locate it when testing the new wiring. Might be that when I plugged in the lights that caused the wire-strand to move ...

 

[Sean.]

How did you arrive at the conclusion that you have a 100 amp limit? What size is your main breaker? What size is the transformer feeding your unit/house? Is that transformer shared with other units/houses?

I ask because if you have a 100 amp breaker/fuse, your maximum limit should be 80 amps.

If you’re on a 15 kVA transformer, your limit is actually ~65 amps. If you’re on a 25 kVA transformer that’s shared with your neighbor, your limit is ~110 amps minus whatever they’re using.

It’s different here in the UK @Big Earl - for a start we don’t share transformers with neighbours here. They serve much larger groups of houses, estates typically. Our LV supply from these is 415v three phase on the street and most domestic connections are 230v single phase with a 60, 80 or 100A sized cutout fuses from the distribution network operator.

 

[Sean.]

Our local supply is getting worse in fact. I’ve even seen a voltage of under 200V on one occasion. The local network is pretty useless with lots of properties on some weedy looking PMTs. Usual crap British infrastructure.

That a definite complaint to the DNO to upgrade the transformer or at least set its output correctly.

 

[Big Earl]

The car derates by 25% when it detects a 10% drop from the voltage at which it started charging. If it was 237 when the car started, it will derate at 213 volts. If it was 230 when it started, it will derate at 207 volts. The car will happily charge at anything from about 100 volts to 285 volts as long as it's within 10% of the starting voltage.

 

[Big Earl]

That a definite complaint to the DNO to upgrade the transformer or at least set its output correctly.

It seems to me that the output is set correctly - it's just undersized for what it's being asked to do (feed 22 houses). It was probably adequate when it was installed, but people have added electric vehicles and time-shifted a lot of their use to off-peak times, which probably didn't exist when the transformer was installed. Utilities don't usually upgrade transformers and distribution until there is a problem, which the OP appears to be encountering.

I am still skeptical about loading a 100 amp fuse to exactly 100 amps continuous. This would be against code in North America, but perhaps the UK is different. I don't know. Either way, it's probably worth having a conversation with the utility.

 

[Sean.]

It seems to me that the output is set correctly - it's just undersized for what it's being asked to do (feed 22 houses). It was probably adequate when it was installed, but people have added electric vehicles and time-shifted a lot of their use to off-peak times, which probably didn't exist when the transformer was installed. Utilities don't usually upgrade transformers and distribution until there is a problem, which the OP appears to be encountering.

I am still skeptical about loading a 100 amp fuse to exactly 100 amps continuous. This would be against code in North America, but perhaps the UK is different. I don't know. Either way, it's probably worth having a conversation with the utility.

Yeah a regular single phase supply here is 230v +10% and -6% at the incomer. If it’s dipping beneath that here then the DNO should be made aware, as it’s clearly going to have downstream issues with certain equipment in the premises.

By rights here we’re supposed to notify the DNO when connecting EVSE here, and certain other equipment / generation and it should be kept on a database shared between the DNOs.

 

[Astrape]

The car derates by 25% when it detects a 10% drop from the voltage at which it started charging. If it was 237 when the car started, it will derate at 213 volts. If it was 230 when it started, it will derate at 207 volts. The car will happily charge at anything from about 100 volts to 285 volts as long as it's within 10% of the starting voltage.

Oh well thats not good. My car will start to charge at 11:30 when IO schedules its first slot. It may well see 240 volts with light evening load. That will fall to 230V with the voltage drop in the garage run, and when the central heating kicks in the 20kW will cause about another 10V drop to 220V. It only takes a little voltage sag from other user load and it will have dropped 10%. Nothing is wrong and the supply is still within normal operating limits. To counter that I would have to restart a new charge after my night storage heaters kick in.

Is this a permanent reduction for the rest of that charge?

What is the source of this information, is it documented?

 

[pdk42]

When the car is reducing the current does it show 24/24A or 24/32A in the app?

24/32

 

[pdk42]

That a definite complaint to the DNO to upgrade the transformer or at least set its output correctly.

Yeah, I know. I’ve complained and complained. They keep saying they’ll fix it, but all I hear is tumbleweed.

 

[pdk42]

The car derates by 25% when it detects a 10% drop from the voltage at which it started charging. If it was 237 when the car started, it will derate at 213 volts. If it was 230 when it started, it will derate at 207 volts. The car will happily charge at anything from about 100 volts to 285 volts as long as it's within 10% of the starting voltage.

That’s very useful info. Explains a lot. Thanks!

 

[WannabeOwner]

That will fall to 230V with the voltage drop in the garage run, and when the central heating kicks in the 20kW will cause about another 10V drop to 220V

Dunno if this sheds any light, but I looked for occasions when I was pulling 23kW (highest I could find) from the grid (2x PowerWall charging and 1 or 2 EV). TeslaFi logs differing charger voltages for each car, one is Tesla wall charger and the other Zappi. My conclusion is that the AMPs did recover (which I don't think you have seen in your case?)

 

[Astrape]

It seems to me that the output is set correctly - it's just undersized for what it's being asked to do (feed 22 houses). It was probably adequate when it was installed, but people have added electric vehicles and time-shifted a lot of their use to off-peak times, which probably didn't exist when the transformer was installed. Utilities don't usually upgrade transformers and distribution until there is a problem, which the OP appears to be encountering.

I am still skeptical about loading a 100 amp fuse to exactly 100 amps continuous. This would be against code in North America, but perhaps the UK is different. I don't know. Either way, it's probably worth having a conversation with the utility.

Seans comment was not to me it was to pdk42.

Drawing 100% of the feeder amps is permitted in the UK as has been explained. The network operator came and pulled the fuse to confirm its rating. The car charger installation approval was submitted with a total load of 100A demand and approved by the network operator. The transformer is supplying correctly within it 230 +10% -6% limits.

The problem here appears to be the car which is unreasonably reducing current to the car in response to the voltage falling by 10% despite the fact that that the supply is still within regulations for the UK.

 

[Astrape]

Dunno if this sheds any light, but I looked for occasions when I was pulling 23kW (highest I could find) from the grid (2x PowerWall charging and 1 or 2 EV). TeslaFi logs differing charger voltages for each car, one is Tesla wall charger and the other Zappi. My conclusion is that the AMPs did recover (which I don't think you have seen in your case?)

View attachment 906640

Ah very interesting. If that last one is caused by 10% voltage drop it looks like it will recover if the voltage rises back to its original level. That wont happen in my installation while I am charging due to the overall demand.

In the car 1 trace it looks like the car is testing the voltage drop and decides its ok, returning to full current. Although that seems rather a long “test” process.

I have no idea what that car 2 trace means, is that one feeling okay ?

How much harmonics do the power wall rectifiers produce? It should all work together but part of my job is dealing with interference caused by inverter infeeds.

 

[WannabeOwner]

n the car 1 trace it looks like the car is testing the voltage drop and decides its ok, returning to full current

Slightly misleading ... the TeslaFi data is missing some data elements for Car #1 (unusual compared to "whole records missing") 12:29 - 12:57, so TeslaFi has basically assumed straight-line from the high value to the low value, and back again, where that dip is. So might have actually fallen lower than that. But either way its only a few minutes, whereas Car 2 stays low for much longer ... maybe car #1 (as in "Its wall charger") manages to grab what it needs, and/because Car #2 drops its AMPs ... but after Car #1 finishes its still an hour until Car #2 increases AMPs

In case it clarifies anything here's what 09-Dec looked like from PowerWall perspective

Interesting that PowerWall throttles up and down, not checked whether it does that every night ...

 

[Astrape]

Slightly misleading ... the TeslaFi data is missing some data elements for Car #1 (unusual compared to "whole records missing") 12:29 - 12:57, so TeslaFi has basically assumed straight-line from the high value to the low value, and back again, where that dip is. So might have actually fallen lower than that. But either way its only a few minutes, whereas Car 2 stays low for much longer ... maybe car #1 (as in "Its wall charger") manages to grab what it needs, and/because Car #2 drops its AMPs ... but after Car #1 finishes its still an hour until Car #2 increases AMPs

In case it clarifies anything here's what 09-Dec looked like from PowerWall perspective

View attachment 906668

Interesting that PowerWall throttles up and down, not checked whether it does that every night ...

What load management is set in each of the devices?