Why AIO manufacturers warn against grounding battery terminals on AIOs (EASUN, PowMR, SRNE etc) and multiple AIOs on one battery bank.

[luk88]

I'm planning to use multiple AIO inverters powering independent loads with one battery bank (EASUN and Y&H). After researching the topic online including this board I believe this should be fine (providing few small charging issues are met and things like not exceeding max currents).

Also after reviewing internal pictures and reading lots of service manuals I noticed all modern AIOs (with HV DC Bus) use a transformer isolated full bridge DC to DC converter for their Battery to HV DC bus connection. So all of them isolate their battery. 

An example schematic is below:

Sometimes there is one transformer. Sometimes there are two (one for the plus leg, one for the minus leg to double the power capacity). But I never saw a non-isolated DC-DC converter for the battery.

So if this is the case why are manufacturers warning against grounding the battery? I've recently asked EASUN about grounding the negative terminal of the battery and they replied that "a short will result and damage to the inverter". Are they simply confusing the PV input with the battery? 

Does anyone know?

@Coulomb @BritishRacingGreen no doubt you know the answer to this question. I hope you don't mind me tagging you here

 

Edit: This EASUN is a 7.2KV clone of Axpert MKSII, The Y&H I'm planning to use in paralell with it are clones of Axpert MAX 10.2kW.

Edited July 29 by luk88

[Coulomb]

17 hours ago, luk88 said:

So if this is the case why are manufacturers warning against grounding the battery? I've recently asked EASUN about grounding the negative terminal of the battery and they replied that "a short will result and damage to the inverter". Are they simply confusing the PV input with the battery?

They are wrong: grounding the battery will not harm the inverter.

I think that they are responding to a general uneasiness about grounding high power batteries. If you do this with a hard (low resistance) connection, then any accidental short to earthed metalwork involves at least a big noise, UV radiation, possible molten metal, and a possibly hazardous arc that can be hard to extinguish. 48 V isn't too bad; the arc will self extinguish if the separation increases to about 48 mm. With higher voltage batteries, it becomes more difficult to quench an arc once it starts.

So generally, high voltage batteries are usually kept isolated. When the voltage gets high enough, it usually is supplemented with automatic leakage detection, so that leakage from one point (which does not result in shock or arc) is detected and fixed, so that you don't get a leakage and a touch.

17 hours ago, luk88 said:

This EASUN is a 7.2KV clone of Axpert MKSII, The Y&H I'm planning to use in parallel with it are clones of Axpert MAX 10.2kW.

Axpert firmware only allows paralleling of identical hardware, with firmware versions that are the same or very close. It's never possible to parallel a 7.2 kW model with a 10.2 kW model, for example. Also, an MKS II or clone thereof can never be paralleled with any MAX or clone thereof, even if the rated power is identical.

[luk88]

1 hour ago, Coulomb said:

They are wrong: grounding the battery will not harm the inverter.

I was almost certain of it, but when you're repeatedly told otherwise one may doubt.

1 hour ago, Coulomb said:

I think that they are responding to a general uneasiness about grounding high power batteries. If you do this with a hard (low resistance) connection, then any accidental short to earthed metalwork involves at least a big noise, UV radiation, possible molten metal, and a possibly hazardous arc that can be hard to extinguish. 48 V isn't too bad; the arc will self extinguish if the separation increases to about 48 mm. With higher voltage batteries, it becomes more difficult to quench an arc once it starts.

So generally, high voltage batteries are usually kept isolated. When the voltage gets high enough, it usually is supplemented with automatic leakage detection, so that leakage from one point (which does not result in shock or arc) is detected and fixed, so that you don't get a leakage and a touch.

This is an interesting take. Perhaps it was the original intent of the manufacturer.

1 hour ago, Coulomb said:

Axpert firmware only allows paralleling of identical hardware, with firmware versions that are the same or very close. It's never possible to parallel a 7.2 kW model with a 10.2 kW model, for example. Also, an MKS II or clone thereof can never be paralleled with any MAX or clone thereof, even if the rated power is identical.

I meant paralleling only the battery connection. Perhaps I should've clarified, but I thought it is obvious  as I said in the first sentence of the post:

18 hours ago, luk88 said:

I'm planning to use multiple AIO inverters powering independent loads with one battery bank (EASUN and Y&H).

However most people when they talk about "pararelling" mean connecting battery and loads together (or setting up loads as synchronised phases). In this sense you're of course right.

Coming back to paralleling-only-the-battery side, unsurprisingly it works fine. I've since connected and tested it in all modes. I plan to resolve the charging issue in the short term by having only one inverter have higher bulk charging voltage than the others. In long term by settings up a "fake bms" connection to all inverters.