Hello guys, I want to ask you for help with connecting these inverters in parallel. My procedure is as follows:
1. In the menu setting point 28. for parallel connection on both.
2 turn off the power / turn on the power
3. I start the inverters and then I get error (51 Over current or surge) or (81 Host loss)
I have parallel boards connected according to the manual.(foto)
I tried to leave the circuit breaker on the AC output turned off. I see that one is set up as master and the other as slave, so I assume that they probably communicate together. But if I turn on the circuit breakers, there is an error again, as if they could not synchronize. Even IGBT transistors have already been destroyed twice, I replaced them  and the inverters work normally, but only individually, not in parallel connection. Of course, the inverters have the same firmware. Another thought occurred to me, is it necessary that they be connected to the  local grid? Because I don't have them connected to the distribution grid.  Thank you for help 

 

 

 

9 hours ago, Miror said:

1. In the menu setting point 28. for parallel connection on both.

Do you do this with the load switch off, and verify that the setting sticks?

FAQ #3

9 hours ago, Miror said:

3. I start the inverters and then I get error (51 Over current or surge) or (81 Host loss)

Please don't power up the inverters with the outputs connected together until you are SURE that everything is working. I like to power up with no output breakers on, and measure the difference between the L outputs with a multimeter on AC voltage range. The difference should be less than about 5 VAC. Then and only then am I confident that the outputs are synchronised enough to turn on the output breakers.

9 hours ago, Miror said:

I see that one is set up as master and the other as slave, so I assume that they probably communicate together.

That's a good sign that they are both in output mode PAL (parallel).

9 hours ago, Miror said:

is it necessary that they be connected to the  local grid?

No, they should be communicating phase over the CAN bus (the light grey cables). There is a small chance that your paralleling boards aren't communicating properly over that channel. There are about 3 channels from memory. I don't know if the firmware can detect such a fault.

9 hours ago, Miror said:

Even IGBT transistors have already been destroyed twice, I replaced them  and the inverters work normally,

Please test them thoroughly by themselves, to say 80% of rated power. That should eliminate any chance that one inverter is "limping along on three legs" or similar. Then do the multimeter test as I mention above and report the voltage difference. It would seem that your inverters aren't synchronising for some reason.

How are your AC-out neutrals wired? Are your output breakers single or double pole? If double pole, the multimeter voltage difference measurement likely won't work, since you have no AC-in wiring.

3 hours ago, Coulomb said:

How are your AC-out neutrals wired? Are your output breakers single or double pole? If double pole, the multimeter voltage difference measurement likely won't work, since you have no AC-in wiring.

Thank you for the quick answer.

I have a circuit breaker L and N output on both inverters separately. If I understand it correctly, I should turn on the N breaker on both inverters and then measure the voltage difference on the L outputs, and if it is less than 5V, then I can also turn on the breakers on L. Is this correct?

40 minutes ago, Miror said:

I should turn on the N breaker on both inverters and then measure the voltage difference on the L outputs, and if it is less than 5V, then I can also turn on the breakers on L. Is this correct?

Yes. In your case, having blown IGBTs repeatedly, I'd like to leave the multimeter connected if that can be arranged safely, and watch the voltage for perhaps 30 minutes or so. That way you can be sure that you didn't fluke a synchronisation, and they are likely to stay synchronised.

Edit: Or just leave the inverters on and sample the difference every  few minutes. The reading isn't likely to change quickly. Of course, if the first reading is way over 5 VAC, then you don't need to monitor for longer, but try and figure out what is wrong.

Edited December 20, 20232 yr by Coulomb

Thanks, I'll try and let you know

So I tried it and the voltage difference between L and L is 2.7V. That should be fine, right?

10 hours ago, Miror said:

the voltage difference between L and L is 2.7V. That should be fine, right?

Yes, as long as it stays that low.

Now the question is, why did your IGBTs blow up twice? Or do you have reasons for that, and they are corrected now?

Edited December 21, 20232 yr by Coulomb

I watched it for about 15 min. and it was still around 2.7V. I'm afraid to turn it on, because I didn't change anything in the connection of the inverters after the IGBTs were destroyed. Is it possible to measure something else so that I can be sure that it is OK and I can turn it on? Because I don't have any more IGBTs transistors .😁

5 hours ago, Miror said:

Is it possible to measure something else so that I can be sure that it is OK and I can turn it on?

Sorry, I have no further suggestions. I think if they blow again now, then there is some serious problem that is impossible to diagnose remotely.

@BritishRacingGreen, have I overlooked anything? I'm assuming that you've carefully checked all the parallel cabling.

The only other thing might be buying a new pair of paralleling boards and cables. These are possibly cheaper and a lot easier to replace than another 8 IGBTs.

Thanks for the help and if I find the courage to turn them on, I'll let you know what happened.

The other thing that you could do is to use my partial schematic trace of the parallel boards to test each set of opto couplers. This would give you more confidence about keeping the smoke in. 

You'd need a bench power supply and some electronics skills.