question

kw4np avatar image
kw4np asked

Multi RS 48/6000/100 Voltage drop with no errors noted

20240621-0600_to_20240621-0614.xlsx20240621-0600_to_20240621-0659.xlsxvoltagedrop.pngThe consumer reported that the Multi RS shut off in the middle of the night for about a minute before restarting. The load was right around 50 watts at the time, so shouldn't have been an overload. I expected to see some error logs on VRM, but nothing was logged except a voltage drop from 220 to 120. Any idea what caused this event? Everything else seemed fine, my only thought was AES or low load shutoff, but I don't see that in settings.

Multi RS
20240621-0600-to-20240621-0614.xlsx (7.7 KiB)
20240621-0600-to-20240621-0659.xlsx (8.4 KiB)
6 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

kw4np avatar image kw4np commented ·
And also why in the world were no errors logged? Seems like the inverter wouldn't shut down for no reason...
0 Likes 0 ·
andrii-podanenko avatar image andrii-podanenko commented ·
What battery do you have?
0 Likes 0 ·
kw4np avatar image kw4np andrii-podanenko commented ·
@Andrii Podanenko I have four EG4 Lifepower4 batteries. They are not communicating with the Cerbo or Multi. They worked fine before with a different inverter.
0 Likes 0 ·
andrii-podanenko avatar image andrii-podanenko kw4np commented ·

what SOC level they were at?

I mean if they were at 100% SOC but Victron though 82% - this could be an issue

I see V is at 53V DC - this is enough for 15S to be at 100% SOC.
Charging at this SOC would call BMS to stop charging and disconnect battery from inverter.

0 Likes 0 ·
kw4np avatar image kw4np andrii-podanenko commented ·

@Andrii Podanenko I can see that possibility, but at 1 AM they were probably at least partially discharged and no chance of charging from solar. I don't have any way to check the battery SOC that accurately, but it seems reasonable that they were close to what the system was calculating.

0 Likes 0 ·
kw4np avatar image kw4np commented ·
So this has now happened twice around the same time of night and I'm waiting on Victron support for answers. The correction in battery SOC happened during a manual shutdown, so seems like that is a symptom rather than a cause.
0 Likes 0 ·
2 Answers
Alex Pescaru avatar image
Alex Pescaru answered ·

Hi @KW4NP

Very strange indeed...

For errors, you can look in several places:

1. In VRM, but you already looked there

2. In the VictronConnect, which reads, when connecting, Multi RS internal memory where are stored Multi RS past errors and warnings.

3. In Cerbo's logs, where, even if you can't find the specific peripheral's errors, maybe you can get some ideas of what happened around that time frame.

Wouldn't hurt to take a look at the battery storage if they have the ability to provide logs and see around that time if something happened.

Also, I believe that you are sure that at that hour the Multi RS was not in Passthrough mode and then maybe the grid has played a trick on the owner...

Anyway, I will be looking with interest at the topic, being owner of Multi RS...

Alex

4 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

kw4np avatar image kw4np commented ·
Hello @Alex Pescaru , Thanks for your response!

I did already check VRM, checking further shows that the internal relay state was off for one minute, the battery SOC (calculated by the Multi) corrected upward by exactly 6 percent, and the Multi indicated that the state was off for that minute and then a few minutes later again off for around 5 hours (which is not true). I will include logs from VRM in the question in case anyone wants to look at them.

I have logged into VictronConnect remotely and have not seen any errors.

I am remote so have not checked Cerbo logs, I don't think it's saving any right now.

There's no battery monitoring currently.

System is off grid so it couldn't be in passthrough mode.


0 Likes 0 ·
Alex Pescaru avatar image Alex Pescaru kw4np commented ·
Up 6% when switching from load to no load means, sometimes, that the battery has problems or some connection is bad (high resistance).
0 Likes 0 ·
kw4np avatar image kw4np Alex Pescaru commented ·
It's always 6%? All connections were good and the batteries were working fine.
0 Likes 0 ·
kw4np avatar image kw4np Alex Pescaru commented ·
Also, I'd expect to see a low voltage or some type of error rather than just turning off.



0 Likes 0 ·
Alex Pescaru avatar image
Alex Pescaru answered ·

Hi @KW4NP

Now I've managed to take a look at the xlsx files posted in the first message.

Those over 80 seconds VRM log time offset at 01:07 - 01:08, leads me to think that Cerbo has became very busy (CPU likewise) at that time and it's possible to have not sent on the CAN bus to Multi RS the status and working parameters.

Therefore, if the MultiRS is missing the Cerbo "beat" for more than 1 minute, will shutdown. It's very likely that this was the reason for that blackout. I can say this for certain, because it also happened to me, but because of my fault, when I've disconnected on the fly the communication cable from Cerbo to Multi RS at some point. Don't know if this is normal or not, it's by design or not, but this is how it's behaves.

But in your case it's clear that the Cerbo was very busy at that time. Do you have any user defined scripts that could make the Cerbo to hang up?

Could you also verify in the logs, in VRM, for the Gateway's (Cerbo's) "D-Bus Round-trip time" for that period of time?

Thanks,

Alex

5 comments
2 |3000
Toggle Comment visibility. Current Visibility: Viewable by all users

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

kw4np avatar image kw4np commented ·
Hi @Alex Pescaru , Thanks for taking a look at those logs. Interesting that this has happened to you before, not a good sign for stability in my opinion. I don't have any scripts that I'm using, I'd tried to install the RS485 battery driver a while back but never actually got it working. So nothing that should be making the Cerbo GX glitchy. I'm not sure how to access the logs you're referring to, could you share more details?

Thanks!


0 Likes 0 ·
Alex Pescaru avatar image Alex Pescaru kw4np commented ·

Happened indeed, but like I've said, it was following an action of mine, that normally isn't happening.... Manual disconnection of Cerbo - Multi RS communication.

Go to VRM -> Advanced -> Widgets (on upper right corner).

Custom Widget Add (+) -> Create custom widget.

Select device -> Gateway.

Select parameter -> D-Bus Round-trip time.


1 Like 1 ·
kw4np avatar image kw4np Alex Pescaru commented ·
Thanks for that tip, I didn't know that there were more logs available. However, that doesn't seem to be abnormal. D-Bus RTT stayed at 1 ms.
0 Likes 0 ·
Alex Pescaru avatar image Alex Pescaru kw4np commented ·

Well, that's good...

You can try to simulate what happened... Agree with the customer and disconnect, on the fly, the Cerbo to Multi RS comm line and see if it's happening again.

Simulate, simulate, simulate... Rocket science... :-)) Kidding...


1 Like 1 ·
kw4np avatar image kw4np Alex Pescaru commented ·
@Alex Pescaru , I might give it a try. But there's no reason to lose connection normally. I'm talking to Victron support and if we find something then I'll try to report back.
0 Likes 0 ·