question

Paul Beskeen avatar image
Paul Beskeen asked

Parallel Multiplus Master failure at end of scheduled charge

Multiplus II dual parallel 48/5000/70 with Pylontech battery bank. See https://community.victronenergy.com/questions/177756/parallel-multiplus-ii-installation-and-config.html for setup details.

The system has been working fine up to now, but in the last few nights I have encountered a strange anomaly at the end of the ESS scheduled charge period - luckily I have been there to catch it.

As soon as the charge ends the master of the pair simply stops inverting and the second unit starts outputting too much current, such that the grid feed-in goes up to around ~600-700w and sticks there!

Using a clamp meter I can see that NO current is flowing on the DC input to the master, current is being consumed ONLY by the second unit.

The system does not report any error condition!

Rebooting the Cerbo does nothing, the only way to fix seems to be powering cycling the inverters (via virtual switch), at which point sanity is restored and everything works fine again.

I have the latest firmware for the MP-II's (501) and latest Cerbo (v2.93) update. I have previously tried the beta firmware (v3.0~18) and this also exhibits this failure. I checked and when I reverted to v2.93 the MK3 firmware was reported as reverted by the Cerbo to 212.

Is this a hardware failure requiring an RMA, or a result of the recent v2.93 update, or as a result of swapping to and back from beta firmware, or some other configuration issue or aspect I wasn't aware of?

Thanks, Paul.

photo-2023-02-21-11-28-12.jpg

photo-2023-02-21-11-28-12-3.jpg

photo-2023-02-21-11-28-12-2.jpg

Multiplus-IImultiplus in parallel
photo-2023-02-21-11-28-12.jpg (141.0 KiB)
photo-2023-02-21-11-28-12-3.jpg (149.2 KiB)
photo-2023-02-21-11-28-12-2.jpg (147.5 KiB)
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.

3 Answers
nickdb avatar image
nickdb answered ·

What are the first 4 digits (after HQ) of the unit's serial numbers?

What you're describing doesn't sound like anything a GX could influence.

1 comment
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.

Paul Beskeen avatar image Paul Beskeen commented ·
Master and second unit are both HQ2234. So relatively recent manufacture and should be compatible with each other for parallel use.
0 Likes 0 ·
nickdb avatar image
nickdb answered ·
You are good with that.

The obvious things are to make sure all the cabling is symmetrical, the inverter configs also and that there aren't any losses on one side vs the other.

Failing that, you can reflash them and reconfigure them with the slave unit as master, and see if that changes anything.

3 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.

Paul Beskeen avatar image Paul Beskeen commented ·

Yep, all cabling is symmetrical, the system is balanced within ~6% - which hopefully is "good enough"? - this is measured with clamp meter on DC and AC lines. I will rerun VEBus Quick Configure again on both and ensure that they have identical configs, but it's not as if anything should have changed! This was all running fine... When that doesn't work I'll try swapping their roles and see if that has any effect.

0 Likes 0 ·
nickdb avatar image nickdb ♦♦ Paul Beskeen commented ·
6% isn’t bad. Mine are closer together but it all comes down to the cabling and connections. Sometimes the cables can be too short or too thick, that counterintuitively can spoil the symmetry.
0 Likes 0 ·
Paul Beskeen avatar image Paul Beskeen nickdb ♦♦ commented ·

Success! Today I swapped master and second unit roles around as suggested. Loaded and set exactly the same VE Configure config as I had used for previous setup. The only difference should just be the roles the Inverters play. Tonight saw no aberrant behaviour, nice and stable operation when the ESS exited from the scheduled charge.

Hopefully the inverter roles this way round will prove to be reliable, I'll have to keep an eye on it for a while. I don't like unexplained behaviour.

The measured imbalance seemed higher at perhaps ~8% rather than ~6% - when the charge cycle is running flat out. It's still the same physical unit that is drawing more power. I also saw the highest battery charge rate I had yet seen at 146.7A - which is slightly higher than the stated rating (70A charge each)...

Cheers, Paul.

1 Like 1 ·
Paul Beskeen avatar image
Paul Beskeen answered ·

Just to note, another issue I found shortly after swapping master and slave was that my (utterly incompetent) installers had not properly torqued down the ET112 positive connection within my consumer unit. Frighteningly there were signs of scorching on the cable. This may have been a contributing factor to the issue. Worth checking if you experience a similar issue.

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.