question

colj avatar image
colj asked

ESS failover - unexpected behaviour from multiple brief (instantaneous?) power failures?

HI All,

Last week my ESS system was installed (Quattro, Cerbo GX and batteries).

We often have short/instantaneous grid failures, and within the first week, during a storm, we had three such failures in about an hour in the middle of the night.

The first Grid failure also tripped my main Consumer unit RCD (30MA) which caused other complications (and an alarm which woke us a couple of hours later when a backup battery failed). Note that this historically would also occasionally occur pre ESS (with these sorts of power failures) but i had hoped that the ESS failover would have avoided the problem so not sure whether the ESS failover contributed or not.

I have not examined log files before but have noticed a few strange entries. The ESS was in a scheduled charge window (started at 23.31) so 'active' at the time of the grid failures. There are a couple of instances in the log where for no apparent reason this goes 'not active', the VE Bus state goes into 'passthro' , alarm relays appear to be triggered and some data fields are blank. There are also several minutes of log missing around the third power failure. Note that ethernet connectivity would have been lost after the first grid failure when the CU tripped.

Note:The Quattro is set to 230Vac (UK standard), but my house is the first feed from a substation/transformer and has a fairly steady 245Vac (this makes the failover times in the log obvious from the voltage) but also raises the question as to whether I should set this higher (240Vac?) to avoid such voltage drops to the house supply on fail over?

In the attached log file I have added comments/questions to the first column . Would appreciate knowing if this represents expected behaviour or whether there is a problem that should be reported please?

Screenshot of one instance of some errors:

MTHGridFailure01NovSS1.png


Full log of events: MTHGridFailure01Nov.zip

Colin

ESSoffgridgrid
mthgridfailure01novss1.png (292.9 KiB)
mthgridfailure01nov.zip (73.3 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
wkirby avatar image
wkirby answered ·

I have briefly studied your log.
One thing that seems to coincide with your schedule going inactive is that your grid meter may have lost communication for that period. Column BW "No grid meter alarm" does not say "OK".
When the grid meter comes back the system would into passthrough first for a few seconds before changing to bulk, in that sequence.
There are a great many sensor values for this system to read and they are read in sequence, not all at once, so there will always be some time between readings.
That's my 2p on this, maybe someone else will have better ideas that mine.

The meter going offline is a bit odd. How long is the cable run to it? If it is longer than a few metres, are the terminators employed? Maybe better to employ the terminators anyway to improve the integrity of the RS495 data bus.

The behaviour of your RCD, we discussed this in your other question.

2 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.

colj avatar image colj commented ·
Thanks for reviewing the data. My ESS doesn't have a grid meter! All loads and Solar PV are connected on the AC 1 Out side of the Quattro (well not quite all loads - there is a EV charger connected directly to the grid but this is not yet in use and would only be used during the charging window).
0 Likes 0 ·
colj avatar image colj commented ·
Just realised that although I don't have a grid meter per the standard ESS schematics there is a grid meter installed on my PV output (forgotten about this addition). This is because my PV inverter (SMA) is over 10 years old and doesn't provide any data to the Quattro and this provides the PV info. Although the grid outages were the middle of the night I assume that this would also be 'rebooting' as a result of the blips. Is this at all relevant to your observation please?
0 Likes 0 ·
colj avatar image
colj answered ·

Having studied the logs more, I may have an explanation. Both of the sections with the string unexpected log entries are preceded by about a 3 minute gap in the log data. My thought is that the instability of my grid supply when it tries to restart multiple times over a few seconds caused the Quattro/Cerbo GX to reboot (3 minute loss of data?). As the log entries restart, the system is slowly configuring itself to where it should be and the 'unexpected' entries occur?

Colin

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.

colj avatar image
colj answered ·

Further investigation shows that the strange entries are probably nothing directly to do with the grid instability.

The Cerbo GX is rebooting since the original grid failure tripped an RCD that caused my network become unavailable. Default VRM config options then cause the Cerbo to reboot at hourly intervals until the network returns.

Colin

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.

Related Resources

What is ESS training video

ESS Quick Installation Guide

ESS design and installation manual

Victron Offgrid page

Additional resources still need to be added for this topic