question

Tom Ranson avatar image
Tom Ranson asked

Cerbo GX causing MPPT's to cut out periodically

Hi all,


In summary, I have a setup as follows. I appreciate this may not be all of the information you require to assist with my problem, so please ask further specific questions and I will gladly provide further info.


- 8x LEOCH DT126 6V traction batteries: 2x parallel strings of 4x 6V batteries - 24V, 480 Ah total.

- 2x Victron 100/30 MPPT, each with 2x 460W panels in parallel with a Voc of c. 49 V (so well within the capabilities of the MPPT). 4x panels in total.

- 1x Victron SmartShunt 500A with mid-point voltage monitoring enabled.

- 1x Victron Cerbo GX.

Everything is VE.Smart networking enabled and communicating.


I am experiencing an issue with the SmartShunt "cutting out" (audibly "tripping") every 20 or so seconds, during periods of high sun intensity when both MPPT's are enabled. When it "trips", charging stops and the mid-point alarm activates (in excess of 2.00%), with values of approximately 15.00%.


If I disable one of the two MPPT's (either of them) at a time, then the system "works" and the SmartShunt does not "trip".


This suggests to me that the combined power output of the two MPPT's may be the cause of the problem, however I genuinely can't see why this might be the case. We are only talking a maximum power output from the two MPPT's of 60A peak.


What are the possible causes of a SmartShunt "tripping"? I have researched this endlessly and cannot find any answer as to under which conditions this should happen, presumably as a protection mechanism. Is it tripping *because* of the mid-point deviation alarm, or is the mid-point deviation alarm a *result* of whatever is causing it to trip? My belief is that it is the latter.


I look forward to your thoughts and advice. Many thanks, Tom

MPPT ControllersSmartShunt
2 |3000

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

4 Answers
JohnC avatar image
JohnC answered ·

@Tom Ranson

If you have the bluetooth neworking enabled, then disable it. Presumably you have everything wired to the Cerbo, so it's not recommended. Also, look up DVCC in the Cerbo manual for alternate functionality.

8 comments
2 |3000

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

Tom Ranson avatar image Tom Ranson commented ·
Thanks @JohnC, I wasn't aware that the Bluetooth networking should be disabled if using a Cerbo GX with wired VE.Direct. Could you please point me to the documentation on this?
0 Likes 0 ·
nickdb avatar image nickdb ♦♦ Tom Ranson commented ·
Smart networking is not enabled by default, it has a specific use case which this is not.

Operating both is not recommended, it will serve no purpose other than to confuse things.

0 Likes 0 ·
Tom Ranson avatar image Tom Ranson nickdb ♦♦ commented ·
That makes sense. Apologies, this appears to have been an oversight on my part. When my Victron system started it lacked a Cerbo GX (and MultiPlus ii), hence I was making use of VE.Smart in the early days. I still use Victron Connect for some things.


Is this stipulation documented in the manuals anywhere?

0 Likes 0 ·
JohnC avatar image JohnC ♦ Tom Ranson commented ·

There might be better links, but this.. https://www.victronenergy.com/media/pg/VE.Smart_Networking/en/limitations.html

The only reason I can think of where bt might have an edge is when syncing multiple mppt charge states. And that may be why you only see this issue when you run 2 together. The syncing thing can probably be done anyway with the mppt Tail function over DVCC.

0 Likes 0 ·
Tom Ranson avatar image Tom Ranson JohnC ♦ commented ·
Yes I see where you are coming from.


It seems like the problem I've experienced shouldn't really have happened. Surely the software should recognise such a situation and either warn the user, or better, prevent the odd situation from occuring here.


I didn't have DVCC enabled, so if I understand correctly, the Cerbo GX was only operating in passive mode anyway? So why would the MPPT's be reacting the way they did?

0 Likes 0 ·
nickdb avatar image nickdb ♦♦ Tom Ranson commented ·

These gremlins are tough to triage over the internet, it's always best to go back to the beginning and configure all the basics to best practice.

DVCC is an integral part.

With everything connected but settings missing and an alternate form of comms active, it is just asking for trouble.

These systems are more complex and nuanced than people realise and need to be installed with that in mind.


0 Likes 0 ·
Tom Ranson avatar image Tom Ranson nickdb ♦♦ commented ·
Indeed, what you say makes perfect sense. Every day is a school day. I now understand more about how VE.Smart and DVCC operate and why a mixed scenario like this may have caused this odd behaviour to manifest. Perhaps this isn't a scenario which has ever been tested for and observed?
0 Likes 0 ·
Tom Ranson avatar image Tom Ranson commented ·
I don't want to speak too soon, but having removed all of the devices from the VE.Smart network and rebooted everything, the strange behaviour appears to have ceased... I wasn't aware that it wasn't possible/recommended to have VE.Smart and the Cerbo GX with VE.Direct simultaneously. Can anyone furnish me with some more info on this please?
0 Likes 0 ·
Trevor Bird avatar image
Trevor Bird answered ·

hello @Tom Ranson ..The Smartshunt has no capacity to "trip" or "cutout" as it is essentailly just a very accurate piece of brass without any circuit breaker in its makeup. If the current to the batteries is stopping completely is cannot be the Smartshunt causing it. Maybe a connection may be loose failing under high load?

2 |3000

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

Tom Ranson avatar image
Tom Ranson answered ·

Hi @Trevor Bird, many thanks for your reply. OK, that's a great start in helping me understand what might be going on here. I was under the impression that the SmartShunt could break the connection in certain circumstances (however the fact that it can't would explain why I couldn't find any information on this).

The pattern of the "trips" is too regular to be caused by a loose connection. I can demonstrate this via a graph from Victron Connect which shows the battery voltage drop to zero, recover some seconds later, drop again, recover, etc. etc. and the cycle repeats. The screenshot shows only one event, however they repeat fairly frequently.

I have determined that the "trip" noise that I am hearing is actually coming from one/or both of the MPPT's (they are situated close to one another so it's hard to tell them apart). It sounds like a relay triggering at the start of the "event" and then triggering again at the end of the "event". When the event starts, the Smart Shunt shows a charging current of 0A and the voltage spikes to just over 29V. When the event ends, the charging current recovers and the voltage drops back to expected levels for Bulk charge.

I'm sorry that I am doing a bad job of explaining this!

Do the MPPT's contain relays that can behave in this fashion? Then comes the question of identifying *why*.

I look forward to your thoughts.

To clarify, I am only seeing this behaviour when *both* MPPT's are operating and connected. When either of them are working in isolation, the behaviour is absent. So it doesn't appear that the "fault" is with either of them individually.

screenshot-2022-09-24-12-51-59-80-30b6efbd53acd6f2.jpg


4 comments
2 |3000

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

klim8skeptic avatar image klim8skeptic ♦ commented ·
The charge current dropping to zero whilst the voltage rising, indicates a cabling/connection/fusing fault.


Look a bit closer.

0 Likes 0 ·
Tom Ranson avatar image Tom Ranson klim8skeptic ♦ commented ·
Thank you for the reply.


Upon closer examination, what I'm seeing (or rather, hearing) when this happens is a "clunk" (like a relay) coming from the MPPT. This coincides *exactly* with the start of the charging current dropping to 0, and then when it recovers there is another clunk.

Why would the MPPT be doing this? The pattern repeats periodically, every 20-30 seconds.


I have checked all cabling, terminations and fusing etc.

0 Likes 0 ·
klim8skeptic avatar image klim8skeptic ♦ Tom Ranson commented ·
Interesting. There is no relay in that model mppt.

Do you have a circuit breaker next to it?

0 Likes 0 ·
Tom Ranson avatar image Tom Ranson klim8skeptic ♦ commented ·
Yes, I am using resettable circuit breakers. In this case, I have 30A ones between the MPPT's and the common busbars on + (which then go to the battery bank on + and SmartShunt on -) The charge controllers are currently set to max output of 25A, so as to be safely under the value of the circuit breaker. The breakers are not tripping, and they will not auto-reset, so I don't think they are related to the issue.
0 Likes 0 ·
Tom Ranson avatar image
Tom Ranson answered ·

Ok, I have some further information.


The noises are coming from the Cerbo GX. They coincide exactly with when the MPPT's cut out, and then cut in again.


It looks very much like the "problem" is being caused by the actions of the Cerbo GX. It appears to be instructing the MPPT's to cut out- they are both doing so simultaneously. There is a clear pattern to the behaviour which I can demonstrate in the screenshot below.


How can I better debug the Cerbo GX to understand what condition is occuring which is making it instruct the MPPT's to cut out on this recurringscreenshot-2022-09-25-12-21-18-94-30b6efbd53acd6f2.jpg pattern?


For clarity, the cut outs only happen when both MPPT's are online. If either of them are online alone then the issue does not present.


What could be upsetting the Cerbo GX?




6 comments
2 |3000

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

nickdb avatar image nickdb ♦♦ commented ·
According to your earlier post you only use smart networking. Is this right?

If so, this does not work with a cerbo, it will not manage anything, you need to connect the devices via VE.direct, then you can use DVCC and coordinate charging.

ESS does not support smart networking.

0 Likes 0 ·
Tom Ranson avatar image Tom Ranson nickdb ♦♦ commented ·
Sorry, I failed to mention that the Cerbo GX is connected to the two MPPT's and the SmartShunt via VE.Direct cables. The Cerbo can see all components registered.


So, presumably the Cerbo is instructing the MPPT's to "cut out" in this odd pattern. I think the root of this problem is, why is the Cerbo doing this? I've been trying to see if there is any verbose logging that can be enabled for it, in order to see what is triggering this action, however so far I have drawn a blank on this.

0 Likes 0 ·
nickdb avatar image nickdb ♦♦ Tom Ranson commented ·

That only happens under (supported) BMS control and these aren't lithium batteries.

Can you post your inverter remote config file here perhaps?

All the cerbo does is coordinate, if DVCC is enabled (please check), it will set limits, that's about it.

MPPT's don't "cut out" under normal operation, they will throttle as the battery reaches it target voltage. They will turn off if PV drops beneath the battery +5V threshold.

They track every 10 minutes which appears as an output dip per mppt.

0 Likes 0 ·
Tom Ranson avatar image Tom Ranson nickdb ♦♦ commented ·
DVCC wasn't previously enabled while VE.Smart was configured, so I don't believe the Cerbo should have been coordinating anything, but apparently it was still trying to buy controling the MPPT charger status?


VE.Smart is now disabled and DCCC is enabled.


I will obtain and post the Multiplus ii config here.


Thank you for the other bits of information which are very helpful.

0 Likes 0 ·
klim8skeptic avatar image klim8skeptic ♦ Tom Ranson commented ·
Unplug the mppt's from the Cerbo to test.

Mppts will operate in standalone mode just fine for a test.

0 Likes 0 ·
Tom Ranson avatar image Tom Ranson klim8skeptic ♦ commented ·
Indeed, they operate just fine in standalone. But the strange thing is, that the Cerbo was configured with DVCC DISABLED, so it should have only been acting as a data collector and not trying to actively control the MPPT's? It was clearly telling the MPPT's to stop charging on that very predictable pattern.


Since disabling VE.Smart things seem to have settled down and are behaving as expected so far. DVCC is now enabled, too.

0 Likes 0 ·