Odd Battery outage on one MPPT, not on other

Question

question

David asked Sep 10, '22

Odd Battery outage on one MPPT, not on other

I have 2 SmartSolar 100/20 controllers fitted physically next to each other. The Battery +ve and -ve connections are each coming off adjacent connections on the same busbars that is located around 6" (15cm) away.

The controllers are configured identically, and have the same revision of (the latest) firmware. The only difference is one is a 100/20 and the other is a 100/20 48V (both are on the same 12V system though of course).

The problem:

One of these controllers is dropping the battery voltage every evening at around the same time and re-establishing it around the same time the next morning. There is no physical intervention at the time the voltage goes or returns. The times of drop and return are matching the time the harvesting is finishing and restarting, which surely cannot be a co-incidence?
I am using VRM to see the voltage so this is what the controller is reporting and possibly not what the voltage could be if a DMM was used to read the actual physical terminals.

This is a VRM chart of the two controllers:

On the controller that is loosing voltage, I have the LOAD terminal in use with a constant load of minimal current, but this load does not really vary and I don't think this should be relevant, just adding for completeness. The Load output is going off of course, as there is battery power to keep it alive (Load setting is "always on" ).

I'll probably fit new cables tomorrow as I want this to be working 100% and that can be the only possible physical thing at fault I can imagine outside of the actual controller.

Anyone got any ideas why this is happening? Got to be suspecting the controller to have a weird fault.

MPPT Controllers

screenshot-2022-09-10-at-22-30-59-monty-vrm-portal.png (106.1 KiB)

2 |3000

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

David answered · Sep 11, 2022 at 11:52 AM

Took out the controller to get better access to the fuse (have you ever tried to change a fuse on one of these once it is fitted in place??) and decided to have a general look at it to see if I could see anything obvious ...

I think I have found the culprit! not a great photo (used the phone on my camera) but examining under a magnifying light on my bench, I am pretty sure this looks to me like a dry joint on the BATT +VE terminal.

That line (crack?) on the post goes along all the visible sides and to me looks like a classic dry joint. Thinking back, the ferrule on the cable going into the BATT +VE had showed signs of heating some time back. I replaced the ferrule (maybe the whole cable?) and though not much more of it, but a dry joint could well have caused a localised heat issue.

Now do I send this back to the distributor for testing or do I put an iron on it myself for a couple of seconds?

I suspect with a low cost item like a 100/20, Victron will probably just either swap it out and not get to the bottom of the fault*, or worse - test it briefly and return it back as NFF and I'm back to square one with a suspect controller.

*the last two items I have reported as faulty - a BMV-712 and a Cyrix-ct, they were not checked out, Victron told the distributors to just replace and destroy the originals.

mppt100-20-dryjoint.jpg (108.2 KiB)

1

2 |3000

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

klim8skeptic ♦ commented · Sep 11, 2022 at 12:00 PM

Looks like a dry joint, return to supplier.

1 ·

Modifications Space

Node Red Space

Answer 1 · 2022-09-10T23:03:30Z

klim8skeptic answered · Sep 10, 2022 at 11:03 PM

Given that mppt 277 shows no battery voltage during the night, and also goes to 13.5v during the day would suggest that the battery fuse is kaput.

Can you contact mppt 277 after dark via bluetooth? Does the blue led flash during the dark hours?

Please check the battery fuse, and the fuse in the bottom of the mppt.

2 |3000

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

Answer 2 · 2022-09-11T10:40:51Z

@klim8skeptic , that is a very plausible explanation. I went to the electrics cupboard last night to have a look to see what was going on ... blue light flashing on one MPPT but not this one. Removing the +ve BAT cable and reinserting it "woke up" the unit and it has been alive all night, but could have disturbed the fuse in the process (loose and I pushed it back in without realising it).

I have compared the battery voltages of the two controllers AND the battery as reported by the BMV and that would seem to confirm your theory I think?

Have to ignore the actual voltages as we can't set the axis values with the VRM custom widgets, but the pattern is the impotant thing ....

This is showing when both controllers seem to be mostly normal (although the start and finish are exhibiting variance). The 'good' controller (Red) is tracking the BMV voltage (Black) perfectly and the 'problem' controller (Blue) is as well apart from start and finish. The last part of the chart is from when the problem is starting to get more obvious.

Looking at a time section when the controller is on during the day going off when it gets dark is revealing. You can see on the graphs that the battery voltage dips on a regular basis - this is when the Multiplus kicks on to run the AC fridge and then recovers. On the graph below, the 'problem' controller is not showing that dip at all, which would suggest it is not seeing the battery at all at this point? And at 15:00 at the end, the blue voltage drops suddenly, and that did coincide with drop in general harvesting (heavy cloud coming over probably).

I'm about to go out and waggle things while checking on the apps and with a DVM to compare values to see if I can pinpoint the issue and hopefully it will be a physical thing that can be easily fixed :)

question