Are my flooded batteries ok?

[gooseberry]

Hi all. I recently decided to build myself a little "raap en skraap" solar setup using parts I could find that was lying around and second hand. My father had 2 250W 60 cell panels and 4 Trojan T105 batteries from a previous install laying about in the garage for a couple of years. I picked up a used 3kVA Axpert MKS Plus 24V, a BMV 702, an ICC pi and bought another 60 cell (280W) panel that I connected in series with the others. At the moment I'm only running a single Bosch standing freezer on the inverter. 

I want to transfer the system to my home for use during load shedding and longer power failures, which seems to be a regular occurrence now, but I'm not sure if I can rely on the batteries. Can anyone have a look at these charge/discharge curves and tell me if it's what you'd expect to see? I've read about equalisation charges on these batteries but don't have a charger to do it myself. I also don't think the BMV parameters are necessarily correct, judging by how quickly the battery displays FULL charge in the morning, after which it still draws a lot of amps from the panels. The previous owner also had 4 Trojans, so I didn't change any settings. It would be good to measure the capacity accurately to get a better SOC measurement, I can't believe that it really stays above 90 throughout the night, when the voltage sometimes drops below 24V.

I also don't see any evidence of the premature float bug that my inverter is supposed to have? If anything I'm a bit worried that it's overcharging by staying at the bulk charging voltage too long?

[Coulomb]

9 hours ago, gooseberry said:

Can anyone have a look at these charge/discharge curves and tell me if it's what you'd expect to see?

It looks OK so far, but you haven't given it enough load to tell if the battery will hold up.

Quote

I've read about equalisation charges on these batteries but don't have a charger to do it myself.

My understanding is that the Trojans won't last as long without that equalisation charge once a month or so. But they'll still do OK.

Quote

I also don't think the BMV parameters are necessarily correct, judging by how quickly the battery displays FULL charge in the morning, after which it still draws a lot of amps from the panels.

If ICC is getting the SOC from the BMV, then the BMV is definitely not set up well. You need to tell it the capacity of the battery, and as you say, that's unknown at this stage. But a guess of half the new nominal capacity (since the battery is old) might be a better starting point that whatever it's set to now. Things like the Peukert effect are also important. Changing the maximum total charge current (setting 02 in the inverter LCD) higher will increase the tail current threshold, which will shorten the absorb time.

Quote

I can't believe that it really stays above 90 throughout the night, when the voltage sometimes drops below 24V.

That certainly can't be correct.

Quote

I also don't see any evidence of the premature float bug that my inverter is supposed to have?

No, that's a near-perfect charge you captured. But there was only one brief interruption to the PV (?) charging (and that may have just been the freezer starting up, which is less than a second, too short to trigger the premature float bug). The premature float bug is most likely to strike when you have intermittent cloud.

Quote

If anything I'm a bit worried that it's overcharging by staying at the bulk charging voltage too long?

No, you can see the charge current tapering nicely while the battery stays at the absorb voltage; that's what it's supposed to do.

I don't see your maximum charge current setting (max AC charge current is 30 A). If that's very low (it looks like it might be set to 10 A), then the tail current (the battery charge current below which the inverter switches to float) will be very low, which would explain the rather long time in absorb stage.

Edited August 9, 2020 by Coulomb
charge interruption was possibly just freezer starting

[gooseberry]

Thanks Coulomb. I'll connect a load to test the true capacity sometime and look for someone to do an equalisation charge. The system is not connected to AC at all, this was charging from the panels only. It was a lovely sunny day.