SBMS - Solar BMS from electrodacus

[Mrcoolabah1au]

All ways a change one in a bad batch

[T1 Terry]

Could it be faulty?

That iswhy I'm leaving any further exploration into the problem until we get home. we have a second unit there so if I can't get this one to operate as required I'll swap it out for the other unit and see if the problem occurs with that one as well. Then at least I'll have something to report back to Electrodacus, it is only a problem in one unit or it is a problem common to both units. From there I'm sure he will be able to solve the problem, he is a very cluey bloke and the base platform is great so it should be something that can be sorted and if it is an inherent problem, fixed in the software and sorted by linking over the interweb.

T1 Terry

[bagmaker]

Any luck with it Terry?
I am considering one of these to boost the harvest, very unimpressed with a parrellel panel setup running rainbow style

[T1 Terry]

Any luck with it Terry?
I am considering one of these to boost the harvest, very unimpressed with a parrellel panel setup running rainbow style

Ummm.... "Please explain?" The Electrodacus is also a PWM controller, so they would still be connected in parallel. The internal balancing part looks as though it is doing heaps, but there is no real cell balancing occurring that makes any noticeable difference. The fact that the unit decides a cell is faulty and simply turns the charging off without putting up any sort of a warning flag is not good. It can be wired to shut down everything, but there is no way to get the system back up and running to clear the problem.
Each time I was forced to disconnect the solar feeds and direct connect them to the battery to clear the faults by bringing the battery back to 100% SOC. Once that was done, all was fine, till the next time.
Because it was not effectively balancing the cells at the top end, one cell would run away and stop the charging, the result was the lowest cell would show a low voltage when the loads were applied in the morning. This would cut the charging and let that cell get absolutely hammered.
If I hadn't had the Junsi cell logger connected on a separate set of cables, I would have believed the Electrodacus that the cell was faulty and extremely low voltage. The Junsi showed this was not the case until the loads pulled that cell way down. By bypassing the Electrodacus the system functioned fine, I just had to apply manual control of the charging by either switching some of the solar off or applying additional loads, or a combination of each.

T1 Terry

[bagmaker]

Oh, thats not good. I didnt realise they were PWM as well. Oh well mr Dacius......

Hmmm, please explain?-

well, The more I think about the mods made, the more I think there is an error somewhere in my installation
Unfortunately been layed up in bed for a few days so little progress but, on the up side, lots of thinking time.
I have finally switched from 2x MPPT controllers running from 5 x 150w panels to PWM rainbow style, 20a Dingo with 2x 100a relays.
On a decent day 600w input was common enough, victron 700 measured, running MPPT.
The panels are not all exactly the same but rated within a volt or so, 17.5v /18.5v etc fully loaded. Each has an extra diode fitted.
So went up to 10 panels at the same time (and a further 6 coming, to bring the roof to a 2.4kw "rated" maximum space available) as the dingo mod. But harvest has gone down to around 500w at best.
Its my thinking there is a choke point somewhere, I simply cannot squeeze 150amps at 5v difference though wiring thinner than a thumb.
Perhaps its the neg wire, I should take it to the top of the shunt instead of an earth terminal nearby.
The increase in panels was a bit of a sign, virtually no increase in harvest although panels added afterwards.
Of course, the actual battery is also at its life end, probably not helping.
I shall keep updating as I find things out

[T1 Terry]

With the controller still in full boost mode, sun shining on the panels and no shadows made by you while doing the measurements , record the voltage at the terminal of each solar panel. If the wiring is up to scratch the voltage should be a max of 1v above the battery voltage. That is the actual voltage at the battery terminals, not the voltage displayed on any charging device. Is the voltage on the Dingo 20/20N the same as the battery voltage measured at the terminals? If it isn't, there is your first problem, trace through the wiring to find out why ..... The Dingo should be getting its voltage reading directly from the battery, not from the general wiring. The solar output from the solid state relays should have a clear path to the battery with at least the same size 6 B&S cable from each relay to a fuse, then the battery.

T1 Terry

[bagmaker]

There is 20.5v (roughly) at the relays, this seems to be doing all the right things when external (240v) charge is applied as test -ie, fluctuating happily from 20.5v when floating (batt at 13.8v fully charged) and down to the battery voltage when in boost mode. These are the same voltages as I have asked of the Dingo. The output has been the same over a few fully sunny days, panels are squeeky clean
The panels are parellel wired into groups of 3 with shottke blocker diodes in each. These triple legs come back to a junction box with 8g? (not sure, but its damn heavy, about 6mm dia core) wire then feed a busbar, each leg has a 45a fuse on the busbar. Then the busbar feeds a main drop of cable (4m) of 1B&S cable to a 500a electric isolator, the output of which is split to provide the relays using equal length cable.

From the relays the cable is of a reduced thickness -original fitment- maybe5mm diameter core x 2. This could be an issue

The negative sides of the system is the same, without the fuses, and terminates on a body post next to the relays. There seems to be some small fluctuating resistance measurements between this post and the top of the BMV shunt. This could be an issue

I will check if I have an extra volt at the busbar compared to the battery - I dont think that I do - this could be an issue.

I dont quite understand though, Are you saying the infeed should retain a volt or so above the battery -to what point? OK so its +1 at each panel, then +1 at the joiner or busbar, -at what point would it be the same as the battery voltage? Right on the battery, at the relay (in boost)? Could you elaborate?

The dingo is getting a separate sense wire from the battery, it matches the other readouts.

Your thoughts are much appreciated

[Peter and Sandra]

I had similar "problems" when I fitted additional solar panels, only a marginal increase in output. After a few questions and answers from T1 started at the panels and worked through from there all the way back to the battery.
Ended up ditching all the MC4 connectors where they connected to the roof wiring, all now connected to ring terminals on busbars in junction boxes on the roof, some MC4's had melted and others were warm when in use.
Next was roof wiring was marginal and warm to the touch when full sun available so that was upgraded.
The biggest cause of voltage drop was the fuse connections, posts even though tight were too hot to touch due to poor surface area contact at the bottom but fuses them selves were cool to touch. Problem solved by stainless washers sized to fit bottom of posts and provide full contact to wiring ring terminals.
As Terry said regulator voltage sensing MUST be at the battery terminals as even a .1v loss can affect regulator performance, all those little volts add up!

[bagmaker]

I had similar "problems" when I fitted additional solar panels, only a marginal increase in output. After a few questions and answers from T1 started at the panels and worked through from there all the way back to the battery.
Ended up ditching all the MC4 connectors where they connected to the roof wiring, all now connected to ring terminals on busbars in junction boxes on the roof, some MC4's had melted and others were warm when in use.
Next was roof wiring was marginal and warm to the touch when full sun available so that was upgraded.
The biggest cause of voltage drop was the fuse connections, posts even though tight were too hot to touch due to poor surface area contact at the bottom but fuses them selves were cool to touch. Problem solved by stainless washers sized to fit bottom of posts and provide full contact to wiring ring terminals.
As Terry said regulator voltage sensing MUST be at the battery terminals as even a .1v loss can affect regulator performance, all those little volts add up!

I believe these might be my problems too, thanks Peter.
Be wary of stainless though, copper is a much better choice IMHO, the bolt will be doing all the electrical work with a stainless washer

[BruceS]

And yet every house install of solar panels just use the MC4 connectors?
I had them on Kreek Kroozer & now on this boat without any problems?