question

This doesn't fly as an explanation frankly. I have never seen a vehicle alternator, especially after voltage drop from the crossover cable, elevate voltage anywhere near typical absorption values. Most direct alternator links provide about 14v max, and hence usually stop charging house batteries well before absorption completes ( hence why so many DIY campervans sulfate themselves inside a year).

45 minutes still seems like too little abosrob time. Considering the recent reports of controllers overreading their output and floating early, I suspect this unit is doing the same. @lnanne the definitive answer is whether your battery monitor says you've replaced the Ah you used the night before, and whether it says still gobbling amps once on float. If the answer is "No" and "yes" in that order, the Victron isn't working.

What kind of battery are you using? 14,8V sounds like a quite well charged battery to me...

@lnanne

It's a bit tricky to answer your questions excelty, not knowing the full details of your rig, OR the full details of exactly how the magic Victron algorithm works. But I'll try:

1/ a) It IS correct to observe amps, but on the TRIMETRIC. That shows the net amps the battery is losing/gaining. If it's holding 14.8v and still accepting 18amps, the banks (assuming it's around the 3-500Ah size?) is NOT full. Even if there's a load reducing what extra could potentially go to the batteries, if that batteries are capable of absorbing anything over 1% of their rated Ah capacity, they aren't full. In fact, you should REALLY be using the replaced percent (rPC) funtion on your Trimetric, and determining 'full' based on replacing 105-110% of your previous discharge depending on batt type).

@lnanne

1. b) I dont buy that its' because of the alternator proving so much absorption current that the MPPT dropped under 2A. Vehicle alternators are FAR lower than 14.8v (14.0v, usually) , meaning they simply cannot hold house batteries at 14.8v for absorb. ONLY the MPPT can do that - once that battery V is ove the alternator V (14 ish), th MPPT does ALL the charging. Ergo, it's impossible for your alternator to have provided so many amps during Absorb, that your MPPT only had to provide <2A to keep the voltage at 14.8v, ad just tripped to float.

It's clear that in this case, charging from your alternator SOMEHOW has disrupted your Victrons magic algorithm calculation prior to absorb starting. Based on the other reports of short abosrb times, it seems vulnerable to this already. Just why it does this, I dont know.

@lnanne

2. Again - I really dont know. As above, I dont think the 2A cutoff can be responsible for this. More that I think you early alternator charge has somehow fussed with the % of Max Charge formula and incorrectly shortened it.

@lnanne what time did you start driving and was it cloudy?

They said already startup voltage was 11.9v.

Yep. Something definitely off. This looks like the grpahs @Solvan sent to me, particularly the fact that 52% of the power provided alone was in float. Even with some loads still running, that says the batteries were well and truly still hungry when the MPPT stopped absorb. Considering you started the alternator well after first sun, that 11.9 startup Voltage should have given a FAR longer absorption. I think you have a faulty unit :(

Also the reason you have to set the MPPT's voltage so high (14.95) is because it reads V at its output terminals, which in turn connect the wires it pushes the charging current through - hence it's victim to voltage drop induced by the larger current. You need 14.95 at the controller end of the wire, to get 14.8 at the battery due to resistance losses.

The Trimetric uses separate sense wires and thus isn't victim to this. It measures true battery voltage and is the more accurate reading.

To be accurate you will need to determine the voltage just BEFORE the MPPT 'wakes up' @ V batt +5v - once the MPPT 'wakes up' & the bulk phase kicks in it will immediately pump up the battery voltage somewhat. So what i'm saying is that evaluation of the voltage/photos even just after 'wake up' is not even enough to accurately evaluate whats going on. To be thorough I would recommend taking a short video starting from before the MPPT 'wakes up' & coupling it with the actual behavior later on in the same day.

PS. You don't need to prove anything to us, my recommendations are more for yourself to confirm without any doubt. If you are already 100% sure then just return the unit & get it tested - that said clear evidence would still assist with this.

@lnanne

May I ask where you're located?

Hi @Boekel, I just had a look at the info provided & totally agree with you.

It was stated that the battery bank is 12v 420Ah = ~5.0kWh

~11.9v rest voltage for a lead acid is ~10% SOC.

Accordingly the battery bank has been depleted of ~4.5kWh from its total capacity.

Assuming ~95% charge efficiency it would require ~4.8kWh of charge to bring the SOC back to 100%.

Now even if we exclude the absorption phase & just focus on the bulk phase (controlled by voltage only), this phase alone should bring the battries back upto ~80% SOC. According this phase ALONE should require ~3.8kWh of energy to complete.

Now when we look at the Victron Connect data log, the TOTAL yeild for the entire day shown is only 1.4kWh, with the bulk phase only lasting ~4.5h & the battery in float for most of the day.

So in summary, I think the batteries are properly screwed!

The other thing that I also noticed is that the MAX PV voltage recorded is only ~20 to 22V. So with a 12v system it would take a fair amount of sun & time after sunrise each morning to achieve the required 'wake up' voltage of V batt +5v (~17v min to ~19v if the battery is not depleted). If you can rewire the solar panels into a series configuration it would make a significant improvement to this aspect.

Finally, regularly depleting lead acid batteries to ~10% SOC is doomed to fail VERY quickly (even with new batteries). You should design the system for ~50% DOC as the extreme max and ideally only take out ~20% max per day to provide a few days of autonomy.

I disagree with @Mark and @Boekel here.

See below OCV chart for Fullriver DC400s - 11.9v = 30-35%, not 10%. And thats at 25C; I regularly see 11.9-12.0v at 40% in cooler weather. Yes, they're AGM, but point is SoC and OCV varies between different batts.



I admit, even if 11.9v = 30% SoC, @lnanne 's graphs still don't account for 70% recharge. BUT:

That 11.9 was a historical min voltage ie likely UNDER LOAD (as you actually already pointed out Mark...). 11.9v is NOT the resting OCV in the morning. What lnanne ACTUALLY said was 'generally a morning voltage of 12-12.2', and even then thats with a 5a load. Meaning the batteries are very likely far from 30% at all, rather much higher.

Additionally, they said that they added more charge by driving, which wouldn't appear on that Victron chart.

In other words, I very much doubt the batteries are "properly screwed". In fact the data indicates quite the opposite

Furthermore @Boekel and @Mark:

I don't see how this changes the problem with the MPPT. Specifically:

-If the batteries ARE 420AH, and lnanne really IS depleting them to 30% SoC (unlikely) then why is the controller not holding bulk and absorb long enough to fill this?

-If the batteries are LESS than 420 AH (Ie sulphated) then they're basically just smaller batteries, and it shouldn't make any difference. The MPPT should still hold absorb until current dwindles, and it's clearly not doing that.

For reasons explained above, I think scapegoating the batteries is a dead end, considering we have established the controller IS dropping out of absorb too early .


But for the sake of rigor: @lnanne - next time you do a discharge, would you mind checking the Trimetric for this info:
- % state of charg
- Voltage (after no load for 10 mins)
-The actual Ah used (hold 'select' 3s and enter sub menu)
-Model of battery

That should verify battery health somewhat.

Regarding the relevance to the question - the battery should reach ~80% SOC during the bulk phase alone & the bulk phase end condition is simply based on voltage only. Accordingly for a deep discharge most of the energy entering the battery will occur during the bulk phase.

I was attempting to demonstrate that even if you forget about/exclude any possible concerns with absorption phase, based on the LIMITED data provided (from the MPPT only - rather than a proper BMV) there seems to not be enough energy entering the battery during the bulk phase. This is typical behavior from a battery bank that is not performing at full capacity any more.

It was also just some general thoughts/advice on the current system design vs usage profile.

They do have a BMV. Or, rather a better one - a Trimetric.


In fact lnanne posted their average nightly useage figures above - about 140Ah.

Considering they said the batteries are already a bit detracted from previous undercharging, and they're operating in near freezing conditions, hitting the 12.0 - 12.2 mark nightly as they say makes sense to me for a roughly 50% cycle.