Battery Boil & Water

[Trawlerman]

You have had a lot of input, and sounds like it is mostly academic to you at this point, so don't take my suggestion too seriously. I may have missed it in this long string, but did you by any chance happen to read the voltages across each battery when under a heavy load like running your microwave? Often, this info can be a bit more informative than low current info. I did see that you rotated (like an X) your batteries, and the issue seemed to stay with the same physical position, not the specific battery. This would indicate to me (as I think you concluded) that it is likely somehow a physical property (heat/vibration/?) but the high current test might be interesting to check out the cable connector resistance when under heavy current condition. Just a thought.

[skipgundlach]

With (again) apologies to OP, a smallish update:

I reset the TriMetric to 90% efficiency. Depending on where we are in the cycle, it's either ahead of or behind the Balmar SmartGauge.

The last many (didn't track, but call it 5 or more) days I've started the day with a Honda boost, finished with an equalization cycle. That usually is done well before noon. The Xantrex 2000XC specifies 16V@10A for the allotted one-hour equalization period; mostly it's not that high (most observed was 15.8 or so) and usually 9-10A.

The solar controller I've left on the equalize setting; the best it can manage is 15.8 by design, and it usually continues in the 20A range until the sun angle or occlusion drops it.

Yesterday it was at 100% on both meters before noon; the TM has the ability to go over 100%; I've seen at most 102% with a noted +AH in single digits. There have been times in the past where we've had to motor extensively and my old meter would show mid-double-digit +AH; it will be interesting to see what the new TM has to say about that.

The time still has not arrived where we're aboard just after the sun has lost its input so that we can take an appropriate SGR to see if our cells all match (there was one alarmingly different cell, and all were notably low for the expected condition in the only time I've measured), but we're now in a windy period which means that my two meters roughly agreed at low 90s SOC overnight thanks to the wind amps.

I'm alarmed that I should have to burn dead dinosaurs every day to apparently achieve something approaching full charge, when our worst case amps (well, presumed, as it's a new amp counter, and the efficiency is now so low as to suggest an older bank) overnight-level are in the low-hundred consumed (today, e.g., with the wind helping last night, at 6:40 it was -81AH and 91 SOC on both meters but at 5:30 it was -71AH and 92 and 93 SOC TM/BM respectively; in both cases it was 12.55V observed)

So, depending on when I look at it, the TM is either less or more than the BM, and until I can prove it, both are wrong (or at least my SGR didn't support the reading) about how full the battery/bank is.

Sigh...

[sailorboy1]

Well as the OP I also have an update.

My solar controller failed while i was away and my batteries per my battery monitor were at -435ah on my 440ah rated when new house bank. But bm said i was at 35% charge and the voltage also pretty indicated the same.This i think is the first time these batteries have ever been below50%.

So i max charged for a while with both my honda+charger and the engine pumping in 120+ amps (way more than what Trojan says to use) till i got to 70% charge and stopped as i had plans and wasn't leaving the engine or honda running while not on board. The next morning i ran the honda+charger for about 3 hours till the acceptance amps started going down and then stopped. The bm still said i was -125ah, which is unusual as normally the acceptance doesnt drop till around -50ah.

During the day i noted the battery voltage and based on that it appears the batteries were way more charged than what the minus ah suggested. This morning the batteries according the bm are 73% charged and -175ah and is reading 12.31v. That voltage is what it normally reads in the morning at 85% charge under load, so i believe the batteries are more charged than the battery monitor displays.

Now I do have my charge efficiency set at 90%, which is lower than Trojan spec of 92%. But that wouldn't make as much difference as being shown i think as i think the bm is currently showing about 10%+ state of charge error and 50-75ah error.

I have 2 take aways worth noting:
- don't get too fixated on your various “magic” battery monitoring boxes. Know the “normal” readings of your battery voltage and acceptance amps at various usages.
- my 39 month old 440ah bank of T105s appear to still be near new capacity. I took 435ah out of them and ended up at 35% charge. Bet if i do the math based on the discharge rate and factor that works out pretty close.

PS - i am not requesting any help

[mdenize]

Originally Posted by sailorboy1

Well as the OP I also have an update.

My solar controller failed while i was away and my batteries per my battery monitor were at -435ah on my 440ah rated when new house bank. But bm said i was at 35% charge and the voltage also pretty indicated the same.This i think is the first time these batteries have ever been below50%.

So i max charged for a while with both my honda+charger and the engine pumping in 120+ amps (way more than what Trojan says to use) till i got to 70% charge and stopped as i had plans and wasn't leaving the engine or honda running while not on board. The next morning i ran the honda+charger for about 3 hours till the acceptance amps started going down and then stopped. The bm still said i was -125ah, which is unusual as normally the acceptance doesnt drop till around -50ah.

During the day i noted the battery voltage and based on that it appears the batteries were way more charged than what the minus ah suggested. This morning the batteries according the bm are 73% charged and -175ah and is reading 12.31v. That voltage is what it normally reads in the morning at 85% charge under load, so i believe the batteries are more charged than the battery monitor displays.

Now I do have my charge efficiency set at 90%, which is lower than Trojan spec of 92%. But that wouldn't make as much difference as being shown i think as i think the bm is currently showing about 10%+ state of charge error and 50-75ah error.

I have 2 take aways worth noting:
- don't get too fixated on your various “magic” battery monitoring boxes. Know the “normal” readings of your battery voltage and acceptance amps at various usages.
- my 39 month old 440ah bank of T105s appear to still be near new capacity. I took 435ah out of them and ended up at 35% charge. Bet if i do the math based on the discharge rate and factor that works out pretty close.

PS - i am not requesting any help

Hi Sailor boy

I know you are not asking for help but just in case anyone else is reading this and wondering about your assumptions....

Without knowing a bit more I would not be prepared to look at a battery and make your take aways...why not?

Well for one thing battery voltage is only really an indication of charge when it is open circuit and has been rested for some time. Any other situation like the 12.31v "under load" is irrelevant unless you know exactly the charge/discharge current is and the voltage curve for T105s at that charge/discharge rate.

And for this reason it would be also premature to know how close to its new capacity your battery is without knowing what the discharge rate was when it went down to -435Ah (although this does seem good) was (because a very low discharge rate would achieve more than the rated 500Ah for two paralleled T105 strings at C100) and the Peukert's exponent set in your Battery monitor would if set accurately take this into account better than any transient voltage

However if this exponent was wrong 35 % would just be guess too,

With my own Victron BM I usually trust its assessment of charge much more than the battery voltage unless the charge discharge current sits exactly on one of the C5,10,20,100 curves for long enough to make a guesstimate of where the state of charge sits on that curve. This is simply because the BM is watching the current drawn all day and constantly makes Time to go adjustments based on the Puekert exponent, I am not patient enough to do the same,

[sailorboy1]

Thanks, I knew all that and am pretty sure of my assumptions

[ramblinrod]

Originally Posted by sailorboy1

Well as the OP I also have an update.

My solar controller failed while i was away and my batteries per my battery monitor were at -435ah on my 440ah rated when new house bank. But bm said i was at 35% charge and the voltage also pretty indicated the same.This i think is the first time these batteries have ever been below50%.

So i max charged for a while with both my honda+charger and the engine pumping in 120+ amps (way more than what Trojan says to use) till i got to 70% charge and stopped as i had plans and wasn't leaving the engine or honda running while not on board. The next morning i ran the honda+charger for about 3 hours till the acceptance amps started going down and then stopped. The bm still said i was -125ah, which is unusual as normally the acceptance doesnt drop till around -50ah.

During the day i noted the battery voltage and based on that it appears the batteries were way more charged than what the minus ah suggested. This morning the batteries according the bm are 73% charged and -175ah and is reading 12.31v. That voltage is what it normally reads in the morning at 85% charge under load, so i believe the batteries are more charged than the battery monitor displays.

Now I do have my charge efficiency set at 90%, which is lower than Trojan spec of 92%. But that wouldn't make as much difference as being shown i think as i think the bm is currently showing about 10%+ state of charge error and 50-75ah error.

I have 2 take aways worth noting:
- don't get too fixated on your various “magic” battery monitoring boxes. Know the “normal” readings of your battery voltage and acceptance amps at various usages.
- my 39 month old 440ah bank of T105s appear to still be near new capacity. I took 435ah out of them and ended up at 35% charge. Bet if i do the math based on the discharge rate and factor that works out pretty close.

PS - i am not requesting any help

What was the make, model, and age of the solar charge controller, and power rating of your solar array? (Not trying to help you, but perhaps others can learn from your failure.)

[sailorboy1]

Outback FlexMax 60, almost 8 years old, ran 7 years with 290w of panels and 1 with 640w. The installation was set for 600w at start far as everything except the panels.

I got what i feel is my moneys worth out of it.

[ramblinrod]

Originally Posted by sailorboy1

Outback FlexMax 60, almost 8 years old, ran 7 years with 290w of panels and 1 with 640w. The installation was set for 600w at start far as everything except the panels.

I got what i feel is my moneys worth out of it.

OK, so the max current seen was likely around 40 A (on sunny days) for a 60 A controller. It would have been nice to get 10 years but 8+/- is reasonable.

What are you planning to replace it with?

[sailorboy1]

I started a thread for suggestions to consider and looked into those. Overall based on controller features, prices, and physical size i currently think i will replace with another MaxFlex 60. It has nice features that i chose it for at start, i know how to operate it, the price is reasonable and less than I think I paid 8 years ago, and fits right into the current location.

[mdenize]

Originally Posted by sailorboy1

Thanks, I knew all that and am pretty sure of my assumptions

Wonderful Yeah I understand your reluctance to go into detail... you just open yourself up to all those who want to tell you how wrong you are.

However if you are brave enough to share your actual measurements that allowed to to be sure of your assumptions I am sure there would be plenty of people who could learn from it.

I do promise not to be negative about your assumptions.

Best Wishes

Mark

[sailorboy1]

No thanks I am trainable

[Jim Cate]

Originally Posted by sailorboy1

No thanks I am trainable

I do hope that you do sign up for some lessons, then. Perhaps there is hope!!

[sailorboy1]

Originally Posted by Jim Cate

I do hope that you do sign up for some lessons, then. Perhaps there is hope!!

I doubt you mean that

[skipgundlach]

Originally Posted by sailorboy1

I started a thread for suggestions to consider and looked into those. Overall based on controller features, prices, and physical size i currently think i will replace with another MaxFlex 60. It has nice features that i chose it for at start, i know how to operate it, the price is reasonable and less than I think I paid 8 years ago, and fits right into the current location.

I have installed a second controller (identical), having had one in reserve for when the first one died (after about 10 years of continuous use), but that, too, eventually will die.

Would you be able to share the url on the thread so those of us interested could see it?

Thanks.

[sailorboy1]

Its only like 9 posts. Click my user name and search threads started by me. Pasting is beyond my skill set on my tablet

But seriously Skip, maybe you shouldn't touch electrical stuff