Valence XP bad self Discharge?

[Ryantech]

Battery voltage set at 13.7 volts. Within 2 hours of being off the charger battery is down to 13.3 volts. I have 3 12XP batteries. Discharge right from the batteries is 0.31 amps at idle. Went camping 2 weeks ago also and the batteries were only charged up to 13.9 volts due to sunlight limitations and with a 100 amp load, the batteries lasted 5 minutes before low voltage cut off. I dont have time to chase after 10 possible solutions as I'm am home two times a month so any past experiences similar to mine or solid evidence would be greatly appreciated! I'm more so leaning toward a slave bms that is shot or a dead cell.

 

[Ryantech]

Sensor 6 on battery #1 is the only thing that looks like it's broken but that still should not explain rapid voltage decrease. I also have had the thunderstruck plugged in from full to dead so im not sure why 2/3 batteries report under 5% capacity. Not concerned about it but just something im now aware of. Photo as of Thursday 12:00 pm.

 

[Travis]

You can disregard readings from sensor 5 and 6.

 

[Travis]

Any battery sitting by itself without connected to any other battery load or charge and also showing amperage moving needs its shunt calibrated. It's not actually discharging. But it thinks it is so it reduces its state of charge accordingly. It's true state of charge is actually not reducing.

 

[Travis]

If your charger only brought it up to 13.7 then it wasn't full unless it held it there for a while and amperage fell very low. During a quicker charge let's say over 5 amps the voltage will go much higher.

 

[Travis]

You need to charge until the battery shows 100%. It'll stop at 99 for the longest time but it won't actually show 100 until it's actually finished. Sometimes you'll stop at 40% and stay there for a long time then suddenly it'll jump to 100. When it jumps to 100 it resets the state of charge indicator to be actually correct.

 

[Ryantech]

You need to charge until the battery shows 100%. It'll stop at 99 for the longest time but it won't actually show 100 until it's actually finished. Sometimes you'll stop at 40% and stay there for a long time then suddenly it'll jump to 100. When it jumps to 100 it resets the state of charge indicator to be actually

Before I went out camping I had the batteries sit and charge at 13.7 volts for about a week or so. Right before I left I set the controller to charge to 14.2 volts. 13.7 volts was the max durring the trip and a days charge lasted maybe 2 hours before the voltage tanked.

Oddly enough I had time today to dive deeper into this issue. I'm writing this as I'm about to head out the door for work haha. But I had the batteries at 13.9 volts today which I didn't care about. I put a constant 170 amp load on the batteries till they hit 12.2 volts and jumped back up to 12.7 volts. With that load test I drew 267 AH. Not at all full capacity or voltage range but would you happen to know why I could run a 2.5 hour high load on it without issue but a trickle load with me camping just about killed the batteries?

 

[Travis]

When you say 170 amps I assume that's divided by three batteries?

 

[Ryantech]

When you say 170 amps I assume that's divided by three batteries?

That is correct. The amperage stated on each battery were all close to 50 ish amps. No battery was being out "amped". Hopefully when I get home next I can retest from 14.2 volts down to 12.3 volts w/o load. Hopefully I'll be able to delete this post or make an update because I'm baffled. I'm not a novice when it comes to electrical but lithium has always been a struggle for me to fully comprehend.

I also noticed one of the batteries showed 3.xx% SOC at any voltage and was able to see the amperage while discharging. Going off of what you said charging it up while the bms is awake will have that SOC recalibrate? I don't necessarily care about seeing the SOC as I only see it when I hook it up to valence software for diagnosis. Thunderstruck takes care of 99% on what I need to monitor.

 

[Travis]

That is correct. The amperage stated on each battery were all close to 50 ish amps. No battery was being out "amped". Hopefully when I get home next I can retest from 14.2 volts down to 12.3 volts w/o load. Hopefully I'll be able to delete this post or make an update because I'm baffled. I'm not a novice when it comes to electrical but lithium has always been a struggle for me to fully comprehend.

I also noticed one of the batteries showed 3.xx% SOC at any voltage and was able to see the amperage while discharging. Going off of what you said charging it up while the bms is awake will have that SOC recalibrate? I don't necessarily care about seeing the SOC as I only see it when I hook it up to valence software for diagnosis. Thunderstruck takes care of 99% on what I need to monitor.

Yes. But only when balancing is complete and it's also at full.

 

[Travis]

I think it has to be over x amount of volts per cell and under 1 amp for one minute.

 

[Ryantech]

Cool. Thanks. I've been told this is way to expensive of a monitor but I have all the data I need to see on one page. Capacity still isn't 100% accurate but hopefully next capacity test will get it near perfect.

 

[ericgrand]

You need to charge each battery individually up to 14.6V, in order to properly balance each cell and reset the SOC.
I'm in this exercise with some batteries sitting for a few months and which need to be properly balanced before being connected in parallel.
You could also connect all batteries together in parallel and charge up to 14.6, but be careful if one cell is too heavily unbalance in regards of the 3 others internal cells, it may lead to overcharge other cells.
Better to closely watch battery and charge up to 14.6 by monitoring them, increasing voltage step by step.
For one of the battery with a severe internal unbalance, it appears only when 3 of the cells reach over 3.4V, where the last cell was still at around 3.35V.
3 cells were almost charged while the last one was few Ah behind... I add to lower the amp around 1A and decrease top voltage to remain below 3.65V/cell for the 3 charged...Internal balancing kicked in, but I believe it is limited to something around 2 or 300 mA...It tooks hours for the last cell to reach the level of the 3 others...
On one battery, the unbalance was so heavy that I opened the case and charge the individual cell via the balance plug.
These batteries are now performing perfectly well, used in a serie of 7 in an electrical car. I have some peak power at around 270A and all 7 batteries remains perfectly balanced!

 

[Travis]

Yep. What he said.

It's a complicated dance getting things balanced and full, even after doing it few people truly understand it.

 

[Ryantech]

I appreciate the feed back. Would 14.6 be their max or highest comfortable voltage? I guess I'm still confuzzled on why I had very little capacity at 13.9 volts one time. I ran a 100 amp load for 5 minutes and hit 11.9 volts or so. When I took that picture the batteries were at 13.3 volts no load. Now granted that was the very first time I did a "fully charged" to "dead" test. The batteries are in storage pretty much all the time so thats the reason why I have them at a lower voltage. Though I'm now thinking it's not as bad if I have them fully charged and they just stay there vs charging up every day or two because a cell dropped voltage.

 

[Travis]

14.6 is 3.65v per cell. From now on you should only think in terms of voltage per cell. When Holding them at 3.65 one cell is likely to hit that voltage first causing you to freeze the voltage at that point. As the lower cells slowly rise in voltage you could increase the total voltage slightly without pushing the first cell over 3.65.

So even though 3.65 theoretically represents 14.6 it actually doesn't in reality because not every cell will be exactly the same.

The upper voltage limits are a bit higher . Don't quote me on this but I think it starts counting the high voltage timer against you around 3.8v and the BMS won't shut you down until over 4v volts. There is a high voltage timer and there's also a high voltage count. The high voltage timer alone isn't high enough to cause damage or log a overvoltage event. The high voltage timer is merely a way to identify batteries that aren't maintaining balance easily. There's a great deal of batteries that have large high voltage timers having accumulated over the years and the battery as a whole still performs very well.

The high voltage count is a different story. That means the voltage was dangerously high and damage was done. The BMS should shut the system down before this happens so that count should be zero.

 

[Travis]

It's possible that you didn't charge to 100%

Yes sure 13.9 is theoretically full but that's only if you were holding that voltage with a very very slow charge rate.

If your battery is at 13.9 volts but it's still charging then it's not full. What I mean by charging is that if it's still accepting amperage at that voltage then it's not full.

 

[Ryantech]

It's possible that you didn't charge to 100%

Yes sure 13.9 is theoretically full but that's only if you were holding that voltage with a very very slow charge rate.

If your battery is at 13.9 volts but it's still charging then it's not full. What I mean by charging is that if it's still accepting amperage at that voltage then it's not full.

I guess that makes sense. I'm used to the FLA setting the charge controller to the profile and it always being full. Seeing 13.9v after being in storage mode for about 3 weeks I assumed they were charged.

In a nut shell I can charge to 13.9v 3.475v per cell and it's not at full capacity till the amperage starts to dip below an amp per bank? Like it would show 3.475v on avg cell but they'll still accept the charge and not raise the voltage? Am I on the right thinking path?

 

[Travis]

If They were at 13.9 after sitting for a long time (days) then yes they were full.

 

[Travis]

Yes it should be below 1 amp per battery at that voltage. That's when it's pretty close to full except if you have anything out of balance you'll have to hold it there for a amount of time possibly many weeks for the low cell Bank to catch up. If the state of charge shows 100% then balancing is finished and of course it's also full.