11 replies to this topic

[bajaphile]

Hey All,

Just got back from a trip to the Sierras for 4 days. This was our 3rd trip on our new SOK Lithium 100 amp-hr battery. On our final day, I woke up to absolutely no electricity. The SOK BMS app gave me the following screen shot. I had gone to bed last night with it at 68% and powering just the CO/Propane monitor as well as a 60watt electric fridge, pulling about 5 amps. 

 

 0percent.jpg   71.74K   49 downloads

 

I started the truck and the Victron DC-DC charger showed bulk charging at 14.5V @  20-25 amps. The BMS also showed a +20-25amp charge. 

 

Drove 7 hours home, and to my surprise, the battery is now only at 38%. Something must be off... 20 amps @ 7 hours = 140 amp hours....I have 4 ga wires running from the alternator/starter battery in the truck back to the camper. 

 

While we were camped I had my 200w solar panels out, getting about 13.5V @ 10amps. I have the settings on the solar charger to LiFePO4, which I thought would up the voltage to ~14.0V... Still it was enough to keep things relatively topped off. But, to my surprise again, the BMS never did get over 80% charged. Volt meter in the camper was always above 13V. 

 

Any ideas? Any troubleshooting trips? This is my first Lithium battery and before this was very impressed with it. Thank you for any help. 

 

Here is the current reading from the BMS. All cells appear to have voltages that are similar. 

 37percent.jpg   73.38K   43 downloads

Edited by bajaphile, 11 September 2023 - 11:47 PM.

[Jon R]

Your screen shots show a “calculated capacity” of 278 Ah, but you say your battery is a 100 Ah battery. The cell voltages in your second shot look like reasonable voltage for a full charged cell after charging ceases and the cell relaxes. I suspect your BMS is calculating the charge state based on 278 Ah, in which case the 37% up from zero for an actual capacity of 100 Ah would make sense from a current integration standpoint. Is the capacity a setting in the app for your BMS? If so, change it to 100Ah. Fully charge your battery, and resync your BMS monitor to 100%.

There’s not enough load info to understand why your battery might have gone to BMS low voltage shutoff, but it looks like you have your MPPT charging voltage set too low to fully charge your battery. You want to have your bulk and absorption voltages set at 14.2 to 14.4 at least as initial settings until you learn more. Set float voltage to 13.4. Same for your dc to dc charger.

Edited by Jon R, 12 September 2023 - 02:42 AM.

[bajaphile]

Thanks Jon for your reply. 

 

• Calculated capacity @ 278Ah - I hadn't even noticed that. I was browsing through the SOK manual but there doesn't appear to be a setting to reset it. I might try a factory reset of the BMS. Nice catch there... 
• As far as charging goes, the Victron DC-DC charger is set at 14.3V and was seemly working well for the last few months.
• My PWM Renogy solar charger is connected directly to SOK battery while camping and I only have the option to set it to the pre-programmed LiFePO4 charge profile. I am about to look up the manual and see what that exactly is. I was surprised to see it in the low 13.XXV though even in the afternoon sun. 

[Jon R]

Assuming the BMS is using the wrong total battery capacity of 278 Ah, if your BMS was synced to 100% at full charge at the beginning of your trip, then the 68% of a 278 Ah battery it had calculated the evening before the battery went to shutoff was really about 15% of 100Ah, hence the depleted battery by morning.

You might want to try to verify that the battery manufacturer put in the correct BMS if the total capacity figure is not an adjustable setting.

[bajaphile]

Well, it appears that the SOK manual states to re-calibrate the SOC, it requires it to be fully discharged and recharged 2X!? Seems weird to me. I was able to get into the settings and pressed "reboot" and nothing of course happened. I guess I'll try searching a bit on that and maybe sending an email to SOK directly. 

 

Also, apparently there is more settings on the Renogy solar charger for LiFePO4. I can set the voltage to 14.6. I was assuming that since it was charging at ~13.5V and 10amp it was getting at least some charge, but maybe I was wrong! Sierras were pretty warm this weekend so I'm assuming the fridge was working a bit more than usual. 

 

Jon, your logic makes sense to me... thanks for your input. 

Edited by bajaphile, 13 September 2023 - 01:33 AM.

[Jon R]

I’m glad to help. It sounds like you’re getting it figured out.

[rubberlegs]

I've had an SOK 100ah battery, heated and bluetooth, since last January and have used it quite a bit. It's a good battery except for the CALC Capacity that it measures. Jon R got it right -- it creeps up over time. The best place for advice is on their facebook group page: https://www.facebook...oups/sokbattery

 

If you aren't into Facebook, the advice always given there is to ignore the CALC Capacity and SOC that the bluetooth app shows. I was pretty disappointed as I loved the way the app worked. They all say the rest of the BMS functions work well. Many people in that group recommend this inexpensive shunt: https://www.amazon.c...B07FGFFHC6?th=1, so I got one and am quite happy with it -- it's pretty simple to hook up. Or pay more for a fancier Victron, etc. shunt.

 

Some of their latest batch of batteries may have fixed this issue with an improved BMS, from what I've read. I asked if a new BMS was available but at that time, it wasn't. I think some people have gotten new BMSs recently though. I've not heard of any free options and not sure what they cost.

 

I tried discharging to 0% and up to 100% twice as SOK recommends, but my charger is just a couple tenths of a volt too low for the BMS to think the battery is fully charged (I think 14.4 vs 14.6 but don't recall the exact numbers). It seems wrong to discharge to 0% anyway... but they say it'll go thousands of cycles.

 

Here's what one guy said when, like Jon, I wondered if the SOC was based on the CALC Capacity:

Your intuition about the SOC being calculated based on the CALC Capacity is correct.

Some points that may be helpful:

• the BMS has an internal coulomb counter
• it uses the value in the coulomb counter vs the CALC Capacity to determine SOC
• when the battery goes below 11V OR the BMS goes into cutoff due to undervoltage, 
the coulomb counter will reset to 0
• when the battery gets to 14.6V/the BMS cuts off due to overvoltage, 
the CALC Capacity is set to the value in the coulomb counter

So from this you can kind of calculate your way through an SOC estimation or an amount 
of discharge/charge estimation as you described.

One caveat is that the CALC Capacity overflows at about 308Ah so if your CALC Capacity 
drifts to more than that then it will be nearly impossible to estimate the SOC..

Other than the SOC disappointment, the battery is amazing compared to our AGM battery!

Edited by rubberlegs, 16 September 2023 - 06:38 PM.

[bajaphile]

@rubberlegs Thank you for your reply!

Good to know that the SOC is a common issue on the BMS. Guess I'll just ignore it and monitor via voltage as usual. Thought it was cool to see the SOC like you would on your phone battery. Oh well.

Got to admit I am still a bit "surprised" about it kicking off overnight... Maybe the solar charger wasn't setup to 100% but it must have still been charging it to some extent. 200W solar vs a 60W fridge should be more than enough.

I have a Renogy shunt with battery monitor which has been useless since I went to LiFePO4 -- Always says 100% but does monitor amp draw (or charge). After setting it up for a 100Ah battery, it still doesn't really do much. I'm assuming the SOC on that is based on voltage alone and does not have a coulomb meter .

**EDIT** User error on my part (as usual). The Renogy does use a coulomb meter via the shunt. Apparently the settings had reset after I did some electrical work shortly after installing my LiFePO4 battery! Ive set it up again and it's working great and will be my SOC meter. Nice not needing to have the phone for that anyways.

Thank you guys for your replies.

Edited by bajaphile, 03 October 2023 - 01:04 PM.

[rubberlegs]

Unfortunately, voltage isn't as accurate to estimate SOC for lithium batteries as it is for lead acid. The voltage is nearly the same over most of the percent range, until all of a sudden near 100% is goes up. (It also goes down rapidly near 0%). But lead acid batteries have a more voltage range over the SOC range. Your shunt is probably seeing a pretty high voltage (higher than lead acid batteries by the way), thus the 100%.

 

So you're probably right that your shunt is measuring volts instead of being a coulomb counter. So you could get a new shunt... but more money! The one I mentioned is $44.

 

SOK recommends 14.6V charging, by the way, and our Renogy charger is a bit lower, I think 14.4V. So it never charges to the point where the battery's BMS thinks it's absolutely full (i.e., when the C MOS indicator changes state on your battery app), resetting the total range.

[bajaphile]

Just following up with this for anyone else...

1) I've discharged / recharged SOK battery 2x times now and the calculated capacity has not changed and cannot be manually set (from what I can tell).

2) Indeed my Solar charger was setup incorrectly. I assumed I just needed to set it to LiFePO4 and it would be good... Nope. Have to go into a sub menu to setup the LiFePO4 charge settings (14.6V bulk for example) as the standard settings only pump out 13.8v. Probably the reason why my battery died in the first place.

3) Renogy 500Amp battery monitor indeed works well with LiFePO4. Once again, user error. I had set it up correctly but had forgot I went back and rerouted my wiring a week after and it reset back to default. Woops. It will be great for a SOC monitor since the SOK app is showing incorrect capacity.

4) While I've complained about the Victron DCDC charger having horrible Bluetooth range, it's been great ever since... Set it and forget it.