Fogstar 280ah fail, any ideas?

[Andy MacD]

I installed a Fogstar 280ah in my 2021 Adria Twin at the start of the year. Ten weeks in France, most of which off grid and using inverter for kettle and air fryer and no issues. Just back in France, been off grid last 3 nights and when I checked last evening I was at 40% remaining. Overnight the Truma control panel was on but not operating and the only other drains were a compressor fridge and devices charging through usb outlets. A Maxxfan was connected but switched off.

At 07:30 this morning the Fogstar app notification sounded on my phone and on checking no 12v. Fogstar app showed 0v. Checked the battery and 10.4v at the terminals. Interestingly the fridge was still powered, must have been off the cab battery which could have been tricky if left on for a while.

Started our days journey and the battery started charging at around 20ah, now in a campsite for the weekend to get mains so the battery fully charges and my blood pressure normalises.

The battery is now charging, everything seems normal and when powering up the fridge, inverter, Maxxfan I can’t replicate significant power use. I emailed Fogstar who said “Sometimes the BMS gets a little confused, and loses where it is in terms of the percentage indicator. It thinks it has more capacity left than it actually does, and all of a sudden it detects a low voltage and realizes its 0%. I wonder if that's happened here?”

That really doesn’t help as it implies I cannot rely on the app to determine charge left.

Below screen grab of BMS. Anyone any suggestions on what might have happened or how to avoid repetition? I’m going to be in France for another 5 weeks and now a little apprehensive about being off grid for more than a couple of nights at a time.

 

[Gromett]

Fogstar are correct. It is possible for a BMS to get the SOC wrong after a long period of not hitting full charge.
Roughly speaking and greatly simplified
The system works by what is called coulomb counting. This is measuring what goes in vs what goes out starting from a known point.
If you are using high current loads like an inverter and not charging back up to full ie going 25%-50%-35%-80%-38% etc the system can get out of whack because you always need to put a little more into the battery than what you get out.
If you never get up to the full 100% mark the SOC counter will drift.

So, it really does help to get a full charge in periodically to reset the SOC coulomb counter. It doesn't mean there is necessarily a fault with your battery.

 

[Andy MacD]

Thanks Gromett , every days a school day.

 

[Gromett]

Thanks Gromett , every days a school day.

Let us know how you get on over the next few days weeks please

 

[autorouter]

At 16:07 on 31st the voltage was 12.9V, current -3.9A. 12.9V means an SOC of about 20%, according to LiFePO4 voltage charts. The BMS says it's at 40%.

The next reading is at 07:36 on the 1st, ie 11 hours later. If it was losing 3.9A for that time, that's 3.9 x 11 = 43Ah, which is about 15% of the 280Ah capacity. So if it really was at 20%, it would have dropped down by 15% to 5%, at the point when the BMS triggered the alarm.

I don't know what that 3.9A is all about, but could be worth keeping an eye on it.

^{_{Subscribers  do not see these advertisements}}

 

[Gromett]

The next reading is at 07:36 on the 1st, ie 11 hours later. If it was losing 3.9A for that time

Looks like it dropped down to zero sometime between the two timestamps.

 

[Andy MacD]

At 16:07 on 31st the voltage was 12.9V, current -3.9A. 12.9V means an SOC of about 20%, according to LiFePO4 voltage charts. The BMS says it's at 40%.

The next reading is at 07:36 on the 1st, ie 11 hours later. If it was losing 3.9A for that time, that's 3.9 x 11 = 43Ah, which is about 15% of the 280Ah capacity. So if it really was at 20%, it would have dropped down by 15% to 5%, at the point when the BMS triggered the alarm.

I don't know what that 3.9A is all about, but could be worth keeping an eye on it.

If I had a load of 3.9a I could understand it but all I had on overnight was the fridge and a couple of iPads charging.

Fogstar’s final comment was that the primary function of the BMS is to keep us safe not count current and to ensure the battery shuts down without any damage. They say if I want to know the state of charge I should install a victron smart shunt. Funny they don’t say that on their website.

 

[Guigsy]

If I had a load of 3.9a I could understand it but all I had on overnight was the fridge and a couple of iPads charging.

Fogstar’s final comment was that the primary function of the BMS is to keep us safe not count current and to ensure the battery shuts down without any damage. They say if I want to know the state of charge I should install a victron smart shunt. Funny they don’t say that on their website.

I've got a FogStar and a Victron shunt. I can't see how the Victron would be any more accurate.

 

[autorouter]

If I had a load of 3.9a I could understand it but all I had on overnight was the fridge and a couple of iPads charging.

What is the fridge consumption, on average per 24 hours. That figure is usually in the manufacturer's spec list. It's likely to be about 40Ah (= 480Wh). So for 12 hours it's about 20Ah. Add in the iPad chargers to make it about 25Ah. That's not enough to be the whole story of the 43Ah loss I mentioned, but it's a large chunk of it.

 

[Andy MacD]

Would need to work it out. Pre lithium my 100ah AGM managed for 2 days easy so 40% remaining overnight shouldn’t have been an issue. Going by Fogstar’s comments and advice seems anything the BMS tells you should be treated with caution and should not be relied upon to be accurate.

^{_{Subscribers  do not see these advertisements}}