LiFePO4 Install

[john61ct]

> it was the only way I comforted myself in knowing a full balance had been achieved

hoping, not knowing.

Better approach is the very low rate, 1-2A and at the lowest possible voltage, just above "start-balance" setpoint.

Still not really knowing, but at least less stressful to the battery.

 

[luthj]

I would guess what happened, is that one cell group drifted too low or high. Its not that difficult with partial cycling without enough balance time. When that cell group went high/low the bank disconnected. Now, normally the bank will reconnect when the issue is resolved, or in the case of low voltage/cell when a charge is present. Your attempts to charge the bank didn't resolve the imbalance, but a higher voltage was enough to get the deed done? We are talking 50-100mv between cell groups.

Its possible that the bank will disconnect from a load if the difference between two cell groups is too large? In which case it would disconnect even if the cells were within the acceptable range. T

If the bank cycle tests good, its probably going to be fine. I would still suggest a warranty return if BB will allow it.

A once in 250 cycle balance charge at 14.6V for a few hours will not cause significant pack aging, and may be necessary depending on what type of balancing is being used, and how well the cells are matched.

 

[dreadlocks]

should I increase the maximum absorb time in my Victron and back off its voltage to 14.2? from my understanding is there's an underlying algorithm that every morning looks at the starting SOC, and uses that to determine the absorb time.. the MAX of 30 means that if the battery was at like 70% in the morning, then it could of only held absorb for a fraction of 30mins..

Disregarding Battleborn's suggested smart solar settings, and digging through the SmartSolar manual produced by Victron it says the following:

Default setting, LiFePO4 batteries LiFePO4 batteries do not need to be fully charged to prevent premature failure. The default absorption voltage setting is 14,2V (28,4V). And the default absorption time setting is 2 hours. Default float setting: 13,2V (26,4V). These settings are adjustable.

If BattleBorn starts its balancing at 14.2v, then setting absorb to 2h and max charge to 14.2v as the Victron Defaults are would seem to be more aligned with the philosophy @john61ct has on balancing.

I'll find out from the guys at Battleborn if the balancing will occur at 14.2v, and if so toss this PD Charger asap and replace it w/a Sterling that I can set my own params on.

 

[luthj]

It May have value. In your case the bad ground connection may have caused repeated cycling of the battery in the lower knee region? That could have pulled one cell low. I would say 14.2V and 1 hour is reasonable every 30-50 cycled.

 

[dreadlocks]

Well i just charged it back full, this time basically doubled the load w/~8A.. it didnt last 10 seconds before BB went back into protect mode



First flatline is it sitting at 14.58v, 0.0A input.. second flat line is when BMV lost all power because battery is 0v now.. looks like ~7-8A was enough to draw it straight below 12v, right off the charger.. barely had time to flip the circuit and turn around before everything died.

 

[luthj]

Hmm. Maybe the bms solid state relay is damaged? So it overheats and drops out? Sounds like a warranty return to me.

 

[dreadlocks]

yeah thats what i'm thinking.. this will come back and look just fine and if I apparently keep the loads mild.. it does just fine.

applying what I've been experiencing, where it seems fine at low loads, and fails at high loads, to Occam's Razor would seemingly point to a BMS failure of some sort, the cells appear okay externally..

Ive been pumping more amps into it than 8A tho, so I'm not sold on a SSR

 

[john61ct]

Looks bad, whether BMS failure or cell, the latter won't show, need access to test.

But.

Have you actually driven Absorb to 100% Full yet?

measuring with an ammeter that current has tapered down to the 1-3A range?

And once you do, do the capacity testing at a lower current that doesn't shut the (broken?) BMS down

______
Re Victron "black box" AHT algorithm

Victron MPPT Battery Charging Algorithm - Cruisers & Sailing Forums

The Victron MPPT regulates the battery very differently to all other solar controllers. Although some aspects of the algorithm are used by other brands, the net effect is the Victron solar regulator

www.cruisersforum.com

 

[dreadlocks]

Let the LED lights (<5A) draw it down to 22% SOC, never went into protect mode and voltage seemed reasonable at 12.91V



alert is for low SOC, so the battery still has basically full life it seems.. as long as I dont draw more than ~5A outta it.

 

[luthj]

Yeah, sounds like a BMS issue. I would suspect that the solid state relay has an issue. Either with its embedded thermistor, or its damaged. So the BMS is dropping out to protect it.

 

[dreadlocks]

yeah you might be onto something w/the SSR, now that its drained this far and not just a few AH like earlier tests I'm a bit surprised to see the battery only taking like 11A of charge current.. I can turn on as many accessories as I want in the trailer and the charge current stays 11A, indicating the battery is limiting the charge current and not the charger.. otherwise if it were charger, me turning on lights and stuff would see the charge current to the LFP go down equal to the appliances load I just turned on.

I see the Victron LFP's have two IO commands, one to start/stop discharge, and one to start/stop charge.. do these BMS's have two solid state switches behind diodes or something? so one allows input and the other allows output and the BMS can pick and choose?

 

[luthj]

I am almost certain the BB bms uses a single solid state relay. Slow charge current indicates excessive voltage drop across the relay. During discharge that would rapidly overheat the relay, causing drop out.

If BB denies a warranty claim, you could always crack open the case, and replace the faulty relay. I am certain you could find a off the shelf replacement that would work.

 

[luthj]

Typically a SSR fails closed. High resistance failures are not as common. Though, I wonder if BB uses a bunch of parallel transistors to switch current? If so, you could have a bunch of them fail, and the remaining ones are overloaded.

 

[dreadlocks]

yeah im thinking it has a ton of parallel mosfets and I cooked a good portion of em... would explain much of what I've been seeing.. maybe a few let go, and then my inverter took out a bunch more with my coffee maker.

 

[luthj]

If thats the case, then BB is using a bad design. To my knowledge, most commercial SSRs in this capacity range, use 1-2 large custom made semiconductor blocks, not discrete components (many cheap BMS do use lots of commodity sized FETs).

Of course if they did use many small FETs, that does mean folks who are out of warranty can replace them manually, though who knows if access is possible.

I honestly hope that BBs warranty stands the test.