BtrPower - LiFePO4 failure after only 20 uses - WARRANTY WAS HONORED

HansTrio

Member
Region
USA
More Diagnosis of problem.

The battery is a 48v 20Ah unit.
It began to end the charge cycle early going from 55v to 58v in less than 1 minute.
A capacity test showed only 8.13Ah from the battery.

After taking the battery apart.....
Cell #7 was at 2.7v while the other cells were at a higher voltage. I think one more may have been a bit low.
I checked the resistance between each BMS wire (16) at the battery and the connector and found no issues.
I checked cell 7 by individually charging it to 3.6v then running a capacity test down to 2.7v and capacity was nominal (normal).
I'm checking each cell for capacity but it will take a few days. So far, cells 1,2,3 and 7 tested near full capacity but the lowest capacity was
Cell 1 - 19.430Ah
Cell 2 - 19.330Ah
Cell 3 - 19.770Ah
Cell 7 - 19.640Ah

I've also checked nearly every component on the BMS since I have datasheets for all the components but have so far found no issues with the BMS
It could still be there since I have a few components left to test.

I visually inspected the soldering of the wires from the BMS connector to each cell and I did find some questionable soldering where part of the wires at the end were not embedded in the solder.
One or two looked bad enough to possibly have caused a higher resistance to current from the BMS so I will have to re check that.

In the process of this I discovered something potentially bad for the battery pack.
The BMS has a circuit board that uses a special IC chip called a HY2112-CB to detect overvoltage/overcharge/overdischarge etc.
The problem is that this IC chip does not detect Overcharge until 3.75v +/- .025v
That means that some cells will possibly reach 3.75+.025v = 3.775v
The Maximum specified SOC for Lithium Iron Phosphate batteries is 3.65v. Therefore, some or more or the cells in the BtrPower battery will likely have reduced lifespans.
It could be significant. I would HIGHLY recommend you pay careful attention to your battery as it is charging and watch for a sudden rise while charging to full charge.
That would indicate one of more cells are reaching full charge before the rest and tripping an overcharge response while other cells are at states of less charge.

I would recommend NOT fully charging this battery pack (with this BMS) but rather disconnecting at the 80% state of charge point.
That would be at around 55.2v
This 55v is different that the one listed about because the one above is caused by the detection of a cell that was overcharged, resulting in lower capacity of the pack.
Cutting it off manually at 55v means it has not yet detected an overcharged cell (at 3.775v)

Anyway, the amount of capacity you get for going over 80% is so small it is not worth the risk of overcharging and shortening the life of the battey pack.

I will be changing the Chips in my BMS to a different chip to lower the Overcharge Detection Voltage. Unless you are an experienced electronics tech good at soldering, and know and are practiced in the techniques for SMD soldering, this would be very difficult.
Alternatively, you could locate a better BMS. and just install a better BMS.

HY2112_ProtectionChart.jpg
 
Last edited:

tomjasz

Well-Known Member
Brilliant!!

Well done. I hope EBR posters read this in it’s entirety. A great demonstration of potential problems and a glimpse into what we may get on a unrealistic budget.
 
Last edited:

HansTrio

Member
Region
USA
Truth be told, I've looked around and haven't really found a better ebike BMS. usually they have to be small and compact.
It seems they all use these same chips. Even the really expensive ones.

I don't think BtrPower is trying to sell a low quality battery (actually, the longevity of the cells themselves determines that. And keeping them under 3.6v and over 2.9v plays a large part in that.)
I would VERY much like to autopsy a "$800" top quality battery and it's BMS to see what the real world difference is.

I have the ability to do that if I could find someone interested in donating a no longer needed/ working specimen.
 

tomjasz

Well-Known Member
Been there, done that. There are a few pack breakdowns on endless_sphere
More important perhaps are buyers who insist on standards. I just bought a UPP battery. I had fits with them as a tech support guy for a reseller. I've learned that there ARE Chinese builders that CAN produce a decent battery. (I'm doing a tear down for another thread) BUT buyers need to understand the reliable standard to specify if they are DIY. THIS is the reason EM3ev has been so successful. CONSISTENT. The only significant criticism is contact time and shipping time. But any Chinese source worth a s*it will have longer shipping times.

I'm REALLY liking Jenny Mao. A young woman that is a great communicator. It's tough when we understand that Chinese pack builders are completely isolated from any insurance claims or product failures. We can only watch, share, and learn together. So we have EBR. A kinder gentler endless_sphere

BTW, I'm thrilled with my JM battery!

She's confirmed cell separted pack with a BT BMS available.
 

harryS

Well-Known Member
How are you checking capacity? Is it comparable to ebiking where you're pulling 200watts. or 4amps average thru the battery, with 500-800W bursts?

Maybe your cell #7 no longer can support a high load current. Or maybe it has developed a high internal self discharge. Since everything is the same, I suspect you will wind up with that battery unbalanced again if you try to re-use it.
 

HansTrio

Member
Region
USA
Cell #7 seemed ok at first but on further checking it seems it resists charging.
Meaning it has high internal resistance and / or low capacity.
How are you checking capacity? Is it comparable to ebiking where you're pulling 200watts. or 4amps average thru the battery, with 500-800W bursts?

Maybe your cell #7 no longer can support a high load current. Or maybe it has developed a high internal self discharge. Since everything is the same, I suspect you will wind up with that battery unbalanced again if you try to re-use it.

16 cells.
I'm still testing them. Internal resistance as well.
The thing that stands out is the discharge profiles. I really need a better load (capacity) tester and will probably invest in a good one soon.
Anyway, the rough curves I'm getting indicate that the different cells may have different Internal Resistances, capacities or even different quality solder joints because
I'm getting some significant differences in the discharge curves of the 16 cells.

For example one cell will drop to 2.7volts and have a flatter curve and hang there flat for a long time while another cell will hang at 2.8v for almost the entire curve then drop off sharply (ideal).
The cell(s) that drop to a lower voltage quickly still seem to have nearly the same capacity overall, but they provide that capacity at a lower voltage. Odd.
The cells aren't matched very well it seems at first glance. That or they've already sustained some damage.

But then, right now I'm only getting cell voltage accuracy to .1v. That's just not good enough. .01 would be the minimum I need. ,001 would be ideal so I can feed the data into DataExplorer and get some really accurate graphs representing their discharge curves. Then I could possibly swap cells around to build a more consistent, reliable battery pack.

I'm looking at the ICharger X6 or X8 and also the ET5410
Probably going to have to go with the iCharger X8 to get the higher wattage.
 
Last edited:

HansTrio

Member
Region
USA
So the tracking number I got was for another package......

BUT...I got a call today from a neighbor on another street (I know this neighbor) and he said he had packages for me.
One was the warranty replacement battery.

I had thought I got scammed because the tracking number I had was for a delivery to someone else.

Mistakes on top of mistakes but when the dust cleared I have the replacement.

Now to test it to see what I got.
 

HansTrio

Member
Region
USA
could you have any worse luck with this! Ouch.

Yeah...if I never got the replacement.
But I finally did so things are definitely looking up finally.

The last week we have had hell with deliveries.
Both UPS AND FedEx have been mis delivering packages like crazy.

The series of events is both incredibly ironic and odd, leading to a lot if miscues and incorrect perceptions.
I'm a bit perplexed at the complexity of the whole mess. I could explain it but it would read like a TV soap....so I won't.
If not for good neighbors a lot of stuff would be lost.

But finally...replacement is in hand. It seems BtrPower honors their warranties.
Now if only FedEx and UPS would clean up their act.

Time to capacity test the replacement.
 

PedalUma

Well-Known Member
Region
USA
I have 6 batteries lost in transit for two months. The other 6 in that order were built to my specifications and arrived in 14 days.
 

HansTrio

Member
Region
USA
The replacement battery has different dimensions than the original.

The replacement is both wider and shorter. Dimensions to follow.
 

HansTrio

Member
Region
USA
I have 6 batteries lost in transit for two months. The other 6 in that order were built to my specifications and arrived in 14 days.

I feel your pain. At least now I know it's not just my deliveries. Sorry you had to go through that. It's awful.
 

HansTrio

Member
Region
USA
How are you checking capacity? Is it comparable to ebiking where you're pulling 200watts. or 4amps average thru the battery, with 500-800W bursts?

Maybe your cell #7 no longer can support a high load current. Or maybe it has developed a high internal self discharge. Since everything is the same, I suspect you will wind up with that battery unbalanced again if you try to re-use it.

As you may know, capacity is related to load.
Generally lower loads will yield "higher capacities" due to lower losses from heating of the cells internally.

LiFePO4 are very good at high loads however.

So I am testing them with a DL24P capacity tester which unfortunately can't give a significant load.
I did, however, do a few tests using an inverter and a heavier load of 15Amps.

The results were very close at the high load and the low load, leading me to believe the results I ma getting with the DL24P are probably very close to the results I would get in real world circumstances.
The BtrPower 20Ah battery is good for about 19.3Ah. Close enough for me at the price point.
 

HansTrio

Member
Region
USA
You should know that by using eBay you give up the right to sue in court and accept arbitration as the resolution method.

Not necessary because BtrPower and iMotorBattery did send a replacement.
I'm prepping it for a capacity test now and will post the results.

ebay is sued all the time. Just because the fine print says that "by using this service you agree to blah blah blah....."
Doesn't remove your right to sue and ability to prevail.
Most big companies have similar fine print, utilities etc etc etc.

There's a lot of information on the web about this. Many legal opinions and decisions.
It's a complex legal web.
But yeah, I know where you're coming from.