Batteries wear out.

harryS

Well-Known Member
I have a 48V battery made up with Samsung 30Q cells, purchased in August of 2015. These cells are rated at 3AH, so the battery would be 12AH, so that's about 500 useable watt-hr (576 wh in theory).

Coming up on 22 months, it pumps out half that power, about 220 watt-hr and it's down to 45 volts. I probably helped beat this battery to death. First year I had it, I kept it charged when it wasn't being used much. Bad for lithium ions.

Well, it's still good for about 20 miles.
 

Alex M

Well-Known Member
I probably helped beat this battery to death. First year I had it, I kept it charged when it wasn't being used much.
Keeping them charged - nothing is wrong with that. Recharging after every trip and maintenance charging every 2-3 months while in storage.
If you ride it down to zero every day, 700 cycles in 2 years , it's about as good as it gets.
 

Solom01

Well-Known Member
Actually there is a lot wrong with that. If you're content to have a battery die quickly in order to get longer distances, etc. that is your choice, but if it's done from a lack of knowledge of battery chemistry and bad information given by bike manufacturers it is bad for the environment (lithium battery production is not a green activity), expensive and wasteful. In the car world companies that do it right like the Prius and Chevy Volt go on for over 100K miles, companies that don't like the Nissan Leaf were loosing about 10% capacity a year.Not charging it fully greatly extends their life, not by a bit but by a lot. Of course the Vendors would love to sell you a new battery every few years.
 

Alex M

Well-Known Member
Nobody said about charging it "fully" every time. It's important to recharge it often, so it wouldn't sit empty.

BMS should take care of the top 5-10% so you wouldn't have to worry... if it's designed properly. Not always the case with lower-end systems.

What is clear (to me), is that any battery would benefit from not DIS-charging completely.
 
Last edited:

harryS

Well-Known Member
It's only been discharged fully twice, and both times were this week. My wife ran it flat on Sunday and then I put a wattmeter on it and ran it til the controller started blinking. Confirmed that it no longer had the full range.

This was more of a spare battery, and lack of use, plus I kept it topped off the first year, is probably what killed it. Tens of cycles, not hundreds of cycles.
 

Alex M

Well-Known Member
Lack of use shouldn't be a problem if you keep it at 50-80% and occasionally top off to upper 90s ~100. There are no hard numbers and manufacturers are keeping users in the dark because it's easier this way. Imagine if they told an average Joe Schmoe that his 600WH battery is more like 400WH, and stated max range 30 miles and top speed 25 mph is "kinda true" but not recommended :)

I've been eyeing that programmable Grin charger for a while. It costs almost like a new battery, and then there is a question whether your particular BMS will do its balancing job at 80%, or it will need full 100% for that.
 

George S.

Well-Known Member
If the pack is at 45 volts, that doesn't sound right. There's a lack of troubleshooting advice for battery packs. The cells in a pack are supposed to be balanced and they are supposed to age evenly. If you took all the cells out and separated them, what would you find? The best you can do is open the pack and see if each parallel group is at about the same voltage.

I don't really know what happens if the BMS goes bad, or one cell in a pack goes bad, what can cause a big drop in capacity other than aging of all the cells. The Vruzend system gives access to the cells, and maybe it will facilitate more complete diagnosis or more experimenting with pack failures.

I have a 2 year Lifepo pack from China. Everyone says these packs are outmoded, obsolete, poorly made junk. I tested the capacity a month ago, and it has 92% of the original 15 amp hours. It's a fifteen pound anchor on any bike, but the darned thing works and I charge it to 100% most of the time.

Some of these fancy cells, I don't know. Even the older 2000 mah cells seem to hold up better. We've been oversold battery tech.

For fun, here are 100,000 18650 cells in a one megawatt configuration.
 

harryS

Well-Known Member
45V volts is down from a full charge of 54.6Volts, after 220 watt hr has been expended. Nothing appears wrong except a lack of capacity. I should open it up and check the balance wires. I will after I get a replacement.
 

George S.

Well-Known Member
At 45 volts with a 13 series pack, the cells are at 3.46v. That's not empty. Generally 'empty' is when the LVC trips. A lot of these cells go to around 3v. A bad cell or two can bring down a parallel set and that set might drop a lot, and that might set off the LVC. I get this sense that maybe packs are tossed because of the interactions between the BMS and the weakest cells. But the system is not set up to replace weak cells in a welded pack that might have to be shipped. You don't want to ship a defective pack.

With something like Vruzend, I can remove the buss bars and check each cell. With a welded pack you could get it out of the case, check the voltages in each set, and put the battery on a decent load, then check the voltages. You need a load to get a real voltage. There are ways to assemble a pack where you can check each cell. But that depends on how easily you can replace one cell or one part of the pack.

If every cell is at 50% capacity, that's one thing (and pretty bad). If you have weak cells and the way the BMS works is shutting down the pack at 50%, that's something else. If you have bad cells that's a different risk.

I told Hicks at Luna that he really needs a little computer in his packs to get solid information. He's really pushing the cells, and that might require more nuanced management. He should probably wire the packs to send information back to Luna. Cells are getting cheaper, of course. Replacing weak cells, if you find them, is not for the average consumer. I guess it is really a 'disposable' model. If you have most cells at 90% but the pack is not useable, that would annoy me.

If you take the pack apart you can buy little chargers for a cell or two on Ebay. I bought a little dingus that discharges a cell and shows the mah. Two bucks or something. The Chinese are really into this stuff.
 

harryS

Well-Known Member
No, 45 volts isn't empty, but I bet there's not more than 50 watt-hr in the battery before it shuts down. I'll pack a spare battery this week and ride this one this down to LVC.
 

harryS

Well-Known Member
Put the battery on my 36V bike, and ran it down to the LVC of the battery, which turned out to be 39.6 volts. I got 4.8 more miles out of it until the power winked off and killed my wattmeter, but that's around 50 watt-hrs for that bike.

In any case, 39.6 volts is just over 3 volts per cell, and 3 volts is also LVC for most BMS circuits. I think this says all 13 cell groups are prematurely tired out.
 

George S.

Well-Known Member
Put the battery on my 36V bike, and ran it down to the LVC of the battery, which turned out to be 39.6 volts. I got 4.8 more miles out of it until the power winked off and killed my wattmeter, but that's around 50 watt-hrs for that bike.

In any case, 39.6 volts is just over 3 volts per cell, and 3 volts is also LVC for most BMS circuits. I think this says all 13 cell groups are prematurely tired out.
Thanks. I have a 30q pack that is now one year old and has only been used steadily for a few months. I guess I will do a capacity test with a watt hour meter.

I have a pack with Sanyo GA cells. It's kind of a favorite with performance riders. The data sheet shows a big loss of capacity quickly if you run it hard. Apparently these 30q cells are for power tools, so high discharge, but they aren't great with hard use, section 7.10. But they say 60% after a lot of heavy discharges.

It seems easier to go fairly slow, get decent exercise, and use the battery for hills and headwinds. But we just don't have very good numbers, even if you treat your batteries pretty sanely. I hope the cells get really cheap.:) I don't really want to spend $400 for another pack, since they can't be rated.
 

George S.

Well-Known Member
Thanks. I have a 30q pack that is now one year old and has only been used steadily for a few months. I guess I will do a capacity test with a watt hour meter.

The pack is rated 6 AH and 50-52 volts new. I ran it for 15 miles on the bike, so up to 700 watts draw. Put it on my discharge system, just lights, so 70 watts. That's not an ebike draw. I got maybe another 40 watts, but the voltage was dropping like a rock.

For me it is now a 210 wh pack, or 70%. This is one year and maybe 30-40 cycles. It's still something I can use but now I feel I need to throw another small pack in another bag as a reserve.

So Harry, the good news is it's not you. The bad news is, well, you know the bad news. Your original conclusion was completely correct.

rundown second phase.JPG
 

JoePah

Well-Known Member
WEll that's a pretty nice size batteries.. It really depends on how much load you put on those batteries every time you went out.. If your controller wasdrawing 2C or 3C, around 1 to 1`.5 kW, its discharge rate will definitely degrade capacity quickly.
 

harryS

Well-Known Member
Update on this after a year. Unexpected, but happy ending.

I opened up the case to check the individual cells, and was surprised to find out it was sold as a 48V vattery but what was inside were 14 cell groups meaning it was a 52V battery. The cells were Samsung 30Q, and all were sitting at 3.83 volts, which is what you get if you put a 48 V charger on them. Nice round sticker still on battery that says 48V. Honest mistake by a US vendor.

Full charge with a 52V charger (which I have) should add about 6 more AH into the pack. Now I can go melt my 750W motors. Just kidding.

So I guess I was charging it to 60%, maybe 70% all these years. It should be fine.
 
Last edited:

Gator

Well-Known Member
Update on this after a year. Unexpected, but happy ending.

I opened up the case to check the individual cells, and was surprised to find out it was sold as a 48V vattery but what was inside were 14 cell groups meaning it was a 52V battery. The cells were Samsung 30Q, and all were sitting at 38.3 volts, which is what you get if you put a 48 V charger on them. Nice round sticker still on battery that says 48V. Honest mistake by a US vendor.

Full charge with a 52V charger (which I have) should add about 6 more AH into the pack. Now I can go melt my 750W motors. Just kidding.

So I guess I was charging it to 60%, maybe 70% all these years. It should be fine.

Nice save on those batteries. That extra 30 - 40 % is going to feel real good while your riding =D
 

Reid

Well-Known Member
Update on this after a year. Unexpected, but happy ending.

I opened up the case to check the individual cells, and was surprised to find out it was sold as a 48V vattery but what was inside were 14 cell groups meaning it was a 52V battery. The cells were Samsung 30Q, and all were sitting at 3.83 volts, which is what you get if you put a 48 V charger on them. Nice round sticker still on battery that says 48V. Honest mistake by a US vendor.

Full charge with a 52V charger (which I have) should add about 6 more AH into the pack. Now I can go melt my 750W motors. Just kidding.

So I guess I was charging it to 60%, maybe 70% all these years. It should be fine.
Do you suppose the pack has a 52V BMS? If so, does that mean the pack may not have reached a balancing point voltage? Evidently, it didn't hurt.

WHY, with all the education and lecturing and such from battery authorities do none of the makers offer as an option, a 52V battery with a BMS that balances it out and charges it as a 48V battery? And so we would have a very long life pack.

It isn't like all ebikers are ultra-needful of the highest energy density possible, which is the case with the usual charging to 4.2V per cell.

Hell, I'd buy a 60V battery if it would be charged and balanced out as a 52V battery. Say it was the size of a 20AH pack and only gave 13AH (just making up numbers here), I'd be good with that.