Why does my battery cut out at 51.5 V

[RTLSHIP]

thanks, amberwolf, just rode in to work on old pack. 4.35 miles into headwinds most of the time.18 minutes. The other,newer, pack is in patio still on charge.
The symptoms on the old one are: it charges up to nearly 58v. Once charger is removed, it begins to drop and in say 5 hours
is around 49 v sitting. It can be quickly recharged from 49 v to 58 v within 3 or 4 minutes. I run it hard for 4 miles and running voltage is 45.9. Is it running on 14 out of 16 series? I can go down to around 44 at which point it will shut down regardless of BMS plugs in or out. Does that make sense?
There are only 16 pins on the bms plugs, so to test voltages, what do you mean by pairs of pins?

 

[BVH]

As others have said here, it would appear that you are missing "2S" out of your total "16S" pack, My 16S packs settle to about 55.5 after an hour or two off the charger. They stay there days, weeks and even months, not losing any appreciable Voltage.

 

[e-beach]

...
The symptoms on the old one are: it charges up to nearly 58v. Once charger is removed, it begins to drop and in say 5 hours is around 49 v sitting. I....... I run it hard for 4 miles and running voltage is 45.9. Is it running on 14 out of 16 series? I can go down to around 44 at which point it will shut down regardless of BMS plugs in or out. Does that make sense?
There are only 16 pins on the bms plugs, so to test voltages, what do you mean by pairs of pins?

58v / 16 = 3.625v per cell average.

49v / 16 = 3.062v per cell average.

44v / 16 = 2.75v per cell average. Most likely your controller shuts off at that point regardless of the BMS being plugged in.

The 45.9v under load sounds like weak cells sagging under load. However 4 miles of WOT is still not much for a 12ah battery that is supposed to be fully charged.

As for testing the BMS:

Charge your battery fully to 58.x volts. Then go through the process of checking the voltages of you BMS since (voltage) wires.

1.) Start on one side of your connector or the other. Usually there is one wire colored differently then the others, but not always. Place the negative probe of your DMM on the end of the BMS connector (the colored wire if you have a colored wire) and then place the positive probe of the DMM to the wire adjacent to the end wire. Read the voltage and write it down. (be careful not to short the wires!!!) By doing this you are checking the battery at the end of the pack. We will call that number 1.

2.) Move the positive probe one wire over and then move the negative probe next to the positive one. You are now checking battery number 2.

3.) Move the probes again and you will be checking battery 3. keep checking the wire like that until you are at battery 16.

When you write down the voltages, writing them down like this:

1) X.XXv
2) X.XXv
3) X.XXv

etc. Just keep writing the voltages down until you have tested 16 cells.

Now, fully charge your battery again, (Make sure your BMS is hooked up before you start charging.) Once fully charged ride your bike until you have LVC. Then repeat the process of testing the batteries and write down the voltages again. By comparing the first and second set of numbers, it will tell you if you have bad cells mixed in with good cells, or if they are all bad.

Does the make sense?

 

[e-beach]

In fact, if you want to, fully charge your battery and test the voltages, let it sit until the voltage drops and test it again, then fully charge it, ride until LVC and then test it again. The three sets of numbers will tell you what you need to know about the cell groups health.

 

[RTLSHIP]

I'm referring to another pack at this time, a 48/15 lifepo. The 48/12 is still on the charger, day # 2.
Another question: I can completely disconnect BMS plugs (on old pack) and unsolder wiring. Reconnect all the reds (discharge and charge wires) and insulate. Same with black. Then ride to see what happens and recharge. If results are same, I can reinstall BMS. I say this because this battery rides the same range whether bms molex plugs are attached or not. Is that a waste of time?

 

[e-beach]

No reason to solder or un-solder anything until you have charged, metered, run to LVC (or at least get it close to LVC) and meter again.

It is the best way to know what each cell is doing.

We are working on a process of elimination. We need to eliminate whether you have bad cells as the problem first.

 

[RTLSHIP]

I have muddled things up a little by discussing a second battery pack. I actually have 3 packs. The pack in the above title "51.5 V
" is still sitting on charger with green light on.


After 5 mile trip home, here are the voltages on the old 2009 pack: 2.75, 2.82, 2.83, 2.85, 2.85, 2.86, 2.86, 2.85, 2.86, 2.85, 2.85, 2.85, 2.85, 2.85, 2.86 and 2.85. grand total is 45.49 v. So this old pack is apparently balanced and running on 16s not 14s.
It suffers from a serious voltage sag. I will recheck when recharged but it looks like a very consistently degraded or damaged pack.

If anything, I learned how to read the BMS board correctly.

 

[DAND214]

I know this is about the battery but I think there are other concerns. Your other packs are also shutting down too early in my mind. If you can't get 1.5 miles at less than 18mph there is more problems than just that pack. At that speed I see 2+ M/AH. A 28 AH 44v (12s) pack gets almost 65 miles at 15 to 18mph. I would think to look further than the battery. Even if the controller is low amp.

I thought I still had my old/new Ping cell chargers but didn't see them. Might of left them back north during the move. Where about are you in Florida?
I will look more later.

Yes the battery sounds bad and yes charging for long periods might help.

Dan

 

[RTLSHIP]

Boca Raton. New Port Richey is Gulf coast and gets the stronger hurricanes because of the warmer Gulf. But we're all fair game to a storm. As for the old pack, I can get 5 miles at 17 mph if lucky. I just charged it up and 10 minutes after all series were between 3.44 v and 3.46 v. One hour after removing charger, all series measured between 3.25 v and 3.27v. So by morning, I figure those numbers will be at 3.1v. The controller I'm using is a Lyen version. He makes them and sells them here. I have 2 other controllers, but don't suspect they will increase range. I do have a new pack (48v 20 ah lifepo) that is only ridden twice but it is not all wired down yet permanently or protected in a case.

What can I do with the super old one? Charge it for a week or trade/sell? That other one is in charging right now and I don't want to lose it because it has so few cycles.

 

[RTLSHIP]

12 hours after removing old pack from charger, it still rests at 53 or 53.1 v. It was 53.2v 1 hour after removing from charger.
So this old pack is holding the charge. Also, the 16 series voltages were all in the 3.26v to 3.28v range. So this old pack is well balanced when charged and discharged. But it will drop in voltage to balanced 49 then down to a balanced 46v within 2 miles of riding. The other newer pack is still charging.

 

[e-beach]

Sounds like your "older" pack is just worn out. Usage (cycles), heat and age all take there toll on lithium batteries. If you can get money for a pack that has limited range why not sell it. Put the money into something else.

 

[RTLSHIP]

who would want it. maybe solar DIY?

e-beach, this old pack rests nicely with the bms plugs removed. Once I replug, potted material on board gets warm
and voltage drops and drops. Maybe the bms did parasitic like damage at some point

 

[RTLSHIP]

OK, 1 week mark at 5:20 am today on lifepo4 charger @58 v. I unplugged charger this morning and waited 40 minutes. Voltage before hooking into controller was 56.3. I rode long way to work, 5.34 miles in 21 minutes. Top Speed was 26 mph. Voltage now rests at 52.8v.
There was little wind, and some poorly lighted areas slowed me down. But there were no cut outs. So apparently the prolonged charging helped.
At least, subjectively, it seems like an improvement. The conditions were nearly optimum and I didn't go full throttle much. Previously, this 48 v 12 ah lifepo pack would drop to 51 to 51.5v and cut out with 4 miles riding.
I use an SAE connector from pack to controller and it has a nice crackle sound to it when plugging in even at 52.8 v. This suggest to me the battery is strong. I will recharge and ride it harder on the way home.

 

[john61ct]

I would not use an SAE connector for much over maybe 6-8A.

And they do corrode in use.

Replace with Anderson if space allows

 

[RTLSHIP]

I tried Anderson connectors and had no luck putting them together even with the correct tooling. The crimp piece had problems nesting into plastic housing and so I gave up on them. I probably still have some laying around. The SAE's I use have pretty thick wires and they have hardly been a problem for a dozen years with the rare exception of fraying . Corrosion can be remedied.
I have a brand new pack that came with 2 big gray 600A andersons with lead wires, but I only use it as an extension/disconnect.

 

[e-beach]

.... But there were no cut outs. So apparently the prolonged charging helped.
.....

Well, that is some good news. Have you ever put a watt meter on your bike to get an average of watts per mile you use?

 

[RTLSHIP]

I have never used a watts meter and don't necessarily want to hassle with one. I can say that re- charging time was about the same as usual, so it would seem there was an improvement.

 

[RTLSHIP]

Second Update: I rode it to gym, home and then again after dinner for a total of 10.2 miles (in 39 minutes) on a single charge up. The voltage was 52.4 at end of ride. This is much better and I felt it could go another 5 miles.
I am also trying prolonged charging with an old 48v lifepo pack that is 15 series (not 16 as I thought) and I'm using a SLA charger as it is lower in voltage that makes it a better and safer fit. Charging is mostly with bms wires out as bms gets hot with bms wires in.

 

[e-beach]

Just so you know it won't balance at sla voltages.

Also, when you say the bms gets hot.....how hot?

 

[99t4]

I tried Anderson connectors and had no luck putting them together even with the correct tooling. The crimp piece had problems nesting into plastic housing and so I gave up on them.

Counterfeit Andersons?

 

[e-beach]

I have a bag of legitimate Anderson's and all the parts needed to make them work in my e-bike box of spare parts. The reason they are there is because I have had the same problem as many, they are hard to work with. I prefer bullet plugs. Much easier to work with.

 

[amberwolf]

I am also trying prolonged charging with an old 48v lifepo pack that is 15 series (not 16 as I thought) and I'm using a SLA charger as it is lower in voltage that makes it a better and safer fit. Charging is mostly with bms wires out as bms gets hot with bms wires in.

WIthout those wires plugged in, the charger can overcharge cells because the BMS can't tell when they are full, so when they're not balanced, the ones already full get overcharged and increasingly damaged over time (when those cells probably started out as the best cells in the pack, vs the ones that didn't fill as quickly and drained faster, so were lower in voltage).

The BMS should get at least warm, and if it has large balance resistors (like some of the old Ping pack Signalab BMSs did) then they do get hot, because they are diverting a lot of current from full cells so the low cells can catch up faster.

If all cells are full, and the charger continues to output current because it's voltage is a little high, then the balance resistors will stay warm (or hot if current is still high, like if the charger is significantly too high a voltage for the pack, like a 16s charger on a 15s pack, etc).


The BMS can also be hot if it's being charged thru the discharge FETS instead of the charge FETs, if the discharge port was turned off by the BMS (like if the pack ran out of charge while in use). Or if the charging thru the charge FETs but they are damaged and can't turn on. Both of these are because the FETs would act like diodes and resistors instead of switches, and get hot when enough current is flowing thru them.

 

[e-beach]

I have never used a watts meter and don't necessarily want to hassle with one. I can say that re- charging time was about the same as usual, so it would seem there was an improvement.

I know what you are saying about the hassle. I stopped using one on my bike when one night I was riding back from work. It was about 8pm and dark. Suddenly my bike started cutting in and out. It took me about 1/2 an hour to figure out what the problem was. It turned out that the wires of the watt meter I was using had began to melt inside the insulation cutting out the power to my controller. After that I moved to a shunt style watt meter. It worked much better, but then one day a thief stole the meter off my handle bars when I was in a store. I haven't had any kind of meter on my handle bars since then. That has been at least 6 years. However, in that time I learned very well how many watts I burn per mile on average, that would be 27. So now I ride by dead-reckoning. Meaning that if I need to go a certain amount of miles I can reckon if I need to take a charger with me or not.

The point being is that if you have an accurate knowledge of your energy usage, you can account for an aging battery and judge your distances.

Or something like that. Anyway watt meters can be useful if the thief doesn't get it. :wink:

 

[RTLSHIP]

I am also trying prolonged charging with an old 48v lifepo pack that is 15 series (not 16 as I thought) and I'm using a SLA charger as it is lower in voltage that makes it a better and safer fit. Charging is mostly with bms wires out as bms gets hot with bms wires in.

WIthout those wires plugged in, the charger can overcharge cells because the BMS can't tell when they are full, so when they're not balanced, the ones already full get overcharged and increasingly damaged over time (when those cells probably started out as the best cells in the pack, vs the ones that didn't fill as quickly and drained faster, so were lower in voltage).

The BMS should get at least warm, and if it has large balance resistors (like some of the old Ping pack Signalab BMSs did) then they do get hot, because they are diverting a lot of current from full cells so the low cells can catch up faster.

If all cells are full, and the charger continues to output current because it's voltage is a little high, then the balance resistors will stay warm (or hot if current is still high, like if the charger is significantly too high a voltage for the pack, like a 16s charger on a 15s pack, etc).


The BMS can also be hot if it's being charged thru the discharge FETS instead of the charge FETs, if the discharge port was turned off by the BMS (like if the pack ran out of charge while in use). Or if the charging thru the charge FETs but they are damaged and can't turn on. Both of these are because the FETs would act like diodes and resistors instead of switches, and get hot when enough current is flowing thru them.

The voltage on the old pack is 54.7 with SLA charger on, and this appears to be a very balanced pack despite age. My mistake apparently was that I thought there was 16 series. There are only 15 wires and the board goes from B1 to B15.
So if the SLA charger is putting out 55.1, I don't see much harm. It can get hot with BMS unplugged when using the 58 v charger. As far as warm vs hot, I can touch it with my fingers with no burn.
The heat is right on the plastic potting and some on the 16 little fets or whatever. I have been using pack with unplug for some time and the range and voltage are unchanged. Whatever happens, I can't complain as the other pack works great after 1 full week of charging. I don't understand why I get a 55.1 v reading and a 54.7 v from using different negative spots on board. The charging negative part is 55.1, the other 54.7. Perhaps the .4 v drop is due to resistance?

 

[RTLSHIP]

I have never used a watts meter and don't necessarily want to hassle with one. I can say that re- charging time was about the same as usual, so it would seem there was an improvement.

I know what you are saying about the hassle. I stopped using one on my bike when one night I was riding back from work. It was about 8pm and dark. Suddenly my bike started cutting in and out. It took me about 1/2 an hour to figure out what the problem was. It turned out that the wires of the watt meter I was using had began to melt inside the insulation cutting out the power to my controller. After that I moved to a shunt style watt meter. It worked much better, but then one day a thief stole the meter off my handle bars when I was in a store. I haven't had any kind of meter on my handle bars since then. That has been at least 6 years. However, in that time I learned very well how many watts I burn per mile on average, that would be 27. So now I ride by dead-reckoning. Meaning that if I need to go a certain amount of miles I can reckon if I need to take a charger with me or not.

The point being is that if you have an accurate knowledge of your energy usage, you can account for an aging battery and judge your distances.

Or something like that. Anyway watt meters can be useful if the thief doesn't get it. :wink:

the hassle part would be installing it between battery and controller. But I think my Lyen controller possibly has a plug in for this. I may want to ask him.
Is it possibly to have a watts meter or analyst plug to controller instead of between battery and controller? I know that voltage range can be changed on my controller through a plug in.