48v 20Ah Ping battery dropping from 19Ah to 17Ah in days

[Frakentrike]

Hi guys,
My brand new 48v 20Ah Ping LiFePo v5 battery gave me a nice solid 19Ah of battery life on the first day I did a full commute on it but it has been steadily dropping after only about 4 full cycles (4 days of commuting) to the point where it yesterday only delivered 17Ah barely getting me home from work. All measurements taken with my new CA v3. (I have the 40A Grinfineon controller)

Last night I plugged the battery in to charge and it stopped charging a lot earlier than usual despite my going for a short night ride in addition to the usual commute. I plugged it in again this morning to top it up, but the charger fan didn't start up - normally after a night's resting the charger will always push a little more current into the pack before switching off.

The motor is a 1KW hub motor which has been drawing up to 35A at full load and pulling up to 1.5KW according to the CA thanks to the battery having the high rate Ping BMS which is rated to 40A continuous with peaks up to 60A.

I'm wondering if the 48v-12v DC-DC converter I hooked up to drive my lighting is pulling too many Amps as it is rated for 20A max. It is a 240W DC/DC Step-DOWN Power Converter Voltage Regulator. It is only driving a 20W LED headlight, LED tailight and Aurora LED flagpole though so I wouldn't have thought this should be an issue? This regulator has a No-load input current of ≤10mA and I'd have the lights on usually for much less than half the commute.

I've been very careful to follow a gentle breaking in period for the battery only going for short rides to start with and allowing the BMS to fully balance all cells overnight on several occasions (and not leaving the battery on charge after that) before I went for my first full commute.

I'll have to try it out with the DC-DC converter disconnected to find out the real Ah I get from the battery as the CA doesn't measure the amps used by the DC-DC converter.

Anyone think I should be worried about my battery pack? (says he nervously!) :|

 

[dustNbone]

You could connect the DC-DC Converter up to the shunt output so that it's measured by the CA. That would give you a more accurate measurement. One thing I recommend is to measure the cell voltages before you charge next time and see if there's one markedly lower than the rest. It could be the BMS hitting LVC (Low Voltage Cutoff) on one weak cell or cell group.

 

[Frakentrike]

You could connect the DC-DC Converter up to the shunt output so that it's measured by the CA. That would give you a more accurate measurement.

I was wondering about that but not sure how to do it with the CA v3 connected via the DP connection to the controller rather than using an external shunt. The DC power output from the CA itself only supports a max of 0.5A which might not be enough to drive my lights and flagpole (and airhorn which I am thinking of adding!). However, perhaps this is a good thing if I want to limit the drain of the DC-DC converter. I might just buy a 48v horn instead to avoid trying to put the high amp draw of that device thru the CA.

One thing I recommend is to measure the cell voltages before you charge next time and see if there's one markedly lower than the rest. It could be the BMS hitting LVC (Low Voltage Cutoff) on one weak cell or cell group.

Sounds sensible. I'll see if I can do so on the Ping without pulling apart insulation.

 

[dogman dan]

Its out of balance. Hopefully not for any permanent reason, like a cell leaked. If it stinks, there is your problem.

Sometimes it can be helpful to force a balance by charging full then ride around the block, charge full again. repeat 5 times or so. This restarts the charger so the cells that need a bit more can get some, while the bms continues to run discharge on the too full cells when it fills each time.

The purpose of the ride around the block is simply to lower the voltage of the whole pack enough to get the charger to restart.

In the past, ping stuff was set up to restart itself and continue balancing if just left plugged in overnight. The charger would turn back on for 30 seconds or so multiple times in the night if the pack is out of balance.

 

[amberwolf]

I was wondering about that but not sure how to do it with the CA v3 connected via the DP connection to the controller rather than using an external shunt.

Unless you open up the controller and solder the negative wire of the DC-DC to the FET side of the shunt (best method), you'd have to tap into the controller-side (cuz it's probably thicker/easier to splice) wire on the CA connector of the controller, for the Shunt + wire.

That way the power provided to the DC-DC comes off the side of the shunt being measured by the CA.

You can also use this same tap point to hook the negative charger wire to to be able to use the CA to monitor charging as well as discharging.

 

[Frakentrike]

Hi guys, is it dangerous not to charge replacement cells up to full charge before connecting the battery pack back together?

Li Ping kindly shipped out 4 replacement cells for my pack at no charge and a skilled friend very kindly de-soldered the old cell group which included one cell that had visibly swelled up, but in the rush we didn't get a chance to charge the cell group up to 3.7v with the little singe cell charger that Ping included with the replacement cells, before closing up the battery package and heat shrinking the new shinkwrap around the battery.

I’m now worried about plugging it all together again with the BMS in case the out of balance cells cause damage or reduce the lifespan of the pack. Would it be safe for me to slowly charge the pack for short periods of time and let the BMS balance the cells or should I cut off the new shrink wrap (drats!) and charge that cell group up separately with the single cell charger as per Ping's instructions?

The replacement cells had a charge of about 3.4v (if I recall correctly) in each of them as per good practice for shipping and I'd previously charged up the entire 2x16s pack to the full 57-59v (3.7v per cell I assume) before pulling it apart so there is a bit of a voltage differential there.

Sorry for my noob questions!

 

[dogman dan]

I think I misunderstood in your other thread. I thought I read something about parallel connection.

If you just have a series connected pack that is unbalanced, plug it in and let the bms work. If you have two parallel packs, they need to be close to the same charge before you connect in parallel.

I'm curious about the puffed cell,, was it the end one? did it look like it got damaged by the shipping? That might be what happened, if the end cell puffed.

I had a ping pack do that, not new. It took a big hit in a crash, and flew off the bike landing on one end. Protected in a box, so no apparent damage. But a year later, that end of the pack puffed up bad.

40 amps controller is still asking a lot from that size ping. I'd strongly suggest setting the amps limit on the CA to 30.

 

[Frakentrike]

I think I misunderstood in your other thread. I thought I read something about parallel connection.
If you just have a series connected pack that is unbalanced, plug it in and let the bms work. If you have two parallel packs, they need to be close to the same charge before you connect in parallel.

You are correct Dan, my 20Ah Ping battery is composed of two separate packs. The cells in each pack are in series, but the two packs are designed to plug together in parallel when fully assembled. I talked with Li Ping who indicated the replacement cells were charged to 80-90% full for transit so were close enough to the voltage of the rest of the pack to be safe to connect everything back together and let the BMS do its work. Fingers crossed it all works without reducing the life of my pack!

I'm curious about the puffed cell,, was it the end one? did it look like it got damaged by the shipping? That might be what happened, if the end cell puffed.

Interestingly enough it was cell number 23 which is 9 cells from the end of the slave pack so no, not an end cell. You can actually see the puffed cell in the photo I attached to my other post (sorry about the double posting - I thought I'd post separate thread to see if I could get more feedback).

40 amps controller is still asking a lot from that size ping. I'd strongly suggest setting the amps limit on the CA to 30.

I paid the extra for the high rate BMS with the battery pack which Li Ping describes thusly:

"Suitable Motor Controller: up to 60 Amps (High Rate BMS), up to 35 Amps (Regular BMS)
Applications: Electric Bikes, Electric Scooters, Electric Trikes, Electric Pedicabs
Voltage: 48 Volts (51.2 Volts Nominal)
Capacity: 20 Amp Hours
Charging Voltage: 59-60 Volts
Charging Current: <8 Amps (<15 Amps with High Rate BMS)
Rated Discharging Amperage: 20 Amps
Max Continuous Discharging Amperage: 40 Amps (60 Amps with High Rate BMS)
Maximum Discharging Current (Pulse): 60 Amps (1 second with Regular BMS), 100 Amps (1 second with High Rate BMS)"

I've actually also installed the 3-way switch available from Grin Tech connected to my CA v3 and set up 250w, 1000w and unlimited (1500w) presets. I only ever switch over to the 1500w setting if I have a steep hill to climb and a heavy load as the reduction in range is too annoying to run continuously at that power setting. As such, that means most of the time I am only drawing about 30 amps so hopefully my battery shouldn't get too stressed.

 

[999zip999]

A 40 amp controller on flat land or hills and how much weight ? 40amp controller are hard on a 20ah ping. Riding style.

 

[Frakentrike]

After running the repaired battery for several partial cycles, I had the chance today to run the battery flat and it delivered an impressive 20.8 Ah, up significantly from 15.7Ah (and that was with my powerful lights and animated LED flag pole being used a bit and not being metered by my CA).

That increased my range from 36.6km to 46.8km which is excellent - thanks to Li Ping for the prompt delivery of replacement cells and my friends Paul and Doug for the expertise to install the replacement cells.

 

[ColinB]

Frank, I think you are being pretty hard on your battery. I know many people consider them as consumables, so using them up goes with the territory. That said, I'm Mr. conservative, and in this for maximum usage. So I suggest:

- Limit power to less than 1c, which is 20a or 960w. "Rated Discharging Amperage: 20 Amps"
- Use the 80% rule for everything. Consider your battery empty when it hits 80% capacity, or voltage hits 44.8v.
- Watch for the "sweet spot" where adding more power doesn't accelerate the bike any faster. At that point you are just making heat. Back off and save big.

What type of riding do you do? Maybe we can help you get your bike to 100km range?

Colin

 

[999zip999]

I found the older cells with 20ah pack live well at 28amps. 40amps things go puff. Plus a set of sense wires for cell meter help with having a bms is very handy.

 

[Frakentrike]

I think you are being pretty hard on your battery. I know many people consider them as consumables, so using them up goes with the territory. That said, I'm Mr. conservative, and in this for maximum usage. So I suggest:
- Limit power to less than 1c, which is 20a or 960w. "Rated Discharging Amperage: 20 Amps"
- Use the 80% rule for everything. Consider your battery empty when it hits 80% capacity, or voltage hits 44.8v.
- Watch for the "sweet spot" where adding more power doesn't accelerate the bike any faster. At that point you are just making heat. Back off and save big.Colin

Thanks for your suggestions Colin, 80% capacity is indeed quite conservative(!), not sure if I can be quite that disciplined! My daily commute is 26km to work and back so most of the time I would run the battery down to about 60% so that's not too bad. I do in fact already have the power restricted using the CA v3 to 1kw or 250w maximum the vast majority of the time, the former of which is pretty close to your recommendation of " 1c, which is 20a or 960w." I find there is a fair bit of difference in perceived torque for hill climbing between the 1kw and 1.5kw setting on my CA so hopefully I'm not just adding heat, but I will definitely only continue to use the "unlimited" setting on the CA rarely off-road and with caution.

What type of riding do you do? Maybe we can help you get your bike to 100km range?

My vehicle is a homebuilt full-suspension tilting recumbent trike meaning it's a fair bit heavier than a regular mountain bike but 100km range might be possible when running at the 250w setting. However, that's okay as I use this beastie as my daily commuter vehicle and for off-road riding.

 

[dogman dan]

Now there is two votes to limit your controller to 30 amps, from guys with thousands of miles of experience with pings battery.

I know you don't pull 40 amps continuous,, Its the spikes. Pings hate a big spike like that starting up.

 

[999zip999]

And going up hills with 40amps.

 

[Frakentrike]

Now there is two votes to limit your controller to 30 amps, from guys with thousands of miles of experience with pings battery.
I know you don't pull 40 amps continuous,, Its the spikes. Pings hate a big spike like that starting up.

I do appreciate the feedback and advice guys.

The good news is I specifically checked out the maximum current draw shown on my Cycle Analyst yesterday when going up hill from a standing start with the "unlimited" power setting (which maxed out at 1,590w) and the current never went above 31 amps. On the 1kw limit setting, the maximum amps was around 22 amps uphill if I recall correctly, so looks like I'm never getting near that theoretical top limit of 40-60 amps mentioned in Ping's specs.

Does that sound ok to you?

 

[dogman dan]

Yes,, that's real world, what it actually is seeing. 30 amps from 20 ah of ping cells is more like what my experience with pings says is ok. 1.5c is fine.

I significantly lowered the lifespan of a ping, by hitting it with lots of 2c spikes. It immediately lost some capacity, and it ended up with two weak cell groups for the rest of its 3 year lifespan.

What a single cell can do in a lab, and what an assembly of cells can do in the rough world of riding around on a bike, is two different things. Cut stated c rates in half has become my general rule. Ping says 3c,, give it 1.5c max, and the battery will be happier.

I did not say the ping cannot do 3c. I said give it 1.5c and it will be happier. It will stay better balanced, and have less possibility of swelling cells and dying.

Look how fast you swelled those first cells. Not entirely your fault, but clearly those cells did not like it. The point is, all your cells in a pack of many cells will never be perfect. You will have stronger and weaker ones in the assembly. Your limit is the weakest cell.

 

[Frakentrike]

Yes,, that's real world, what it actually is seeing. 30 amps from 20 ah of ping cells is more like what my experience with pings says is ok. 1.5c is fine.

That's good to know Dan.

I significantly lowered the lifespan of a ping, by hitting it with lots of 2c spikes. It immediately lost some capacity, and it ended up with two weak cell groups for the rest of its 3 year lifespan. What a single cell can do in a lab, and what an assembly of cells can do in the rough world of riding around on a bike, is two different things. Cut stated c rates in half has become my general rule. Ping says 3c,, give it 1.5c max, and the battery will be happier. I did not say the ping cannot do 3c. I said give it 1.5c and it will be happier. It will stay better balanced, and have less possibility of swelling cells and dying.

Understood. Makes a lot of sense.

Look how fast you swelled those first cells. Not entirely your fault, but clearly those cells did not like it. The point is, all your cells in a pack of many cells will never be perfect. You will have stronger and weaker ones in the assembly. Your limit is the weakest cell.

Thankfully only one cell swelled, the other 3 in the problem group all tested fine which is a relief. I don't think I've ever seen the max amps go higher than that 31 amps I measured on the unlimited setting - my mention of 40A was purely reading off the specs of the controller and battery. The very first full cycle ride I took with the new battery (after 2-3 short rides to gently ease the battery in) only gave 19Ah and the next full cycle a couple of rides later it had dropped to 17Ah and then 15.7Ah, so I'm hoping the problem cell was just bad to start with and not due to excessive amp draw. The battery is certainly performing beautifully now delivering almost 21 Ah when I did a full cycle 2 days ago so I have no complaints.

I will limit the Cycle Analyst to never supply more than 30 amps though just in case of unexpected spikes as per everyone's advice. Thanks again for this info. It greatly helps out a battery newbie like me. I'm learning very quickly thanks to the wealth of experience on ES.

-Martin

 

[999zip999]

Now that it delivers 21ah just use 19ah.

 

[Frakentrike]

Now that it delivers 21ah just use 19ah.

Yep, makes sense. The vast majority of times I commute, I'll only be using up to 12ah so should be good.