Typical 'Duct Tape' performance ? / Need more power

Trying to get a feel for what performance I should be able to expect from a Ping battery. I don't know the specifics of what the BMS is and is not capable of. I am running a 24v20Ah pack on a 650W Cyclone. Electrical stats are taken from a Cycle Analyst.

Off the charger the pack seems to stabilise at around 29.5V. As soon as I pull any sort of load the voltage dips to around 25v. For the first 9Ah or so the battery is great, I have accidentally pulled up to 800w but much more than that will trip the BMS. No idea what the voltage is when the BMS trips as I loose the CA at the same time.

After about 9Ah I try to keep my draw to less than 450w and even at that level my voltage dips below 23.3 and starts to bring on the amber caution light on the Cyclone's throttle. If I back off the throttle the voltage picks up to above 23.8

Deeper than 14Ah and I need to be very careful about current draw. Much above 350w will cause the BMS to trip. Again, not sure what the pack voltage is at this time due to loosing the CA. The CA has logged min volts at 18 odd, but I wonder if that is just the transient during the power down.

Is this normal behaviour for LiFePO's, am I flogging my pack excessivley ? I am aiming to not draw more than 10Ah, partly because it is annoying having to monitor my watts but mainly to get the maximum number of cycles out of it. My main trigger for this is I got dusted by an infernal combustion bicycle when I was near home and already used 14 ish AH's. Bloody annoying to go for max power and have all the lights go out :? . Need more power... Cyclone says the controller will take 36v, so I am now torn.. do I get a pair of 36v20A's. Or a second 24v20 and run it in series and risk the controller. Fets are IRFP064v with a Vgss of 60v. And on top of this, if I am going to have a major power storage upheaval, do I wait for Ping's Mk2 pack?

To be honest I briefly considered going the infernal combustion route but I just can't bring myself to listen the the damn thing. Other option I am seriously considering is a 405 on 48v. I got the Cyclone specifically for hill climbing, but 60kph with your butt 3" off the ground is so much fun. I have had a taste of real speed and I want more . I don't HAVE to climb the couple of big hills around here, I just do it for the down hill run afterwards :wink:

Hmm, that doesn't sound normal. One of the big draws of LiFe is its super stable discharge curve. I've never run my 48V 20Ah very low, but it doesn't seem to want to sag below 47V under a 40A load. I have no BMS on mine, BTW.

But, then again, these aren't the best of batteries. They don't like more than 1.5C very much. :?

I am sort of abusive towards my ping packs and have never experienced anything close to what you describe. I have a 72v 20Ah setup and my nominal voltage is around 80 and i never drop below 76 even when I'm pulling 40-60 amps out of it. Im actually not sure exactly what my numbers are since I don't have a CA, but i can guess pretty well off speed and the lvc on my controller is set to 70, so in theory it would drop the draw somewhere around 76 and Ive not felt any sudden slow downs while accelerating up a hill or anything. Its possible that you got a slightly inferior brick of battery's or it could just be the BMS. I don't discharge with a bms so I cant be sure.

It could also be a bad cell, now that I think about it. After a hard ride, feel for any warm spots. Get voltage readings off individual cell groups if you can, too.

It sounds like you may have a wire disconnected inside or something. My ping is a bit hotter at the first couple miles of the ride, and then settles in at about 1 mph slower than when I leave the house. That stays the same pretty much till the pack is dead for me. It may be that at 24 volts, things are noticeable to you than are not to me at 36 volts, and toward the end of the ride, the hot motor may pull more current.

Finally got off my ass and fitted Andersons to the pack. Up till now it was hard wired to a terminal board inside my case.

I can't see any obvious breaks on the individual cell tabs.

No obvious differences in cell temperature after pulling 4Ah , but it is so cold here that it takes a while before I even register the warmpth of the BMS. Am I feeling for a couple of degree's, or 10 to 20 (C) ?

I will see if I can come up with a safe way to get cell group voltages under load... more to follow

Edit:
Voltages for cell groups when drawing 650 to 750w. Pack discharged 7Ah before measuring. Rest voltages between 3.3 and 3.328 after a few minutes
3.15
3.032
3.024
3.044
3.036
3.052
3.06
3.024

Damn near impossible to get steady state conditions so I don't hold much faith in the deviation. I plan to redo the measurements on a decent hill tomorrow, after pulling 14 ish Ah. I am beginning to suspect a broken tab burried in the pack someplace. But it probably won't show up till I get close to 90%DoD on the 15Ah group.

The sun came out for a bit, so new figures from running up my test track. 700w +/- 30, steady state for about 30 seconds for each measurement. Pack would have started out at around 5Ah discharged, and last measurement would have been about 8 Ah discharged. Same cell order as the previous measurements but the top figure was the last measurement.

2.91
2.81
2.956
3.0
2.936
2.99
3.030
3.0

It wouldn't have been much, but a warm cell would have been noticeably warmer if that was the problem.

Hmm, that 2.8V cell could be it, but it's really not deviating enough to give any definite answer, methinks.

Is this a real Ping pack or something from Anna or NewJourney? Supposedly, Ping and they use different cells, so that could be it...

It is a legit Ping pack. I might have to find a couple of driving lights to run the pack down to LVC gently then do another set of voltage measurements

Maybe you just need more volts, or a different controller. Some setups just draw more power than any prismatic cell can give, so maybe you are just flogging the batt. Back on the old duct tape thread, most of the folks with a problem had the 10 ah size. The 20 ah size should be giving you enough amps, but more than 20 amps draw continuous would be hard on it according to my understanding of the specs for mine.

dogman said:

Maybe you just need more volts, or a different controller. Some setups just draw more power than any prismatic cell can give, so maybe you are just flogging the batt. Back on the old duct tape thread, most of the folks with a problem had the 10 ah size. The 20 ah size should be giving you enough amps, but more than 20 amps draw continuous would be hard on it according to my understanding of the specs for mine.

Click to expand...

Need and designed seem to be quite different in this case. I am running the 650w Cyclone so it has it's own, unique controller. How unique?, I am not sure but the Cyclone runs around the 3600rpm mark and the controller does not appear to have an input shunt at all. The throttle led's are setup for a 24 volt system but Cyclone says it can (should?) be run at 36v. The throttle could just be because it got pinched from the 24v product lineup and Cyclone didn't bother to change the voltage points.

Looking at the datasheet, aren't rated and max continuous supposed to be the same thing ? Hmm, I will take that on board and try a long run at a lower C rate to see if I am hitting the LiFePO equivalent of the Perkut effect . Dosen't seem right though since Ping makes a point of his cells being 2C and the BMS trips at 1.75C.

So the decision stands, do I get a second 24v20 and risk running at 48v considering the fets have a Vgss of 60v. Or do I retire the 24v20 in favour of one maybe two 36v20 packs.

Ok, pulled 16.3 Ah at no more than 1C. at around the 15Ah mark I couldnt pull more than about 200w continuous, without tripping the BMS.

At rest cell voltage in the same sequence as the other two ;
3.229
3.053
3.214
3.243
3.181
3.240
3.243
3.236

Is there enough deviation in the #2 group to be significant ?

The second cell (3.053) and fifth cell (3.181) are looking a little low. Not much, but may be worth a closer look.

Since they aren't completely destroyed (negative voltage) after drawing more than 15Ah, however, we can rule out broken tabs.

After looking at how small / thin the cell tabs are, I am not at all keen on messing with trying to replace cells. I agree that it must be a weak cell not a missing cell so I am pretty much stuck with living with the pack as is. I am not going to attempt to replace cells and it is not worth the shipping to send the pack back to Ping.

I'm no good at the math, so I'm not sure how many amps you are drawing when you use 200 watts, but I suspect that since you run 24 volts you are pulling a discharge rate closer to the limits of the cells that the same watts pulled from a 36 or 48 volt battery. Just because the specs say it should handle 1C is no garantee that it will actually like it in the real world. A few cells may just be unhappy with the discharge rate enough to trip the BMS. Maybe not completely drained, but unable to keep up with the rest of the pack. Another battery in parallel with a diode, or getting a higher voltage battery may be the only solution other than switching to a real wimpy controller. I love the Ping battery, but they definitely do run into bikes that will trip the bms. The other possibility could be that the bms trips earlier than it should. Might be worth a try to get a second one from Ping and seeing if that helps. Another thing worth trying would be charging with a different charger. I'ts harder to tell state of charge by voltage with Lifepo4. Maybe you just aren't getting as full a charge as you think? Mine shut off early one time for reasons unknown.

If U have a Wattmeter or the like, check the charge in and also max amps on your ride.
otherDoc

The lower Ah ping packs do not like hills. The BMS will prevent you from damaging the cells by getting aggressive about cutting out ... I have to pedal-assist on steeper hills with my 36V16Ah batt.

When the voltage goes down after some miles of hard use, simply blipping the the throttle full-on from a standing start will trip that BMS.

Interesting thread, and I admit, I haven't been on ES for quite some time.
My Ping pack is performing quite well, except that my throttle LED's are now going 'off' much earlier than when I first got the pack.
i.e., green goes out fairly quickly, (200m ?), and amber goes out under load before I get to work (about 6 k's). All lights usually come back on at rest, or when coasting. I've had no issue doing a 25k run, even with the red throttle LED being the only one lit, and ending voltage is still quite healthy. (checked with DMM, no Watts-Up connected)
I'm thinking that it may be related to our drop in temperature, as we are in the middle of winter here, and my pack is charged outside in the garage. Temps ATM are only around 6 C. I have noticed that when I park my trike in the foyer at work, that the battery seems to perform better on the way home, although it is more generally downhill, and usually have a tailwind!
Does anyone else have an opinion on whether LiFePo4 chemistry performance is degraded in cooler weather, or if I should be looking for a cell that is under performing?

Cold does degrade these a bit.

Once I wrapped mine in closed cell foam in side my ammo can, this problem abated ... it doesn't get extremely cold in the Pacific North-Wet.

Thanks,
I e-mailed Ping for his opinion, and he has confirmed the same thing, saying the performance will drop off in cooler weather, and that full performance should return in the warmer weather.
I decided I would/should mount my Watts Up, to see what the pack is actually doing. I did this yesterday, and noticed that my pack is sagging down to 34.5v!
No wonder my LED's are going out early!
Max current draw is 14.8 A, so confident that I'm not abusing the pack. (1.5C or less).
What did surprise me, was that my motor is drawing up to 560 watts (from a motor that was rated at 24v/200w). My WHr usage is around 9.5 Whr/k, which seems reasonable, seeing that I don't spare the horses on my commute! If I kept my speed down to around 30 kph, I'm sure that figure might improve, should try that tonight..........................

Freddyflatfoot said:

I decided I would/should mount my Watts Up, to see what the pack is actually doing. I did this yesterday, and noticed that my pack is sagging down to 34.5v!

Click to expand...

From a 36V pack? That's not bad at all. My old SLA one sagged down as low as 30V. And it was a 48V pack.

But then, that's the beauty of LiFePo4!
BTW, I tried to limit myself to 30kph riding to work today. Managed to drop my WHr usage to 7.5 WHr/k.

Time to resurect an old thread.

After many delays (Not Ping's fault) I finally have 4 new cells to swap out the slightly dodgy set. Two thumbs up to Ping for taking the effort to diagnose the problem over email and to send replacement cells at no cost to me.

Since we were stuffing around with international postage anyway, I ordered a 12v20A module to splice in with my soon to be repaired 24v20A pack. And obviously a 36v BMS + Charger. Pity the Aussie dollar died in the ass against the greenback. My 'upgrade' ended up costing almost the same as the original setup thanks to currency conversions.

More to come once I replace the cells and strap everything back together.

Ahah!
Just stumbled on this old thread, after my battery 'died', with two low cells, maybe that was my original problem?
I wonder.
I've e-mailed Ping, so I'll see what he says.

Dredging up a really REALLY old thread. Haven't been near ES in a couple of years and it is amazing to see how far things have come.

I ended up getting a 12V20A block from Ping and have spliced it together with the original 24v20a. I had another look at splitting the pack apart to find the weak cells, but I really didn't want to be messing around with those foil tabs. At least the pack works in its current state - who knows if I would ever get it back together without damaging the tabs.

The trike pretty much hasn't moved in 2 years thanks to life and work, but I took it for the first real run with the 36v pack. No smoke, which is a plus. I still need to protect the pack properly, so it was just a spin around the block.

The chain path is lousy with the mid-drive. It needs a lot of work to get the combination of mid drive and rear derailure to run reliably without dropping the chain off the Cyclone and the front chain path is really tight, sucking leg power. As much as I like being able to run the motor on it's own, I am probably going to move the motor drive up to the LH side of the front crank and mount the motor down on the front crossmember.

The other alternative is to use a Rohloff rear hub like the Greenspeed setup, but I can't bring myself to drop $1500 on a hub no matter how much it would simplify the build.