Ping 2.0 Reports

[dnmun]

i think charging at up to 1C is feasible, and i think you can look at the BMS and maybe replace a few parts to allow it to handle higher currrent during the charge, but i would watch the heat because the shunts would get hot, hot. might have to build up the traces on the PCB like they do for the power fets on the controllers.

the discharge rate is .1C for the capacity ratings that the manufacturer uses. so 2C and 3C discharges reduce the effective capacity, even though it is not the classic peukert effect, it acts similarly, it appears.

 

[nomad85]

i think charging at up to 1C is feasible

The batteries can charge at a higher rate, but the bms can only handle 8 amps charging rate. I use a 5 amp charger and it works wonderfully. I have a ping 1.0 so maybe the new bms can handle a higher rate of charge.

 

[ZapPat]

The one part on the BMS that would directly limit charging current is the single FET, since the FETs on my BMS are 24 milli-ohms, which is pretty darn crappy! I've already changed all the 4 other FETs, and this has cut down on the BMS heating while discharging a lot! Also, I don't think the shunt is a limitation here, since I seem to recall that it is not used by the charge parts of the circuit.

 

[dogman dan]

I was sorta thinking along the lines of an opportunity charger, that you would never really leave charging for more than the time you spent in a store or restaruaunt, and then use it bypassing the bms. Just a 20 amp power supply. As long as you did full charging at home later with the bms, why not?

 

[swbluto]

I'll post more when I get my logger back from Eagle Tree. I'll do a complete discharge test down to LVC, and will post the discharge graph and the pack's measured capacity.

*still waiting*

 

[ZapPat]

Well swbluto, as it happens I just got my eagle tree logger v3 back from RMA on friday, and have just ran my first full discharge of the 48V 20Ah ping pack. I rigged up some ni-chrome (heating element wire) as a load, and this resulted in a 4.5 dropping to 4 amp discharge current. It's too low a discharge rate I know, so I will rig up a bigger test load for the next run at 20 amps or so.

Before showing you the discharge graph, I want to point out a couple things about this particular Ping battery. Although it has only 20 cycles or so at 1.5C average, I have to admit I have put it through some abusive handling... Namely, there was first of all dropping the positive wire and watching it fall onto the exposed negative connection of the battery and breaker (the wrong side of course)... BAM! The wire welded itself to the connection bolt, and wouldn't pull loose for 2 or 3 seconds, sparks flying as I yanked away at it. Finaly it came off, and there was thick smoke eminating from the BMS by then! Oh oh! Had to replace the four discharge FETs, and it's now better than new with much lower resistance (less heat during high discharge). Then came accident #2 just the next day - I guess I was very distracted because I did two bad things while charging the pack: I used the regular discharge wire for charging, and also forgot to put the voltage feedback resistor back onto the DC power supply I use to charge. What was the result of this supidity? Overcharging the pack up to 72V (60.5V normaly!) and without any balancing or high voltage cutout functions! Ouch again!!! I thought to myself if this battery is even 75% good after all this I'll be lucky! And as we see here it looks pretty darn good at 19.5Ah capacity. That's like 95% of advertised capacity still remaining!



I'll post the individual cell voltages read at the end of discharge at the same time as I add the higher current discharge graph, in a day or two.

All I can say is I'm happy with what this Ping battery can take and still deliver the goods!

 

[swbluto]

So it sounds like over-current for a few seconds and overcharging. 19.5 AH sounds like a good amount of left and its nice to know they can stand such abuse with grace!

(That's actually 97.5% of the rated capacity)

I'm looking forward to your testing at 100-200 cycles. :wink:

 

[Microbatman]

ZapPat

Thanks for the test data

60 volts down to 45 under a 5 amp load.

These batteries are not that good. According to the data from your eagle tree logger (nice piece of test equipment!!)

Here is my test data.
The test data is from 1 of my 6cell packs

The green line is volts. I can post another chart with amps but wanted to make it as simple as possible to understand.
Box to the left shows Max and Min numbers.

Setup is 18s2p

Each 18 is divided into
three subbanks of 6cell packs
3x6=18

Data in the test is taken from 1 of those 20volt/6cell modules.
Had to do it that way due to voltage limit of the DPR-50
It is only good to 50 volts and did not want to blow it.

Anyways this pack was sujected to 33 amps!!!

And only dropped 2 volts under that load.

 

[swbluto]

ZapPat

Thanks for the test data

60 volts down to 45 under a 5 amp load.

These batteries are not that good.

I will post my data of A123 at 60 volts and everone can be the judge.
Nice to finally see some hard data on this.

Can you elaborate? You probably mentioned something earlier in this thread that I forgot. :lol: Sorry about that.

About the A123s. I'm just interested in cycle-life data for high discharge rates. A123systems has a nice cycle-life chart with a lifetime of ~7000 cycles with 85% capacity left(And it seemed that the capacity decrease was linear. Maybe it's different with LiFePO4 than SLA?).... BUT that's at a mere 1C, not the 10C that their cells are famous for.

 

[Microbatman]

Same data

Magnified with current

Blue=Amps
Green=Volts

 

[geoff57]

hi all
microbatman don't do a Don Harmon on us we all know that the ping batteries can't compaire with A123 cells no one has ever tryed to ping batteries are a good battery that does what it says it will do, it has a place in the LiFe market as comuter batteries,you just stay within the safe limits and it will give you a long and rewarding life. A123 are of the high discharge type like small R/C lithium batteries and Lifebatt from don. The price reflects this.
If you have a bike that you've tured into nothing short of a trials bike and use as a hill climer then A123's are needed no question, a lot of us out there are commuters trying to replace their car with an electric bike they don't want to mak a hole in their bank account the size of the grand canyon for people like this ping batteries are ideal.
geoff

 

[Microbatman]

The last thing I want to do is pull a Don Harmon.

If high C rates and low cost are not a priority then A123's may not be the ticket.

LIFePO4 may not even be the ticket.

Nickel-metal hydride also becomes a candidate. Some comparison test here may would be nice.
One Nickel-metal alternative would be Nilar. They are one of the lowest cost alternative to PHEV prius conversion kits.
But you are still looking at $500 plus.
Another alternative is individual 10Mah cells around $8cell 40 cells would be $320.

All I was trying to do is to provide some sort of data to compare it to.

Fechter says is best that "1 test is worth 1000 opinions".

 

[ZapPat]

60 volts down to 45 under a 5 amp load.

These batteries are not that good. According to the data from your eagle tree logger (nice piece of test equipment!!)

Microbatman,

All lithium-phosphors have a similar discharge curve, always with a long flat region ending with a fast drop-off. Ping's BMS cut off as soon as the first cell hit 2.7V as would any well-protected LiFePO4, thus explaining the voltage difference between fresh off the charger (~60V) and the end of discharge (~45V.). This has NOTHING to do with a battery being good or not - it's just the chemistry's characteristic!

Of course we wouldn't expect a Ping battery to be on par with A123 cells - Ping's completely assembled and tested battery (including BMS) costs about 1/3rd the price of your hard to find bare A123 cells of the same capacity! Now, I calculate Ping's cells to have about 3 to 4 times more DC internal resistance than equivalent capacity A123 cells - sounds bad you might think? Well, think of it this way:

I would rather buy 60Ah from Ping than 20Ah of A123 cells for the same money, since what I want mostly is better range, and don't want the hastle of assembling my pack myself for the moment. So for the same cost, I'll get a battery with similar internal resistance (meaning similar voltage drop either way), but by going Ping instead of A123's I get three times more capacity! And 60Ah worth of Pings can easily dish out 120A continous, and 180A intermitant - good enough for many applications.

The only time I might use A123's (at present prices) is if I would be making some powerfull vehicle that I will be completely discharging in 15 minutes of use. Not my type of ebike for the moment (but does sound fun!).

What do you think of this?

 

[Microbatman]

I agree that if range is what you are looking for don't overspend to get high C rated batteries.

What would be nice would be to see your graph with just the volts displayed.

As far as LiFePO4. There is quite a difference in performance/discharge curves amoung different manfactures of LiFePO4 chemistry. Seen it on RcGroups and other sights. There is plenty of data to support this.

I too am interested in finding a low cost LiFePO4 battery keep up the testing and info well appreciated.

If the priorities are cost and range and discarge around 4C
then we may not want to rule out Nickel Metal Hydride.

Im not trying to pick any fights or disrespect any product. Saying not that good is disrespecting a product. My apologies.

 

[Pete]

All lithium-phosphors have a similar discharge curve, always with a long flat region ending with a fast drop-off. Ping's BMS cut off as soon as the first cell hit 2.7V as would any well-protected LiFePO4, thus explaining the voltage difference between fresh off the charger (~60V) and the end of discharge (~45V.). This has NOTHING to do with a battery being good or not - it's just the chemistry's characteristic!

Of course we wouldn't expect a Ping battery to be on par with A123 cells - Ping's completely assembled and tested battery (including BMS) costs about 1/3rd the price of your hard to find bare A123 cells of the same capacity! Now, I calculate Ping's cells to have about 3 to 4 times more DC internal resistance than equivalent capacity A123 cells - sounds bad you might think? Well, think of it this way:

I would rather buy 60Ah from Ping than 20Ah of A123 cells for the same money, since what I want mostly is better range, and don't want the hastle of assembling my pack myself for the moment. So for the same cost, I'll get a battery with similar internal resistance (meaning similar voltage drop either way), but by going Ping instead of A123's I get three times more capacity! And 60Ah worth of Pings can easily dish out 120A continous, and 180A intermitant - good enough for many applications.

The only time I might use A123's (at present prices) is if I would be making some powerfull vehicle that I will be completely discharging in 15 minutes of use. Not my type of ebike for the moment (but does sound fun!).

What do you think of this?

I think you are blurring the lines a little between AH capacity and C rating. Yes, a Ping and an A123 are both LiFePO chemistry with the same broadly similar discharge characteristics ie very flat (within it's C rating) , but the internal mechanical construction affects the voltage behaviour under heavy load. As I see it, LiFePO's also seem to have a Perkut effect, some worse than others.

Your example is a good one though. But it also depends on how much extra weight you are prepared to tolerate. If you only need 18 AH for your usage, why carry the weight of the extra 40AH of lithium storage ?

I really dont want to get Safe started on this thread , but there is more than one variable that helps to define what is the MOST appropriate battery setup for each rider. And for the record, I ride a 24v20AH ping battery

 

[dnmun]

if you want lifepo4, why not join the headway cell group buy. the 48V20Ah pack was $630, and the shipping was gonna be surface to toronto to save costs. you may want to go over tot he lifepo4 headway group buy yahoo group. i don't know if they have shiped yet or not, but it appears the newest headway cells are much superior to the previous batches, and headway has been very generous in helping when people have problems. even when they have received defective cells, they were replaced quickly. might check and see if you can join it before they ship.

 

[ZapPat]

Here's the results of a 0.9C discharge of a Ping battery, the same one as in my last post. I've included only the V vs time graph this time, changed time to minutes, and zoomed in on only the 40 to 55V range for more precision on the discharge curve. I also redid the first 0.22C discharge graph in the same way (included after) for comparison.

*EDIT*: Just added the full voltage range graphs, since some people like to see the overall discharge picture as well as another with the details of the voltage drop during discharge.

[Abused] Ping 48V 20Ah, 0.9C discharge - 19.2Ah capacity
View attachment 1
View attachment 4


[Abused] Ping 48V 20Ah, 0.22C discharge - 19.45Ah capacity

View attachment 3


So not much difference in capacity from 0.22C to 0.9C, (0.25Ah) - So far so good.

Next will be a 1.5C - 2C test... but I have to boost my test load first, cause this one gets a wee bit hot!
View attachment Discharge test - 1kW load (small).jpg

BTW for anyone wanting this data, I have the eagle tree data files saved away.
Ciao!

 

[nicobie]

Thanks for doing the testing!

Nick

 

[dnmun]

you could use several of those radiant electric heaters where the nichrome wire is wrapped around the ceramic cone in the center. they don't have fan so would work ok for DC.

can you scale down your eagletree logger jpg so the scales and full elapsed duration all shows on the page? i think you are the first documented discharge of a ping battery, this shoulda happened 6 months ago when the first ping was purchased on ebay. thanks a lot.

 

[ZapPat]

you could use several of those radiant electric heaters where the nichrome wire is wrapped around the ceramic cone in the center. they don't have fan so would work ok for DC.

can you scale down your eagletree logger jpg so the scales and full elapsed duration all shows on the page? i think you are the first documented discharge of a ping battery, this shoulda happened 6 months ago when the first ping was purchased on ebay. thanks a lot.

I added the full graphs in the post above, so you have both full voltage range of battery, then a close up of the voltage drop.

Full times are shown until battery hits LVC (which is first cell down to 2.7V).

 

[dogman dan]

Does anyone know, straight from Ping himself, the c rate of the new 4 ah pouches? I've heard 3c and 4 c.

Perhaps Zap Pat's excellent tests will show the reason why the recomended discharge rate is still 1c, if the cells c rate has now gone from 2c to 4 c. If the new cells are really 4c, why not a recomended discharge of 2c?

 

[ZapPat]

Does anyone know, straight from Ping himself, the c rate of the new 4 ah pouches? I've heard 3c and 4 c.

Perhaps Zap Pat's excellent tests will show the reason why the recomended discharge rate is still 1c, if the cells c rate has now gone from 2c to 4 c. If the new cells are really 4c, why not a recomended discharge of 2c?

This is an email I got from Ping before I bought my battery:

We used to sell packs based on 3.2v4950mah cells. But now they are out of
stock and we are selling pack based on 3.2v4ah cells. The 4ah cell has
better 2C performance than the 5ah. They can discharge more than 98%
capacity at 2C continuous rate. Internal resistance is less than 15mohm. The
cell is rated 2C max continuous discharge rate and 4C peak. 1C max charge
rate. More than 85% capacity after 1500 cycles.

Note that during my constant 0.9C discharge test the battery got up to 40oC by the end, BUT it was sitting inside a foam padded box (insulated), with only the top exposed. Ambient (room) temperature during the test was about 20oC, but finished at around 24oC or so due to the 1000W test load heating the room.

 

[Kurt]

I have been looking forward to some hard data on the ping packs for some time now. I think when people are spending several hundred dollar's on a battery they really need to know if the battery is going to meet there needs. Until now all people could do is ask around for opinions and hope for the best .Now we are starting to get some hard data people can make an informed decision.

It is the lack of this information that has held me back on buying a ping pack to date.So if the results turn out good it might even boost sales for him.

keep up the good work,

kurt.

 

[Microbatman]

ZapPat

Thanks for doing the testing. Great work!!

Yes would like to see the data files. If you have a chance post them or send them in a PM.

Great idea on using wire as a resistance source.

Is that nichrome wire you used?

 

[dogman dan]

Thanks for the info on the c rate, it sorta answers my question. I had suspected that a continuous 4c rate was not a good Idea. But it's darn good to know that 2c continuous is not expected to harm the battery all that much. I still consider all the duct tape batteries to be 1c technology if you want the max life out of em. Apparently Ping does too, or he'd change the specs for recomended discharge amps on them.

Pings batteries rock in my opinion, the only way I would want to trade mine, is for 48v instead of 36. But I have about 1500 miles on mine, about 120 cycles, and it's balanced perfect. I don't know how many ah I pull out of it, but I do know I haven't lost an inch of range since it was new. I haven't seen the bms do any balancing since June. Mine is v 1.0 cells.