Adding balance wires to Ping 48v battery

[zelectric]

Hi all,

I am considering the purchase of my second 48v 10ah battery from mr. Ping. My fist battery lasted for two years of commuting and I got about 4000 miles out of it. As with most of these batteries one bad cell group had the battery cutting out in just minutes of use. With the help of mr ping and the many threads here I went through all the trouble shooting by opening up the shrink wrap and load testing the cells to determine with group was causing the problem. I ordered a few cells and attempted to perform the repair. As it turned out some other tabs got torn on 'good' cell groups and I managed to short out 2 channels on the BMS so the repaired battery didn't last very long for me. Rather than throw any more $$ at the problem I decided to donate the battery locally for someone to repair.

I believe that if I had known the status of each cell group and was able to balance charge the pack it might still be on my bike today. This brings me to my question about adding balance wires to a ping pack. Mr Ping has stated that they can add the balance wires to the pack during assembly but asked for a diagram. I am not sure how to create such a diagram to sen to my ping. A 48v ping pack is 16s so would be best to have him install two 8s balance leads?? I may have a few more questions about this but I wanted to get some opinions on what and how to do this.

Thank you for your help in advance,

Todd

 

[dnmun]

ping installs the balance leads when he makes the battery. it will come with the balance leads attached and they run to the BMS which balances the pack.

 

[zelectric]

Thanks dnmun,
I understand how the pack comes as I had it for 2 years and opened it up in an attempt to repair it. I am asking about adding additional balance wires like the ones on all the lipo packs. I was thinking then I can monitor the cell groups better with a cellog8 or something similar.

ping installs the balance leads when he makes the battery. it will come with the balance leads attached and they run to the BMS which balances the pack.

 

[dnmun]

adding cellogs will cause the pack to become unbalanced. they do nothing to make it balance any better than the BMS. it might be best to leave it alone. the ping BMS will balance it better than you can do manually.

 

[zelectric]

Thanks again dnmun!!
So you are saying there is no way to monitor each cell group without opening the pack or unhooking the bms everytime? The ping pack is plug and play until a cell group goes bad then its either replace said group or replace the entire pack?? I am just trying to prevent what happened to my first pack. It sounds like there is no solution.

adding cellogs will cause the pack to become unbalanced. they do nothing to make it balance any better than the BMS. it might be best to leave it alone. the ping BMS will balance it better than you can do manually.

 

[dnmun]

i have been thinking about this very idea. how to use a monitoring device we can buy cheap that would create a data file associated in time, or in cycles, with various cell parameters such as internal resistance and aamount of heat generated by a given amount of current flowing through it over time.

it is very difficult to integrate a sampling system into the normal operation for the motor/battery as the agent of this prospective idea of an "ageing"calculator.

if you read the thread on the cellog hacking that richard did at the end Otmar added the depletion mode mosfets so i thought it would help to do that for the cellogs in order to get the drain uniform across all 8 cells. this would allow you to leave the cellogs on all the time without consuming so much of the ping BMS balancing current.

the cellog is using about 15mA and the ping has 60mA of shunt current so the cellog is 25% of the balancing current. so to use the cellog to monitor the pack constantly the 8S source for the cellog circuit current kinda has to be done.

then use the recording facilities of the cellog 8S to record the voltage of the individual cell as it is charging, and the data could be recorded in time. then to estimate the internal resistance you would take that data and divide it by the current that the time the cellog is recording voltage you could generate a graph of the changing voltage/current relationship. then see if we can find some point int he data where the cell suddenly or slowly changes and see if we can use that as an indicator of cell health.

it seems the simplest way to record enuff data to start figuring out what matters in terms of cell health and longevity. my first idea of the hardest thing is using the pack down to low voltage with a large current flowing through it.

 

[dogman dan]

8s balance plugs spliced as a Y into the leads to the bms is what you want. Then you can BRIEFLY plug in cellogs or other monitoring devices to check cell levels as you desire. It will also allow a convenient place to tap in to single cell charge a low cell if you wish to do that. It can be fun to watch the bms work, by having a cellog in place as the pack finishes charging. I was quite amazed at how high a charge the highest cells were getting in the process.

It would also be possible to hack an adapter harness, that would plug into the existing bms plugs. That harness would need a clip on cell one -, since that lead is the main - to the bms.

He is correct, you don't want to leave stock voltage monitors like cellogs plugged into the pack for long periods. Even just regularly using them while discharging tends to unbalance the pack. But you won't need that. It will be nice though, to easily find out which cell is always the one that tripped the bms. Which cell is not holding a charge so good, etc. It may not make your next ping last one day longer, but it does help your peace of mind to easily check what's going on. Knowing which cell group is weakest can help you plan on when to fix it, or whatever you plan on doing about it.

I'm not anti bms. But I've come to believe in checking on what a bms is doing from time to time. Some 8s plugs make that quick and easy.

 

[migueralliart]

It is good to use a cell log every once in a while to check the cells are actually balanced appropriately by the BMS. This way if a fully charged pack shows you any cell that is pretty out of voltage you know that the cell might be damaged or the BMS is not doing its job well.

You can then charge that cell up to the voltage of the other cells and check the voltage after a ride. If the voltage stays even with the other cells you know the BMS might be playing tricks on you.

 

[dnmun]

and some cells just never charge up to full voltage because the leak down rate is high enuff that the balancing current cannot supply enuff current so they go back to the 3.35V-3.45V level and never get to the full charge under the small amount of balancing current that the ping supplies.

but the 8S cellog hack is not hard, just requires steady hands and a deft touch with the soldering iron. then the current is spread over all 8 cells.

 

[dogman dan]

Maybe you should offer hacked ones for sale? Sounds good for those that want to run one and log the graphs.

But even being able to just briefly use a cellog to quickly check 16 cells sure beats fiddling with a DVM and tiny plugs, working your way down the row. So if you don't have a hacked cellog, or have some other cheapie cell checker, it sure is convenient to have those 8s jst plugs on the pack.

 

[dnmun]

i agree it is useful to have the cellogs. i use them too. but i was observing that if he leaves them on the ping pack that the cellogs will consume 25% of the balancing current that is needed to balance the pack and it will make the pack more unbalanced.

i would never make stuff for sale. no money in it and no time to do it. maybe someone who is younger and who has stable hands and can solder could do it but not me. i can hardly solder anything now because my hands shake so much.

but the point is that there is nothing he can do that will improve over that balancing that the BMS does already. even if he knew what the voltage is, because the voltage is not an accurate indication of the state of charge. so when he says it is a bad cell because it is only charging to 3.42V that doesn't mean anything except that he has the voltage set too low on the charger so the low cells cannot ever get enuff juice during balancing. but people always turn the charger voltage down so the pack does not balance and then don't understand why.

 

[dogman dan]

True, no money in doing that mod, especially if you can't do it easy or quick. I agree, you don't want to run a permanent unmodified cellog, or other monitor, simultaneous with a bms.

He could have done nothing wrong, his pack is getting old, and acting just like mine did as it kicked the bucket. I had cells that would charge to full charge, but drop to 3.3v instantly when the charger was taken off. Capacity of those cell groups was down more than 30%. Likely his is doing the same.

If I understand right, he just wants to be able to check on his new pack easier when it comes.

 

[zelectric]

If I understand right, he just wants to be able to check on his new pack easier when it comes.

The original plan was to trouble shoot the pack but much easier than using the muliti-meter. I also thought that using a good balance charger like the ones you guys use with lipo would balance the pack better than the bms but I guess that is not the case.

I never intended to leave the cellog connected longer than it would take to determine which cell group may be going bad.

At this point if I were to order another Ping pack I think I would have him add the balance wires just to be able 1. trouble shoot the pack easily if needed and 2. be able to single cell charge very easily if needed.

So back to one of my original questions. Ping has requested a diagram of said balance wires. Can anyone assist with this diagram?

I really appreciate all the input on this subject.

I guess I was just bummed my ping pack didn't last longer than it did and was looking for ways to increase the longevity of a new pack if I were to purchase another.

Thanks again,

Todd
Zelectric

 

[dnmun]

if this is gonna be a 16S ping pack then you can add the 9 pin plugs like i did to this 24S bestechpower BMS. they come with 300mm leads and i solder them right onto the base of the pins that mount the sense wire plug on the pcb. it would be easy with the ping signalab. you don't have to make ping do it but you could get him to have his people solder them for you i bet. it is not too hard to do, but not easy to keep the leads from shorting when the solder wants to walk over to the adjacent pin. i would add a small dip of solder to each pin first and then add a little bit to the end of the wire, then hold the wire right on the edge of the pin, sometimes the solder would short and you have to just pull it away with the tip of the soldering iron. but you have to double check to make sure they are not shorted before plugging the sense wires back in.

the 8S hack is not hard either. i was gonna order some of those depletion mode mosfets because i do wanna be able to keep the cellogs 8S attached permanently during the charge cycle to monitor the battery pack health which i feel can be done more easily during the charge cycle. that way the data collected on each cell during each charge cycle could be recorded along with other parameters like charging current to allow interpretation later by analyzing the data.

but i can see the imbalance even with the big D131 BMS i used for this since the top two cells #7 and #8 are always the high cells and the D131 has almost 200mA of balancing current. more than three times as much as the v2.5 signalab.

this is what i use for the cellogs: http://www.ebay.com/itm/JST-2-5-XH-9-Pin-Connector-Plug-w-Wire-x-5-sets-/111093380099?pt=LH_DefaultDomain_0&hash=item19ddae7403