Lithium Polymer Battery Packs

[tom_pay]

Hi All,

A friend of mine has asked me to make up some rather large LiPo battery packs. I am familiar with LiPos, but I have only used pre-made packs, I have no clue how to make them from scratch. So I have a few questions.

The finished pack needs to be approximately 15Ah and operate at 35v. Because of the high capacity needed I thought that the LiPo cells would have to be connected in parallel and, because of the high voltage, in series. So there would be a few cells in parallel connected, in series, to other sets of parallel cells. Can this be done with LiPo cells?

Once the pack has been made, how does it have to be charged? Is it just a constant voltage until a certain voltage is reached?

Also, are there any special safety requirements for LiPos? From what Dr. Google has told me they seem pretty dangerous!

Thanks in advance,

Tom

 

[Diver300]

You can parallel LiPo cells, but you need to make sure that they are at the same voltage before you connect each set in parallel. Once connected in parallel you can regard them as a larger cell.

When you put LiPo cells in series, you should make sure that the charging rules apply to all of the cells, so you shouldn't assume that the total voltage is evenly spread between the cells.

You should only charge slowly (0.1 C or so) if the lowest cell voltage is less than 3.4 V. There is hardly any capacity below that voltage so charging up to 3.4 V per cell does not take long.

You must not charge any cell to more than 4.2 V. Now that can be where you need balancing devices that allow the cells with lower voltages to be charged when other cells reach 4.2 V

You should stop the charge once the voltage is at 4.2 V and the current drops to 0.05 C. However I don't think it would hurt to leave some cells at 4.2 V while the others hit full charge.

When I needed 3 Li-Ion cell in series, I gave each one a separate charger.

Having said that, many commercial products just stick the cells in series and hope for the best. I don't know what cell matching they do. I would guess that they make sure that they are all at the same state of charge at assembly.

 

[tom_pay]

Thanks for such a prompt and thorough reply Driver300, I have some new questions though.

You can parallel LiPo cells, but you need to make sure that they are at the same voltage before you connect each set in parallel. Once connected in parallel you can regard them as a larger cell.

Does this mean that the first time that I connect them I need to use a low-ish resistor to limit the transfer current, and then I can safely solder them together, without any resistors?

Having said that, many commercial products just stick the cells in series and hope for the best. I don't know what cell matching they do. I would guess that they make sure that they are all at the same state of charge at assembly.

The way in which this pack will be used is going to be 'seasonal' for a few weeks it will be used daily, and then left for a few months. Would it be OK to wire them up in series, charge them and after the trip put them on a proper balance charger?

If this is OK, would a voltage source of 4.2 x cell count, connected to the battery and then stop when the current drops to 0.05C work as a charger?

When doing calculations to determine the cell count, what cell voltage should I use? Looks like they go from 4.2 to 3.4v, so a medium of 3.8v?

Thank-you so, so much for all of your help, I really appreciate it.

 

[jpanhalt]

Here are three sources of information you may find useful:

https://www.powerstream.com/BatteryFAQ.html
**broken link removed**
**broken link removed**

The third link is from the second source, but I thought it was worthy to emphasize.

The configuration you are considering is often referred to by using S and P to indicate the number of cells in series and parallel. Thus, your pack would be called something like an S8P7 (i.e., 8x4.2= 33.6V; Px7 (assuming 2,200 mAH cells) = 15.4 AH). Most LiPo chargers I have seen charge the individual cells (or parallel cells) individually. That is, they are not charged in series. There is a connector brought out of the pack to allow doing that.

John

 

[tom_pay]

Thanks for those links they seem very good, but maybe a bit over my head!!!!

I think for the trip charger Ill use a power supply voltage of 4.2 * series cell count at a current of approx 1C, until the charge current falls to 0.05C. With no individual cell charging . Does this sound any good? At home Ill treat the pack to a proper balance charger!!

On the pack itself, what low voltage cut-off voltage should I use? The link reckons 3.0v but I think that 3.2v would be better, just to be safe, or would this reduce the pack's capacity?

Would 4.2v zener diode/resistor combinations be necessary, to limit the max voltage across each parallel cell set?

Also, a more fundamental problem, is there a place where you can buy large LiPo cells? I can only find pre-made packs.

Thanks for all the help.

Tom

 

[jpanhalt]

There have been significant improvements in LiPo batteries that allow faster charging. Link #1 recommends 0.2 to 0.7C. I still charge my 2200 mAH motor batteries at 0.8A (about 0.4C) and have had no problems. They are 3S1P configuration with spacers for heat dissipation. Older versions without spacers did tend to overheat according to the manufacturer. If you go to 1C, heat dissipation might be a problem. Newer cells may take the faster charge rates. I don't know, but I would certainly check it out with the manufacturer for assembled packs, not individual cells.

If you puff a bunch of cells in your pack of 56 or more, it will be expensive.

Link#1 sells single, unassembled cells. It has several in the 4000 to 5000 mAH (e.g., GM0845131) range at about $20 each in volumes of 100. There are many other hobby sources too. It is hard to recommend a source without knowing where you are.

John

 

[audioguru]

Lithium rechargeable batteries catch on fire (a very hot fire that is impossible to extinguish) when charged wrongly like you are planning.
You should not simply charge Lithium cells in series, each cell should be monitored separately by a balanced charger.

Nobody makes a 4.2V zener diode and a zener is not accurate enough nor sharp enough.

 

[tom_pay]

With more knowledge comes more questions, so here goes...

Would the cells heat up much? The pack will only draw about 5A (0.3C for 5000mAH cells) Would some cell spacers would be a good idea?

I would make a little circuit board to monitor pack voltage and if any cell had puffed and cut power if there was a problem. Though, what cut-off voltage should be used?

I think that designing and building a LiPo charger may be a bit too hard for me at my knowledge level. A 9 cell balance charger looks as though it is hard to get, though I have an 8 charger. Could I use 2 cell connectors and charge one group of 4S3P and another group of 5S3P cells one after another?

When I was looking for a source of LiPos there were a few warnings on soldering their tabs to form a pack, is soldering their tabs difficult or dangerous?

Lastly, I have found a source of Turnigy branded cells, has anyone used this brand before, are they any good?


Thank-you very much for all your help, Id be totally lost otherwise,

Tom

 

[jpanhalt]

I think for the trip charger Ill use a power supply voltage of 4.2 * series cell count at a current of approx 1C, until the charge current falls to 0.05C. With no individual cell charging . Does this sound any good? At home Ill treat the pack to a proper balance charger!!
Tom

Would the cells heat up much? The pack will only draw about 5A (0.3C for 5000mAH cells) Would some cell spacers would be a good idea?

I am getting confused. Are you planing on using 1C or 0.3C for charging? Second, you don't use voltage as a signal for end of charge with LiPo's. Please review the links I gave. LiPo fires are for real.

John

 

[tom_pay]

Charging probably only 1C or less, discharging only about 0.3C, taking into account the 3 cells in parallel (each cell having a capacity of 5000mAH).

Wasn't the charging considered complete when the charge current drops to 0.05C? But making sure that no cell went over 4.2V.

Tom

 

[audioguru]

Turnigy Li-Po cells are sold by Hobby-King for electric RC airplanes. They are Chinese extremely cheap (CHEEP cluck, cluck, cluck) and do not perform like better more expensive ones. My friend bought some and gave me one but I hardly ever use it. I use excellent ThunderPower Li-Po battery cells and I get a good price from the Canadian distributor.
ThunderPower Li-Po battery cells cost 4 times as much as Turnigy so you could save money by using Turnigy.

 

[tom_pay]

Hmm, that's not good.

Is there any other good brands out there?

Is the two stage charging a decent idea? Or should I stay away from that?

Thanks

Tom

 

[jpanhalt]

Charging probably only 1C or less, discharging only about 0.3C, taking into account the 3 cells in parallel (each cell having a capacity of 5000mAH).

Considering a 15AH battery pack at 35V, 1C charging is 525 watts (ExI). Battery University (**broken link removed**) states that charging at less then 0.8C is 97 to 99% efficient -- referring to the battery, not the charger's conversion efficiency. Higher charge rates probably have lower efficiency. If that is accurate, at 97% efficiency, 15.75 W of heat will be produced. How hot it gets will depend on how that heat is carried away. Unless you make provision for it, 16W can get pretty warm.

John

 

[tom_pay]

Ok, so maybe little pieces of circuit board wedged between the cells, without the usual heat-shrink wrap.

Im still concerned about charging the pack. I cannot seem to find a 9S charger easily, is the idea of having two 'sections', ie cells 1 through 4 on a different set of cell taps to cells 5 through 9 on another, and then charging each 'section' on an easy to find 8s, a good one?

Thanks

Tom

 

[audioguru]

Is there any other good brands out there?

People say that Hyperion Li-Po batteries are excellent.

Is the two stage charging a decent idea? Or should I stay away from that?

You should read about the safety things done in a battery charger IC, Maxim and other semiconductor manufacturers make them. The Battery University on the web also has a good tutorial.
The temperature of the battery is almost always measured and it controls the amount of charging current.

 

[tom_pay]

To get the right voltage and capacity I will have to connect the cells in series and parallel. My first thought was to connect three cells together in parallel, make up 9 of those and connect the 9 little packs together in series to form the pack. However I then thought that if I were to make up 3 sets of 9 cells in series and then connect the three strings together in parallel, this would give me the same voltage and capacity. I thought that the first method was better, but now I am not sure. Which way should they be connected?

Other than the low voltage cut off sensor, a temperature sensor and the 'puffing' sensor, is there any other sensors or circuit boards that should be incorporated into the pack?

Thanks so much for the help,

Tom

 

[Diver300]

To get the right voltage and capacity I will have to connect the cells in series and parallel. My first thought was to connect three cells together in parallel, make up 9 of those and connect the 9 little packs together in series to form the pack. However I then thought that if I were to make up 3 sets of 9 cells in series and then connect the three strings together in parallel, this would give me the same voltage and capacity. I thought that the first method was better, but now I am not sure. Which way should they be connected?

The first method is better, as you only have 9 voltages to worry about. If you have three separate parallel strings, there could be a voltage imbalance in any of them, so you would need to measure each cell voltage.

 

[tom_pay]

When the LiPos are about to explode and catch on fire, do they puff and get hot? Do they do anything else if something is going wrong?

Thanks everyone for your help

Tom

 

[bountyhunter]

Thanks for those links they seem very good, but maybe a bit over my head!!!!
Tom

I would respectfully recommend you not try to build high power batteries from Li-Poly cells. They are dangerous and have a habit of catching fire. The world has moved on to the safer Li-Fe chemistries. We developed some control IC's for Valence Technology (the originator of Li-Po) back in the early 90's, and they never found a solution for the problem. Steer clear of Li-Poly.

The video link above illustrates what can happen can happen to an overcharged Li-Poly battery pack. Pictures have been published in modeling magazines of the burnt remains of airplanes and helicopters whose owners did not properly recharge their batteries while the packs were still in the aircraft.

 

[bountyhunter]

Would 4.2v zener diode/resistor combinations be necessary, to limit the max voltage across each parallel cell set?

Li batteries require a charge voltage control with a tolerance better than about 0.5%. A zener will never do it, it takes a precision charger control IC.