Alternative to Thunder Sky LCP

[granada203028]

With the increased cost and difficulty in obtaining the Thunder Sky LCPs I have been looking at alternatives. Other Chinese LiFePO4 being the obvious one but also a parallel approach of polymer cells. The poly cells perform very well but are still expensive. The links show pictures of a 100Ah cell using 10 Li poly cells in parallel, with charge/discharge performance. Temperature rise was 7, 18 and 33 deg C for C/5, C/2 and C discharge rates. Temperature actually fell approx 4 deg on charge for the first 2 hours of the 6 hour charge.

http://i246.photobucket.com/albums/gg11 ... yCell2.jpg
http://i246.photobucket.com/albums/gg11 ... Charge.jpg
http://i246.photobucket.com/albums/gg11 ... harge2.jpg

Total weight is just over 2Kg giving 185Wh/Kg - usefully better than the LCPs and twice as good as LiFePO4 alternatives.

I'm considering converting a 100 - 125cc motorbike. A 30V x 200Ah, 6KWh/6KW battery should provide 60mph and 60 mile range. The LiFePO4 suffers a big penalty of size and particularly weight. The Li polys are PL9059156 from:
http://www.batteryspace.com

Total pack price is likely to approach £3000. The 10 cost me approx £220 using my employer to forward them and avoiding VAT and duty.

I have asked the supplier to comment on the parallel approach. Also they don't spec the end of charge condition.

[milo0105]

Li-POLYMER battery has the soft cover ,while LiFePO4 battery has hard cover .Compared with our LiFePO4 battery ,Li-polymer has many disadvantages:
1) Soft cover is easily to be destroyed thus get short-circuited .
2)Li-polymer is not safe .It may not be easy to explode but easy to get fire .
3)Li-polymer has short cycle life and short continuous mileage.

[retepsnikrep]

Nice job Grandad. I too liked the idea of lighter weight cells but the cost was enormous. I like very much your neat casing and interconnection of the smaller cells. Looks vey smart, interconnections on those flapy terminals was always a worry especially for the bigger Kokam type cells.

How long did it take you to make up one of those 100ah cells, can we have some more pics of how that is assembled and the attachments to the bus bars.

The more members on here who investigate different battery options the better please do your own thing and report back with all the details.

[Jeremy]

As I understand it, the Thundersky "cells" are just plastic boxes that contain several pouch-type cells in parallel, albeit with some pressure/temperature equalising fluid added around the pouches.

As long as the risks are understood and managed appropriately then I can't see any real problem with this approach. A good BMS would be key to ensuring safety, as would robust cell packaging.

There are many tens, or hundreds, of millions of LiPo cells in everyday use, in everything from mobile 'phones to laptops. Fires are exceedingly rare events now, and only seem to result from extreme abuse.

For an EV that doesn't need to draw extreme discharge rates, say no more than about 5C, then I honestly can't see a problem with using suitably mounted and protected LiPo cells like this.

It may well be possible to buy suitable cells for a better price direct from China, rather than vis batteryspace, though. There are several RC model battery suppliers in China that sell direct to the UK, for example.

Jeremy

[granada203028]

I have proceeded with the project and bought 200 for approx £3K + £1K to get them here. Re - safety they were air freighted without comment and come with a comprehensive data sheet. This includes destructive tests of over charge/discharge, puncturing etc guaranteeing no fire. Some of these Li poly cells are UL listed. I accept though a large pack would need more care and of course I'm not going to take any risks with my investment! It was seeing the dismantled Thunder Sky cells and the difficulty in now getting them that gave me the idea of paralleling li poly. Testla take massive parallelism one step further with their 7000 18650s.

I have bought a Honda H100 for the conversion. Maximum discharge rate will only be 1.5C and about 1C continuous. I'm familiar with power and analogue electronics so the BMS will be a fully ground up design. Not particularly difficult but a fare acreage of strip board required! The main problem I see is a KWh gauge to ensure I know I have enough range for any journey and use the cells at middling charge state to maximise life.

The connections are quite fiddly, particularly soldering the 22 swg copper wires to the copper bus bar. I plan to build my 20 "bricks" like this and have ordered some professionally bent plastic.

http://i246.photobucket.com/albums/gg11 ... lyCon1.jpg
http://i246.photobucket.com/albums/gg11 ... lyCon2.jpg
http://i246.photobucket.com/albums/gg11 ... olies1.jpg
http://i246.photobucket.com/albums/gg11 ... olyCon.jpg
http://i246.photobucket.com/albums/gg11 ... olyOne.jpg

[retepsnikrep]

This looks very encouraging good luck with it. Good luck with your ground up BMS as well the more the merrier, I look forward to seeing how you go about the whole thing. Please keep us updated with the details. Peter

[Jeremy]

The LiFePO4 pouch cell pack that I have on my electric recumbent uses a faily simple connection method that you might be able to adapt. The cells are paralleled using a strip of PCB material with slots milled into into it; the slots are spaced to align with the tabs coming from the top of the cells. The cell tabs are fed though the slots from the insulated side of the PCB, then folded over and soldered.

One way of adapting this technique for higher current connections would be to cut slots into thicker copper strips. The slots can be cut from one edge with a hacksaw or a thin slitting saw on a CNC mill might give a better result.

Here's a picture of the top of my pack. It's connected as a 12S 2P pack, with 24 cells. The interconnecting PCBs are the solder blobs in the photo, sorry it's not clearer. The black stuff is thin neoprene strips that insulate and provide some support for the cell tabs from vibration etc. The white tabs are bits of plastic insulation between cell groups.




Jeremy