Karolis
10 W
I think it is only good for recycle. The contacts were cut for not to use those cells any more.
Terminating the 15Ah and 20Ah Cell-Man cells
- Thread starter cell_man
- Start date
cell_man
100 kW
As was mentioned I'm kind of sold out of 20Ah cells. That's not strictly true, I have a customer that has already purchased a big quantity who is due to take most of the remaining 20Ah cells. With what's left I could probably do some smaller packs but not sure I'd have enough for a bigger pack. Situation on the 15Ah cells is similar but have a few more 15Ah cells to spare. There's no news on more batches of cells at the moment.
oatnet
1 MW
I spent a few hours trying different methods of wrapping foil around dowels inside the clips. I came up with a great solution, then realized that the tabs were not long enough to execute it. Finally, I worked out that just wrapping the two tabs around a quarter-inch dowel, from opposite sides, with the balance wire shimming them in place, was the ticket. Someone else suggested that configuration earlier, and I think it popped into a lot of our minds.
Armed with a solution, I tried to apply it to the two cells I connected earlier. I discovered an interesting thing - I had to really bear down to undo just the first fold on the two tabs. These things are bonded together really well. The uneven crimping from pliers really helped too, so I will continue to use that instead a vice as I planned earlier.
So I refoldeed the tabs, put the clip on, and shimmed it with the 1/4" dowel. With the dowel in place, the tabs aren't going anywhere. I was worried about the metal of the tabs slicing through the end of the heatshrink, so I cut a longer clip and dowel, and 4 really short pieces of 1/16" dowel. 2 short pieces slide in against the end of the tab, both sides, keeping it away from heatshrink, eliminating the risk. A quick layer of heatshrink, and fast, easy, reliable connection. I tried pretty hard to pull it off, and I think the tabs will tear off long before the clip fails. I'm gonna do the whole pack like this.
So to summarize, a contact patch thicker than the tabs is established by a clip. Heat shrink holds the clip in place to maintain the contact patch, retain the clip, and assume the compressive load after the plastic in the clip fatigues. Folding the foil keeps the tabs from sliding out of the jaws of the clip, crimping them supplies additional contact patch and mechanically bonds the tabs together. Dowels keep the tabs from unfolding, compress the folds for even more contact, and keep the tabs from coming in contact with/cutting the heatshrink.
Simple, cheap, effective, and easy to service.
-JD
Repost of what the tabs looked like before the dowel:
And Now:
View attachment dowels_9733.JPG
View attachment heatshrink_9735.JPG
Armed with a solution, I tried to apply it to the two cells I connected earlier. I discovered an interesting thing - I had to really bear down to undo just the first fold on the two tabs. These things are bonded together really well. The uneven crimping from pliers really helped too, so I will continue to use that instead a vice as I planned earlier.
So I refoldeed the tabs, put the clip on, and shimmed it with the 1/4" dowel. With the dowel in place, the tabs aren't going anywhere. I was worried about the metal of the tabs slicing through the end of the heatshrink, so I cut a longer clip and dowel, and 4 really short pieces of 1/16" dowel. 2 short pieces slide in against the end of the tab, both sides, keeping it away from heatshrink, eliminating the risk. A quick layer of heatshrink, and fast, easy, reliable connection. I tried pretty hard to pull it off, and I think the tabs will tear off long before the clip fails. I'm gonna do the whole pack like this.
So to summarize, a contact patch thicker than the tabs is established by a clip. Heat shrink holds the clip in place to maintain the contact patch, retain the clip, and assume the compressive load after the plastic in the clip fatigues. Folding the foil keeps the tabs from sliding out of the jaws of the clip, crimping them supplies additional contact patch and mechanically bonds the tabs together. Dowels keep the tabs from unfolding, compress the folds for even more contact, and keep the tabs from coming in contact with/cutting the heatshrink.
Simple, cheap, effective, and easy to service.
-JD
Repost of what the tabs looked like before the dowel:
And Now:
View attachment dowels_9733.JPG
View attachment heatshrink_9735.JPG
jonescg
100 MW
cell_man said:
Hope you can get more stock in mate. They clearly sold like hotcakes
oatnet
1 MW
Melting the heatshrink on apparently melted the plastic clip. Overall the join stayed in place, but the clip was not solid enough for me. I removed the old clip/heatshrink/dowels, and tried a new method to cover the existing 'folded' tabs.
1) Apply heat shrink to folded tabs. Just enough so that you can mould it shut at the top of the tabs, without leaving a long tail that won't fit into the plastic clip.
2) Double over the balance wire, so it still emerges at the sides of the cells instead of the center. This helps hold the clip in place, provides a little extra length in case the balance wire gets tugged on, or cut too short.
3) Slide the clip over the folded balance wire and tabs. The heatshrink and wire jackets really grip the clip, so minor force is needed.
4) Wrap the clip in a few twirls of fiberglass-reinforced tape to keep the clip shut.
-JD
View attachment b1_9774.JPG
View attachment b2_9775.JPG
View attachment b3_9776.JPG
1) Apply heat shrink to folded tabs. Just enough so that you can mould it shut at the top of the tabs, without leaving a long tail that won't fit into the plastic clip.
2) Double over the balance wire, so it still emerges at the sides of the cells instead of the center. This helps hold the clip in place, provides a little extra length in case the balance wire gets tugged on, or cut too short.
3) Slide the clip over the folded balance wire and tabs. The heatshrink and wire jackets really grip the clip, so minor force is needed.
4) Wrap the clip in a few twirls of fiberglass-reinforced tape to keep the clip shut.
-JD
View attachment b1_9774.JPG
View attachment b2_9775.JPG
View attachment b3_9776.JPG
oatnet
1 MW
I found some time to build an 8s/24v pack from the 16ah cells, and I'll make a second one later to series into the 16s/48v pack for my BMC V3 bike. I used simple painters tape to bind each cell to the pack temporarily, I have some giant shrinkwrap I bought from ebikes.ca a few years back I'll use on them when I get everything sorted.
The system works quite nicely when bonding tab-to-tab, but I am not happy with tab-to-8ga for my discharge terminations.
The system works quite nicely when bonding tab-to-tab, but I am not happy with tab-to-8ga for my discharge terminations.
cell_man
100 kW
Hi JD,
if you don't have much luck with getting a good connection onto the cables you can always try soldering. The negative tab solders very easily with standard solder, it's just the positive that is a bit tricky, but still do able even with standard solder if you've got a big iron and take care to sink the tabs to avoid excessive heating of the cells. Luke linked to a special solder post specifically for LiPo cells that should make it much easier. I must get myself some of that stuff.
Edit:
here's the stuff Luke recommended: http://www.solder-it.com/shop/item.aspx?itemid=89
if you don't have much luck with getting a good connection onto the cables you can always try soldering. The negative tab solders very easily with standard solder, it's just the positive that is a bit tricky, but still do able even with standard solder if you've got a big iron and take care to sink the tabs to avoid excessive heating of the cells. Luke linked to a special solder post specifically for LiPo cells that should make it much easier. I must get myself some of that stuff.
Edit:
here's the stuff Luke recommended: http://www.solder-it.com/shop/item.aspx?itemid=89
regmeister
100 W
- Joined
- May 9, 2009
- Messages
- 263
deleted
cell_man
100 kW
I believe it's for a BMC V3 1000W setup so max 50A, probably a bit less.
oatnet
1 MW
cell_man said:
Yep, this will replace the a123 M1 pack on my BMC V3, where the discharge is only 50a. I might parallel the packs and test them in my VW bus conversion to see how it feels with an extra 30v, and in the future I might use these as part of a 150v pack on a 5305, although I might just build a pack specifically for that project. I am confident that the folded tabs will handle 15c without issue, I have 10x more contact area than I need. The soft metal of the tabs is melded so tight I'm not sure if they will actually come apart.
I saw Lukes post on the low-temp paste he just ordered a case of, and I'm looking forward to seeing how that sorts out. I'm going to see if I can wick some solder into the rolled end of the tabs with the discharge leads, just a little to hold it in place will do, just gotta heat up the 100w iron and get a wet sponge ready for the cool down.
Once people get used to seeing these cells in use, you will not be able to keep them in stock. Once I used the Headway 38120s in my VW here, lots of people started buying them.
-JD
cell_man
100 kW
To avoid excessive heat going into the cells I clamp copper or aluminium sheet on the insulated section of the tab where it goes into the body of the cell. This will sink away the heat and avoid any heat going into the cells so you can hit the tab with as much heat as you want. You'll probably need to rub the alloy positive tab with a stainless steel brush first. It can be tinned with normal solder but it's not easy and requires the tab to be heated with a big strong iron and rubbed until the solder takes. Better to get the proper solder as Luke suggested and use the technique outlined on the website.
You'll know if you've got good terminations by how hot they get under load. If running a constant 50A a good connection doesn't produce any heat. As soon as you get some attenuation, say 10mA or so it will quickly get pretty hot. Best to do this on the bench of course, if you can rig up a suitable load. If you could get a DMM probe onto the tabs before and after the join you can actually measure the voltage drop at each termination. I personally find spending a few mins polishing all the tabs with Autosol before terminating or testing helps to reduce any attenuation.
You'll know if you've got good terminations by how hot they get under load. If running a constant 50A a good connection doesn't produce any heat. As soon as you get some attenuation, say 10mA or so it will quickly get pretty hot. Best to do this on the bench of course, if you can rig up a suitable load. If you could get a DMM probe onto the tabs before and after the join you can actually measure the voltage drop at each termination. I personally find spending a few mins polishing all the tabs with Autosol before terminating or testing helps to reduce any attenuation.
michaelplogue
10 kW
I understand what you're trying to do with that newspaper, but it will also act as a pretty good insulator - keeping the heat in...........
cell_man
100 kW
Trust me you need to pull A LOT of current to get anything hot if you have good terminations. If there's a poor connection between cells generating heat then you would need to improve that connection. The cells themselves do not get hot until you are pulling very serious constant current levels.
When I first started tested these cells I noticed that the alloy tab and copper bus bar clamped to the tab would get really hot. It was only much later that the penny dropped. I now know that the alloy tab generally has a little more loss to the copper terminal that the plated copper terminal (can be drastically improved if you take the following steps). I also now realise that polishing the tabs and the terminals is very worthwhile and lastly it needs a reasonable clamping force across a reasonable but not necessarily very big area of the tabs. I was only clamping at the edge of the tab and the insulator I was clamping to was completely inadequate so force was not across the tab. The oxidised surface also doesn't help (much more noticeable effect on the alloy tab than the copper tab, but with the right preparation the alloy tab can have just as low losses). When you are trying to pull 400A out of a cell just a very small voltage drop creates serious power and heating.
When I first started tested these cells I noticed that the alloy tab and copper bus bar clamped to the tab would get really hot. It was only much later that the penny dropped. I now know that the alloy tab generally has a little more loss to the copper terminal that the plated copper terminal (can be drastically improved if you take the following steps). I also now realise that polishing the tabs and the terminals is very worthwhile and lastly it needs a reasonable clamping force across a reasonable but not necessarily very big area of the tabs. I was only clamping at the edge of the tab and the insulator I was clamping to was completely inadequate so force was not across the tab. The oxidised surface also doesn't help (much more noticeable effect on the alloy tab than the copper tab, but with the right preparation the alloy tab can have just as low losses). When you are trying to pull 400A out of a cell just a very small voltage drop creates serious power and heating.
cell_man
100 kW
Hey Chris,
here's some pics of a couple of boxes you might be interested in
View attachment 1
View attachment CIMG0691.JPG
These are 4S 3P packs and there will be 8 in total. There are 0.3mm sheets between each cell and these sheets are thermally coupled to the outer case to help cool the cells as they will be working pretty hard. There will be 8 packs in total, so 32S, 3P making a 105.4V nominal, 60Ah pack. They were a lot more work than I had ever imagined. The brass positive and negative terminals will be used to link the packs with bus bars or very big cables as required.
here's some pics of a couple of boxes you might be interested in
View attachment 1
View attachment CIMG0691.JPG
These are 4S 3P packs and there will be 8 in total. There are 0.3mm sheets between each cell and these sheets are thermally coupled to the outer case to help cool the cells as they will be working pretty hard. There will be 8 packs in total, so 32S, 3P making a 105.4V nominal, 60Ah pack. They were a lot more work than I had ever imagined. The brass positive and negative terminals will be used to link the packs with bus bars or very big cables as required.
liveforphysics
100 TW
Beautiful packs!
cell_man
100 kW
Thanks Luke,
still on a steep learning curve with the whole thing but hopefully getting there. I've spent a lot of time on Chris's packs and they could be better (hindsight is a great thing) but I did them at a loss not including the assembly and countless days designing and fretting over the whole thing, so hopefully he'll be happy how they're turning out. Most people really do not realise just how much work goes into these things. I've probably gone way overboard on the terminations for these and the cooling is likely completely unnecessary but you don't know these things until you've got the whole thing up and running. Hopefully some day I can get over to Oz to see the bike they'll be fitted into and maybe even feel what dual Agnis running 100V feels like 8)
still on a steep learning curve with the whole thing but hopefully getting there. I've spent a lot of time on Chris's packs and they could be better (hindsight is a great thing) but I did them at a loss not including the assembly and countless days designing and fretting over the whole thing, so hopefully he'll be happy how they're turning out. Most people really do not realise just how much work goes into these things. I've probably gone way overboard on the terminations for these and the cooling is likely completely unnecessary but you don't know these things until you've got the whole thing up and running. Hopefully some day I can get over to Oz to see the bike they'll be fitted into and maybe even feel what dual Agnis running 100V feels like 8)
jonescg
100 MW
They look fantastic Paul!
I dare say I'll be pulling maximum amps from these guys, so maybe the cooling was well worth it. Just need to fine a way of mounting Rod's BMS units, either externally or on the packs somehow. Otherwise, I had another idea for packaging, but they certainly wouldn't be waterproof. I guess the next e-moto will have 4P or 5 P, so the currents will be lower still
They ended up being 175 wide and 245 long, and 95 high?
I'll hopefully be track testing before I head to Canada, then I'll need to leave the bike in someone's capable hands while I'm gone.
Well done mate, I really appreciate the work you've put into this!
I dare say I'll be pulling maximum amps from these guys, so maybe the cooling was well worth it. Just need to fine a way of mounting Rod's BMS units, either externally or on the packs somehow. Otherwise, I had another idea for packaging, but they certainly wouldn't be waterproof. I guess the next e-moto will have 4P or 5 P, so the currents will be lower still
They ended up being 175 wide and 245 long, and 95 high?
I'll hopefully be track testing before I head to Canada, then I'll need to leave the bike in someone's capable hands while I'm gone.
Well done mate, I really appreciate the work you've put into this!
cell_man
100 kW
Including the brass terminals which extend out from the front panel they are 250mm long. They might be slightly higher than 95mm high as the screws are a bit proud of the boxes and there is no room to countersink them into the top and bottom as they are positioned very close to the edge. They are 175mm wide. I'll get these 2 shipped today so you can have a play with them within a few days and see exactly what's what.
As you can see the battery terminals are on the left of the box, with the negative at the bottom, positive at the top. If you want some with terminals on the other side so they can be more easily connected in the bike, let me know soon. Hopefully get the remaining done within about a week.
As you can see the battery terminals are on the left of the box, with the negative at the bottom, positive at the top. If you want some with terminals on the other side so they can be more easily connected in the bike, let me know soon. Hopefully get the remaining done within about a week.
jonescg
100 MW
Hmm, since the sprocket is on the left hand side it might be wise to have the terminals on the right. No drama - we can flip them over. Also, the ones under the tank will be a bit funny since there will be two packs facing eachother. We'll see how it looks next week anyway. I have to source some angle ally for the trays etc.
Attachments
cell_man
100 kW
I thought you were going to have 2 facing the other direction? If that's the case, then it might be easier to terminate if you have 2 of the 8 packs assembled with terminals on the opposite side from the other 6. You'll need to allow some space for the termination onto the terminals as well, so this might add some additional length on that side. It's no extra work to assemble the pack in either configuration but it's a big job to change it once it's built.
oatnet
1 MW
Nice job on the packs! That is the way I'd want them built on a larger vehicle/higher amps. jonescg, nice build, and these packs are a great choice.
-JD
That is the way I'd want them built on a larger vehicle/higher amps. jonescg, nice build, and these packs are a great choice.-JD
jonescg
100 MW
cell_man said:
I have a drawing on my coffee table at home explaining myself a bit better :lol: I'll try to get it uploaded this evening. There will be two packs with the terminals facing forwards - the one in the under-seat triangle and the one closest your googlies under the tank. All others will be facing backwards, and these will be towards the front of the bike. I'll draw another pic
cell_man
100 kW
Well that's what I remember and to avoid having the connections going across the packs for those 2 facing the opposite direction I thought it would be better to have 2 of the packs with terminals on the other side. Just let me know how you want each of the packs to be made up and I'll do that.
jonescg
100 MW
If my stero-vision is correct, you're making them exactly how they need to be. In order to have the termini on the left hand side, we can simply flip them all over and make the positive go on the bottom. When it comes to joining the foward facing packs, we'll add them as numbers 6, 7 and 8 in the series. No worries!
