Why did you have the trouble to make the holder yourself if there is a cheap solution allready? http://www.ebay.com/itm/10pc-Battery-Spacer-2x-18650-Radiating-Shell-EV-Battery-Pack-Plastic-Heat-Holder-/321114233041?pt=Power_Cables_Connectors&hash=item4ac3e658d1
Battery assembly kit for 18650 cells - first prototype
- Thread starter circuit
- Start date
Gregb
100 W
Has anyone tried their BMS??? $30.00 plus $4 postage !!!!! I spent 3 times that just buying the MB for the Zephyr mk3
Hugechainring
100 W
So, I've been following this off and on. I have a question about the fuse rating. It would seem that the fuse rating would need to be at or slightly above the maximum current rating of the cell used.
It seems to me that if you have a 12S6P pack full of 20Rs (example) and you are running them at 10A surge per cell (5C), your fuse is like 12A, and one of the cells has a fault or for some reason hits 12A and blows the fuse, the rest of the parallel cells in the bank have to carry the same load. However now they are one fuse short, only (5) 12A fuses left, increasing the current load on each remaining cell/fuse to 12A, cascading into a total failure of the whole parallel bank. Of course the possibility of this happening is higher with small parallel banks. If you lose a cell on a 12S20P battery the current increase in each remaining cell is much smaller.
Most of the time a fast acting fuse is sized above the maximum rated current of the circuit, or a time delay fuse at the current which will produce a continuous temperature rise in the conductors. I'm picturing this as a fast acting fuse. I'm trying to understand the benefit of fusing a 10C cell at 5C, or a 5C cell at 2C. It seems like you are inviting failure under normal operation for some users.
Am I missing something?
It seems to me that if you have a 12S6P pack full of 20Rs (example) and you are running them at 10A surge per cell (5C), your fuse is like 12A, and one of the cells has a fault or for some reason hits 12A and blows the fuse, the rest of the parallel cells in the bank have to carry the same load. However now they are one fuse short, only (5) 12A fuses left, increasing the current load on each remaining cell/fuse to 12A, cascading into a total failure of the whole parallel bank. Of course the possibility of this happening is higher with small parallel banks. If you lose a cell on a 12S20P battery the current increase in each remaining cell is much smaller.
Most of the time a fast acting fuse is sized above the maximum rated current of the circuit, or a time delay fuse at the current which will produce a continuous temperature rise in the conductors. I'm picturing this as a fast acting fuse. I'm trying to understand the benefit of fusing a 10C cell at 5C, or a 5C cell at 2C. It seems like you are inviting failure under normal operation for some users.
Am I missing something?
savaoaknyc
10 W
- Joined
- Feb 13, 2011
- Messages
- 69
Hi, circuit,
nice design.So you have tested already the d1 cell.Do you have any data about the temperature at 1.5c continius discharge???Probably i will choose this cell for 3p pack but need to now the temperatures at 1.5c continius.
About the insulator tape, the last year , while searching for some cells i have found a manufacturer of this insulator paper for the cell,the one posted by agnisium.It costs pennys for hundred pieces,if i find it again gonna post it in your thread.
Greets Sava.
nice design.So you have tested already the d1 cell.Do you have any data about the temperature at 1.5c continius discharge???Probably i will choose this cell for 3p pack but need to now the temperatures at 1.5c continius.
About the insulator tape, the last year , while searching for some cells i have found a manufacturer of this insulator paper for the cell,the one posted by agnisium.It costs pennys for hundred pieces,if i find it again gonna post it in your thread.
Greets Sava.
Yes, I believe you are. I have mentioned several times that fuse rating is not only adjustable by cutting half of it off, but also it is possible to make slight changes for higher current rating. Today's design is aiming for lower power cells, but it is very possible to make a change and offer this solution for high power packs.Hugechainring said:
savaoaknyc said:
Yes, I actually do have some data.
See below.
Temperature is not in this graphs, but I do have numbers in xls file. Temperature of cell at end of each discharge cycle is ~48°C at 1.5C discharge, 5min rest and 0.5C charge.
Note that this is for a single cell, lying on a bench. Stacked battery pack would be worse, especially in center of the pack.
BTW, sorry guys for taking so long... I had to change my job, some other changes, lots of stuff happening in my life these days...
Mammalian04
100 kW
Hi Circuit. I about to order my 29E cells in the next couple days for two 20s13p packs. I am still trying to figure out how I am supposed to wire it. Would I be correct in assuming that this project will not be ready in time for me to use in my pack builds?
Cheers,
Mammalian
Cheers,
Mammalian
999zip999
100 TW
Please do keep info in thread and going in or thru P.M.'s. Thanks
Mammalian04
100 kW
Testing and then production soon?
What are the final fuses rated for?
Still more suited for 29E or 29pf rather than 25r or VTC5?
Is your production method able to vary in dimension?
Looks great! Can't wait to hear how it does over the rough stuff.
What are the final fuses rated for?
Still more suited for 29E or 29pf rather than 25r or VTC5?
Is your production method able to vary in dimension?
Looks great! Can't wait to hear how it does over the rough stuff.
Awesome work !
Is their anyway you can make this work on low c crate cells with less then 10 amps draw per cell ? and use magnets to hold the buss bars on the cells instead of welding from this thread here:
http://endless-sphere.com/forums/viewtopic.php?f=14&t=60517&start=25
it would make it a lot easier for anyone to use this as getting the welds is the hard part.
Is their anyway you can make this work on low c crate cells with less then 10 amps draw per cell ? and use magnets to hold the buss bars on the cells instead of welding from this thread here:
http://endless-sphere.com/forums/viewtopic.php?f=14&t=60517&start=25
it would make it a lot easier for anyone to use this as getting the welds is the hard part.
In 8p configuration it is possible to get up to ~90 amps from the pack. And probably over 300A phase current, depending on your controller.
Good enough maybe?
I am not going to experiment with magnets, since I can simply weld it all together, which I think is superior.
However I can not see a reason why it could not be done. You could place a magnet on each tab and it should be fine.
BTW, magnet seems like a good way to quickly test prototypes.
Mammalian04,
I am able to vary pack dimensions. Current tabs are for 8p config, but these can be cut to 6p or 4p configs. Also several tabs can be soldered together to form larger pack, like 10p, 12p, ... etc.
As for plastic trays, unfortunately I have to design and produce each one separately.
Good enough maybe?
I am not going to experiment with magnets, since I can simply weld it all together, which I think is superior.
However I can not see a reason why it could not be done. You could place a magnet on each tab and it should be fine.
BTW, magnet seems like a good way to quickly test prototypes.
Mammalian04,
I am able to vary pack dimensions. Current tabs are for 8p config, but these can be cut to 6p or 4p configs. Also several tabs can be soldered together to form larger pack, like 10p, 12p, ... etc.
As for plastic trays, unfortunately I have to design and produce each one separately.
I think the time has come. Here's a SELL THREAD:
18650 fused assembly kit
18650 fused assembly kit
Mammalian04
100 kW
Great news! I am interested to see how the testing comes out on the first few packs.
Mammalian04
100 kW
Just the typical stuff. In bench testings, the fuses are working as they should (burning at the prescribed output) but it would be good to see the real world testing that it holds up to the bashing of a big offroad bike and its shake, rattle, and roll (well, maybe not roll).
The only issue I can think of right now is the potential for the fuse connection to fatigue from micro movement of cells in the pack over time. This may be a non-issue but there could be other issues we aren't thinking of. I would hate to weld the sheet to my expensive battery pack and then find out that "something" needs to be adjusted. It's pretty hard to unweld cells!
I don't mean sound like a skeptic, as I am super excited about the design. I am just interested to see how it shakes down.
The only issue I can think of right now is the potential for the fuse connection to fatigue from micro movement of cells in the pack over time. This may be a non-issue but there could be other issues we aren't thinking of. I would hate to weld the sheet to my expensive battery pack and then find out that "something" needs to be adjusted. It's pretty hard to unweld cells!
I don't mean sound like a skeptic, as I am super excited about the design. I am just interested to see how it shakes down.
DrInnovation
100 W
Nice idea/kit but I'd need a different size than on the "sale" thread.. So I figured I'd post 2 questions/suggestions. here
Have you considered (or would you consider) designing the fused elements as basically a long strip.. for welding 18650 many people already buy nickle strips.. if they could buy strips with pre-fused connections, especially if they fit within the existing modular 18650 Radiating Shell EV Battery Pack Plastic Holder (like supower sells on ebay) they we could configure them by soldering/welding the strips into what ever confutation we want while still getting the advantage of the per-cell fuse. You Could get to higher production volumes because one fuse deisgn/run would satisfy many kits.
Second.. I wondered how much flex there is in the plates. In particular if they would bend down enough that putting a compression element above them (as in this thread http://endless-sphere.com/forums/viewtopic.php?f=14&t=57810) might allow a weldless build -- that too would get more interest. In the cited thread they used dimpled copper, but if the compression/dimple was a non-conducting piece (e.g. plastic) that just pushed down on the fused plates it would be a great combination (presuming it does not break the fuse...) With the "wings" opposite the fuse I woudl think they would flex enough to make good connection on that side without too much stress on the fuse.. less sure about the bottom element's flex. A compression build would make replacing cells -- or even failed fused, a lot easier and without the need for welding might get more people interested
Have you considered (or would you consider) designing the fused elements as basically a long strip.. for welding 18650 many people already buy nickle strips.. if they could buy strips with pre-fused connections, especially if they fit within the existing modular 18650 Radiating Shell EV Battery Pack Plastic Holder (like supower sells on ebay) they we could configure them by soldering/welding the strips into what ever confutation we want while still getting the advantage of the per-cell fuse. You Could get to higher production volumes because one fuse deisgn/run would satisfy many kits.
Second.. I wondered how much flex there is in the plates. In particular if they would bend down enough that putting a compression element above them (as in this thread http://endless-sphere.com/forums/viewtopic.php?f=14&t=57810) might allow a weldless build -- that too would get more interest. In the cited thread they used dimpled copper, but if the compression/dimple was a non-conducting piece (e.g. plastic) that just pushed down on the fused plates it would be a great combination (presuming it does not break the fuse...) With the "wings" opposite the fuse I woudl think they would flex enough to make good connection on that side without too much stress on the fuse.. less sure about the bottom element's flex. A compression build would make replacing cells -- or even failed fused, a lot easier and without the need for welding might get more people interested
I did some prototypes and there does not seem to be any real need for such product. Or maybe price is too high? If this is the case, please let me know.
Each new 'shape' requires new tooling and costs too much for my hobby pocket, so I am not going to do any new stuff soon. For now, I have tabs for these configs: 2p, 4p, 5p, 6p, 8p and 10p. Almost any S config, but rectangular battery only.
Each new 'shape' requires new tooling and costs too much for my hobby pocket, so I am not going to do any new stuff soon. For now, I have tabs for these configs: 2p, 4p, 5p, 6p, 8p and 10p. Almost any S config, but rectangular battery only.
DrInnovation
100 W
circuit said:
Is the 5P a linear strip of 5 cells? if so how much for 25 pairs of strips?
Similar threads
