18650 battery holder

[leamcorp]

Thanks guys - I've got zapped few times also.

Got it about the Parallel! That will make it simpler for this holder. I just need to figure out how to easily connect holder to holder. I'm still waiting for few important items. Hopefully those will speed things up.

 

[pwbset]

I just need to figure out how to easily connect holder to holder.

Since the "subpacks" will be (most times I assume) connected in series having space for a positive and negative lead off each subpack would be good.. obviously. Some sort of overlapping male/female clip together system would be ideal. Andersons somehow? Just thinking out loud. I still think 8p is like a minimum usefulness for 18650 ebike use. 4p just isn't enough I don't think. Those Tenergy cells are very unlikely to deliver the goods under ebike loads at only 4p. Any 18650 worth it's salt will be 1.5ah or less... that I've read about anyway... 4p even then = ~6ah... not very good unless they are 5c+ sustained cells. What is the hesitancy to do higher parallel packs like a 10p (maybe dual side 5p to create that)? Materials? Weight? Cost? Just curious.

Keep up the good work!

 

[MitchJi]

Hi,

Since the "subpacks" will be (most times I assume) connected in series having space for a positive and negative lead off each subpack would be good.. obviously. Some sort of overlapping male/female clip together system would be ideal. Andersons somehow? Just thinking out loud. I still think 8p is like a minimum usefulness for 18650 ebike use. 4p just isn't enough I don't think. Those Tenergy cells are very unlikely to deliver the goods under ebike loads at only 4p. Any 18650 worth it's salt will be 1.5ah or less... that I've read about anyway... 4p even then = ~6ah... not very good unless they are 5c+ sustained cells. What is the hesitancy to do higher parallel packs like a 10p (maybe dual side 5p to create that)? Materials? Weight? Cost? Just curious.

Keep up the good work!

The issue is that if you only have 8p holders then you are limited to 8p and 16p etc. configurations. So what do you do if you want 10p or 12p?

 

[pwbset]

The issue is that if you only have 8p holders then you are limited to 8p and 16p etc. configurations. So what do you do if you want 10p or 12p?

Okay.. I could see that, but again unless they are konions or the 18650 A123s most Li cells in this format and in 4p or 8p are just not going to deliver under ebike loads so why not just do multiples of 10 to keep it simple? Even a crap 2C 18650 cell in 10p would at least give a reasonable amp output... at 4p... forget it. Plus I wouldn't want to have to series up like 3 4p packs and then series up those again... becomes kind of a balancing nightmare doesn't it? The higher the 'p' the better the self balance... right? Or am I missing something basic here... entirely possible! I'm not very bright this time of night usually. Haha. :lol:

 

[MitchJi]

Hi,

The issue is that if you only have 8p holders then you are limited to 8p and 16p etc. configurations. So what do you do if you want 10p or 12p?

Okay.. I could see that, but again unless they are konions or the 18650 A123s most Li cells in this format and in 4p or 8p are just not going to deliver under ebike loads so why not just do multiples of 10 to keep it simple? Even a crap 2C 18650 cell in 10p would at least give a reasonable amp output... at 4p... forget it. Plus I wouldn't want to have to series up like 3 4p packs and then series up those again... becomes kind of a balancing nightmare doesn't it? The higher the 'p' the better the self balance... right? Or am I missing something basic here... entirely possible! I'm not very bright this time of night usually. Haha. :lol:

If he makes 4p (for example holders) you could combine those in parallel and then combine those in series. For example 3 4p connected in parallel. Then series connect the groups of 3. This would give you the option of 4p, 8p, 12p, 16p vs 10p and 20p with 10p holders. If its easy to connect the holders either serially or in parallel (which is one of the main points of this exercise) then the extra wiring shouldn't a show stopper and the extra flexibility would be valuable.

I think the best option would be a selection of sizes. I have no idea what the best sizes would be but using 4p, 6p and 10p, for example, would allow you to create most even p ratings between 4p and 30p by connecting a maximum of 3 packs in parallel.

 

[leamcorp]

The holder could be whatever we want - I just need a definitive answer as to what combination will work best.

What would be an optimal configuration for a 48v 10ah or 48v 20ah? Would increment of 4P or 8p works best?

4p trustfire is 10ah (per their claim). So 8p would be 20ah, right? 12p = 30ah?

So 8p and 15s would make 48v 20ah (120 cells)?

As for the casting capability - there are limits but I don't think 8 cell holder is a problem. Again, I'm trying to make it modular - I just need to know what would be a good multiplier (4, 5, or 6 cells increment). Some one needs to explain this to me.

 

[leamcorp]

would allow you to create most even p ratings between 4p and 30p by connecting a maximum of 3 packs in parallel.

Confused as ever...

4p I understand.

30p? Isn't that - 2500mah per cell = 75ah? Isn't that bit high for Ebike? or am I not understanding parallel thingy.

 

[CGameProgrammer]

The holders must have the cells in parallel, or not connected at all, in order to be able to balance the cells.

That's another thing - when all these are connected into one whole pack, the pack outputs need to be one ground wire and one positive wire for each parallel-string. So whatever mechanism is used would need to allow for that.

 

[Turboferret]

Nice progress!

Just found this thread - I started a very similar project a few months ago, although for mountain bike light battery packs, so for a smaller number of cells.

It has been a fascinating project where I've learned a huge amount, and has now paid for itself too!

I've now got a few products on sale, single, twin and quad holders, cast in polyurethane resin from silicon moulds.

a few pictures

I was fortunate enough to have a volunteer produce me some rapid prototype patterns, and my holders lend themselves more to moulding due to their size. Prior to that I did make my patterns from cardboard, but the quality is now vastly superior

have a look at Turboferret where you can see a pictorial history of the product development - it looks as though we have followed a very similar thought pattern for some stages

I designed everything in SolidWorks which has been fantastic both for trying out ideas, and for producing a CAD file to rapid prototype from.

If you want any advice, feel free to get in touch.

Cheers, Rich

 

[Turboferret]

Just seen that I should have read all of this thread before jumping in - I seem to crop up on page 3 :lol:

Doesn't look as though you need any advice from me

Cheers, Rich

 

[marty]

Turboferret / Rich,
Good to see you here. Now lets see a design for a big electric bicycle battery holder. 48V 20Ah or so. Do some research on the series parallel stuff. Doctorbass is the one to ask those questions? Go here to see pictures of Doctorbass batteries. http://www.voltev.com/doctorbass/

Want 18650 and A123 26650 battery holders

I know almost nothing about Rapid Prototyping and Plastic Injection Molding, but I did play around at this place http://www.emachineshop.com Designed a part. Click on price and almost fell off my chair. Then emailed design to local CNC Laser Cutting place. Went to talk to CNC guy. Ask how much he charge. He said nothing. I said "thats not enough, how about $10?" He said OK. Go here to see Marty's Torque Plates

 

[Turboferret]

Those Doctorbass are pretty bad ass

The capacity and current draw requirements are probably an order of magnitude above what us bike light guys need, so I'll keep schtum and let you carry on :lol:

CNC and rapid prototyping can cost an arm and a leg, if it's being done as a business, CNC particularly so for small numbers, as set-up time is a major part of the production. I was fortunate enough to have an MTB guy volunteer the services of his work machine as a favour, it certainly wouldn't have been cost effective to pay for parts commercially.

Cheers, Rich

 

[leamcorp]

I use solidwork when doing client work - actually these days I use Alibre more than solidworks. But for this project - I use left over material from work, so no modeling work is required. You discover more options when doing it old fashion way of prototyping. And I hate doing SLA or FDM - they don't come out very nice and can't get the parts thinned out. CNC is still king.

Turbo - next time you make a mold, try to encase your model in a non-hardening clay. Then paint the surface first before puring RTV. This will eliminate the bubble without needing a vacuum chamber. Still it must be a slow process as you have to pour, then wait a while to demold then repeat. kudo for making some profit though.

Oops, hit the submit button too fast - edit.

The goal of this project from the start is to make it cheap or it defeat the purpose of using cheap batteries. The blocks sold by Rechargeable Lithium Power are 80 cents each. So hopefully I could still do it within that spec - with each multi-cell holder costing no more than a dollar in material. The initial development is free since I'm doing all the work. The only cost is the vulcanized rubber mold - which isn't cheap but could be spread out quite cheaply. Good thing is, I could pop these out at rapid rate and i wouldn't have to revisit them after I get my pack (and for few more people here).

 

[Turboferret]

I've learned a lot about moulding, and air bubbles are no longer an issue

The fact that I was pouring the whole mould in one hit, rather than using clay and a 2 sided mould, so there were areas of the pattern which would act as shelves, and prevent air bubbles from rising to the surface.

By vacuum degassing the liquid silicon before pouring, and making sure that the whole thing is rotated at certain points, I can make a pretty much perfect mould at one time.

Not a viable method for the scale of holders you are needing though.

Thanks for the tips though, and good luck.

Cheers, Rich

 

[leamcorp]

Speak of the devil, I've just received some goodies and one item caught my eye. Its a AA battery holder and it has a rivetted 9v connector (I assume to connected to another holder or for making wiring).

Slayer - you've talked about riveting, now I understand. Where did you get the rivets and the gun. Does anybody know where I could get those rivet with 9v connectors? This would make things so much easier.

 

[MitchJi]

Hi,

Hi,

would allow you to create most even p ratings between 4p and 30p by connecting a maximum of 3 packs in parallel.

Confused as ever...

4p I understand.

30p? Isn't that - 2500mah per cell = 75ah? Isn't that bit high for Ebike? or am I not understanding parallel thing.

pwbset suggested 10. I was just pointing out, that an array of sizes would be much more flexible than one sized at 10. This would allow almost any capacity, and only require connecting a maximum of 3 holders in parallel.

4p trustfire is 10ah (per their claim). So 8p would be 20ah, right? 12p = 30ah?

Correct.

The holder could be whatever we want - I just need a definitive answer as to what combination will work best.

What would be an optimal configuration for a 48v 10ah or 48v 20ah? Would increment of 4P or 8p works best?

There isn't any "best" size. It depends on what pack size an individual wants to build. It also depends on the space available. Its possible that someone wanting 12p would be better of in terms of spacing using 4p blocks as opposed to 12p blocks.

If you can make it work well I think the best solution would be something that would hold 10 or 12 cells that could into shorter lengths to accommodate a variable number of cells. These could be shaped like the C Channel below. The cells would slide into the channel.

 

[pwbset]

This 10 cell AA holder is pretty much exactly what I had in mind except for 18650 sized cells and all parallel instead of series. Instead of a 9v rivet type subpack connector I like the connectors in this example as you'd have a lot more flexibility in layout of multiple subpacks.. like fitting into a triangle enclosure better.

Sorry about image quality.. web image.

 

[pwbset]

Very poor image quality, but here it is with the 9v style connector on the side:

 

[leamcorp]

I think I got the concept down now. Unless someone could think of a better way, this might be the final prototype before I start working on a usable version.

This design is even simpler than before and lot more versatile. A modular holder that could be setup in any orientation or # of cells or even different size cells (with different endblocks).

(Please remember that this is just a prototype),

1st pic - the design is based on metal strips that can be inserted into 2 end blocks. This design is not fixed - rather, its the metal strip that will determine the size. For longer setup - use a thicker metal strip (1/8"). The only thing missing from this picture is a support structure or something to hold the battery in place ( 2 metal strip bolted on the back should do the job).


2nd pic - Both side will have clips - I've measure the distant wrong. The clips I'm using are quite strong, so I couldn't put the battery in with tabs on both side.


3rd pic - Hard to see but its one holdeer on top of the other. For this setup, you could use a wider metal strip to cover multiple levels.


4th pic - Side by side setup. I think configuration can be pretty much any shape - as long as its braced correctly.


Once side blocks are thinned and scaled down, only other thing is to come up with support brace and front/back cover.

 

[pwbset]

Bump. What's the latest leamcorp? I'm about to order like 500 211 springs, but want to see what you've got going on first. Looks good so far.. got any completed stuff?

 

[leamcorp]

Bump. What's the latest leamcorp? I'm about to order like 500 211 springs, but want to see what you've got going on first. Looks good so far.. got any completed stuff?

Another day or two and I should have a working model.

I'm not using spring - takes too long to put them in or you'll be so frustrated that you'll give up after few. I've even tried to snip part of the spring off (in the front) so that I could put machine screws in - didn't work so well. I have to say Gary has amazing patient as it must took him forever to put those in.

I'm using the clip, which I think is even stronger. And it doesn't come off as there's a nub on the back that hold it there.

By the way, I'm also making a dummy 18650 cell using these clips. I had one made last year (very crude) and with these clips, its really a cinch. I'll be using those to power my SSC P7 lights and 4 or 8 cell - should last few hours with those !!! Really cheap way to get 1000+ lumen lights on your bike.

 

[pwbset]

Thanks for the update leamcorp. I'll hold out on an order for awhile longer. The springs are quite a bit cheaper is why I want to use them... every penny counts these days. Besides there several feet of snow outside so I'm not going to be riding any time soon... patience is a requirement. :wink:

 

[paultrafalgar]

I didn't read the whole of this thread, but you might like to thing about this:
http://endless-sphere.com/forums/viewtopic.php?f=14&t=8617&p=131191#p131191

 

[leamcorp]

Here's dummy cell for the bike light I've talked about. I made one last year but it was very crude - so I just thinned things down and put the clips on and Viola ! cheap and simple external power source.

This is what the dummy cell looks like. Just slip the attachment - put some cable on the back side... you don't even have to solder the wires but I would. Oh you have to drill the light housing.


Close up of the light housing - I just put a small hole with a drill.
View attachment 2

Assembled!


As for making it water proof - you could use a grommet or for lazy person like me - just use hot glue. Once battery holder is finished, I could have 4 to whatever cell I want. If you use a DC to DC converter, you could even connect it up to your main battery. I use two light combination and it light up the road - and since I'll have much larger external power source, I'll probably use 3 (one in reserve).

 

[pwbset]

I didn't read the whole of this thread, but you might like to thing about this:
viewto ... 91#p131191

Paul what you suggest in that thread is pretty similar to what I'm attempting with my triangle build. I'll have 14 subpacks of 10 konions that will essentially slide in and out of custom "boxes" in a triangle mount that will be side-closed with some sort of bolts that tighten down on the springs/cells. I'll have a little "ram air" slot in the very top left/tip that provides a little air flow, but I don't expect to ever pull more than 3-4C and konions don't get very warm under light loads like that so it should be fine. Each side of the triangle board will have the copper bus bars and springs mounted directly to the wood and wired up to the controller internally in 14s (if I can get it all to fit). It's going to take forever and be a complicated build, but in the end I should have a large 52v/15ah triangle mount pack that will be easy to replace individual cells as needed and last for a really long time. I'll have a ton of extra konions and will carry a spare dozen or so in my backpack (1.2lbs no big deal) on the trail in case one or more dud mid-ride.

This is not to scale or anything, but gives an idea of what I'm talking about.