Charging parallel pack of different capacity li ion cells?

Korxi

Newly Enlightened
Joined
Jan 21, 2008
Messages
39
Location
Denmark
Hi guys,

I'm planning on making a pack for a m@g 3D consisting of 18650's and 14500's.. (one big pack with all cells in parallel)

Naturally I will have cells of different capacities around 7-800mAh and some around 2000mAh..

The pack is going to be shrink-wrapped, so I'm not going to be able to unpack them and charge them individually - now my questions are:

1) first, is this safe??

2) What charging current? at 1C for the smallest capacity cells, my guess would be that the charging current should be (number-of-cells * smallest cell capacity)?

2a) Alternatively, could I use one of the DX og KD two slot chargers and just hot-wire it to charge the whole pack at way less than 1C?

I hope you can help :)

Christian
 
Last edited:
I'm not quite sure what you mean by saying "one big pack with all cells in parallel" but to take you literally, that would mean your total voltage would be 3.7V nominal with a large mAh capacity.

In any event, it will not work to mix different capacity Li-Ion cells, nor is it safe, and saying "Naturally..." makes it sound like you think doing such a thing is common place.

If someone is making this for you, I would immediately be suspect of their competency.

Hi guys,

I'm planning on making a pack for a m@g 3D consisting of 18650's and 14500's.. (one big pack with all cells in parallel)

Naturally I will have cells of different capacities around 7-800mAh and some around 2000mAh..

The pack is going to be shrink-wrapped, so I'm not going to be able to unpack them and charge them individually - now my questions are:

1) first, is this safe??

2) What charging current? at 1C for the smallest capacity cells, my guess would be that the charging current should be (number-of-cells * smallest cell capacity)?

2a) Alternatively, could I use one of the DX og KD two slot chargers and just hot-wire it to charge the whole pack at way less than 1C?

I hope you can help :)

Christian
 
Thank you for your reply. J

Just to clarify, I am making the build my self, and my "naturally" was refering to the fact that different size cells (physical) have different capacities, not that I was assuming it was commonplace -if I did, I wouldn't be asking if it was safe and so forth..

As you suspect, what I'm looking for is one pack of 3.7V and large mAh capacity, but I still don't understand why different capacity cells can't be mixed in a parallel pack?
Won't the larger capacity cells just balance the smaller capacity ones as they discharge quicker??

When you put together a parallel pack of 18650's they won't all be the same capacity either will they?

Christian
 
Hello LuxLuthor,

While not "orthodox," I wouldn't go so far as to say that it couldn't work, and isn't safe...

The bigger question is why limit yourself to the characteristics of the lower capacity cell(s)?

Tom
 
Hello Christian,

You can safely connect Li-Ion cells of differing capacities in parallel and discharge and charge them... IF you limit the current to what the lowest capacity cell can handle. This will limit both the charging and discharging current, and the question remains why do it?

As LuxLuthor has pointed out, it would be far better to use all the same capacity cells and have a back up battery if you need more capacity.

Tom
 
Hi Tom,

Thanks for clarifying :)!

First I'll clarify too.. I'm building a dive light for nighttime spearfishing (it is leagal in Denmark.. :)), and hence I need a rather small body, with large capacity, since I can't just change the pack on open water, but need minimum 3 hours runtime - to have some back-up light if anything goes wrong out there..
Well, regarding the small size, I'm using a 3D body, and this will not fit more than 3 18650's. However, using 3 18650's there is stille room for some 14500's in the body, so to get the total capacity up, this was the best idea I could come up with since I don't feel safe around D Li-ions :)

Any other ideas would be greatly appreciated, since you no knwo where I'm heading.

To increase efficiency of the light and keep thermal issues down, while stille getting a good amount of light I'm thinking of 3 parallel P7's running at aprox. 1600mA each, using a parallel driver configuration with a total of 13 AMC7135's - Oh by the way, can I run this many 7135's off just 3,7V??..
In total this would result in a current draw of 4,8A, which then would mean that I should have at least six cells in total in my pack assuming the smallest ones are 800mAh and hence would be discharging at 1C (btw. my pack would have more cells than that, probably 12, so only 0,5C would put my in the safe zone right?) - or am I getting this all backwards?? - I am kind of a newb here :)

As for the charging, I assume the same goes, number of cells * smallest cell capacity = max charging current?

All comments are greatly appreciated :)

Christian
 
Last edited:
Hello Christian,

I would suggest that you go with the D cells. Less connections means less things to go wrong. Uniform cell capacities mean no special precautions are needed and this also means more safety in use.

Another option is to fabricate a body that will mate to the 3D head. This will give you the room to run multiple 18650 cells in a battery pack. The 18650 cells may be more reliable than the D cells, and they are also less expensive.

Tom
 
Hi Tom,

Thanks again, both are good ideas, the D-cells will be the last resort I think :)

Im also thinking of putting the back of the 3D in a vise and compress it to a shape that fits two 18650 side-by-side, and put in six of them :)

Now I'm left with my last question though. Can I hotwire a two cell charger from KD or DX to charge this kind of pack?

Oh, and is it possible to run 13 AMC7135's off only 3,7V to drive the P7's

Christian
 
Hello Christian,

Yes, it is possible to charge your battery pack by hooking it up to one of those chargers, but the charge time is going to be quite long. A wild guess would put the charge time at around 50 hours.

Tom
 
It is a lot clearer what you are trying to accomplish now. In general terms, as Tom pointed out, my safety concern was based upon managing the differential limitations of charge/discharge, variable internal resistance differences, and safely making parallel connections between the different types of cells. I was not seeing how to safely take into account the resistance of the wire guage & lengths, connecting without soldering the cells, and making sure you have no shorts as you join all the Negative ends together, and Positive ends separately together.

I think you would have a hard time finding the side clearance to run and parallel connect wires with D cells in a Mag3D. If you get the inside body bored out, you can fit 2 wide 18650 cells, 4 wide x 14500, or 3 wide 17600, and have room for wires. However, connecting them in parallel as a "1s" waterproof setup is a significant structural challenge.
 
On paper charging LiCo cells of different capacity in parallel works fine most of the time. There are some circumstances where this would be ill advised but if the pack was always being charged and discharged as a pack, and the charge and discharge rates were well under the maximum safe limits then there would be no issue that I can see...


However, I think it would be much smarter to just try to standardize on one cell size....

If you dual bore a mag body you can fit 18mm cells side by side, or skip the boring and fit 17mm cells side by side, or tri-bore and fit 17mm cells 3 abreast. When you consider the intermediate spacers and wiring and connections you are going to need to make this thing a parallel pack you may want to opt for 50mm long cells. 9 17500s or 6 18500s or 6 17500s...

This really sounds like a job for an elephant or colossus body style light though- 4 18650s abreast would give you the kind of runtime and output you are after.

-Eric
 
Hi,

Thanks for the inputs both of you! I hadn't really thought of the porblem with wiring D-cells in parallel :) good point there - so they're definetly out of the question!

I think I have to clarify one more little thing - I'm a stundent and just bought an appartment in crisis times :( = I'm totally on a budget.! :(

So I'm trying to do this as cheap as possible = DX or KD parts as far as I can go, since shipping costs from the rest of the world would just bring the cost way up.. So I'm kind of "stuck" with the m@g bodies - being the largest on DX, and I don't know anyone to bore the body for me..

The two best options for me now I think would be to either fabricate a wider body to mate with the m@g head (thanks Tom!) as this would be relatively easy I guess - haven't got any experience with cutting threads though :D - or put the body in a vise and compress it till I can fit 18650's side by side

The reason I'm trying to build the light myself is mostly to save the money compared to buying a wiseLED tactical, which is kind of what I'm trying to copy performance-wise :)
If I ever get there - using the 3 P7's - I'm thinking of doing an aspherical setup (individual asphericals for each LED), with different focal lengths for all three, one very narrow spot at 7-10 meters and two floods (diameter of spots: 5, and 10 meters) - any ideas or comments in this regards?

Christian

Edit: Norm, my plan is to get a hobby charger, I was just wondering whether the cheap'o chargers would do the job for a start :) I've been looking at these two also:
http://www.ecpower.com.hk/index.php?main_page=product_info&cPath=66&products_id=182
http://www.ecpower.com.hk/index.php?main_page=product_info&products_id=185
As reviews of the DX chargers are not all that good - not heat sinked for discharge and unisolated from case.. - but then again, I haven't found reviews for those two above..
 
Last edited:
Christian, you can see that great minds think alike with mdocod repeating the various Li-Ion diameter cells that I mentioned. But if you need to do this on the cheap, you would be best using a stock D Mag. Without boring out, you are only going to fit 3 wide of 14500 (750mAh each) or 14670 (1100mAh) in either case there would be plenty of room for wire. I don't see how doing a fabrication and threading is going to work out in an affordable manner with a waterproof application. Only other idea is to use threaded PVC pipe...but you are in for a lot of searching for parts to fit.

You still have the challenge of how to attach wires safely to the Li-Co Ion cells, as no official reputable source recommends soldering them. Adequate waterproofing to your actual depth is another question. Certainly the rubber gasket over the switch is not going to work with the pressure changes underwater. You said this is for diving, but if you are walking in shallow water doing the spearing, then it could be made splash/rain protected much easier.

Unless you are in crystal clear water, I would think a far throwing configuration for the LED's would make the most sense regarding penetrating the suspended fish poo & other particulates.

You don't need a balance charger for a 1s setup, so you could cut some $$$ corners by getting a 0.5C Amp output charger and a cheap DMM and watch the voltage....and stop charging at say 4.15V Charging is gonna take a long time no matter what you use.

The chargers you linked are certainly cheap, but it's hard to know of their quality without being independently reviewed. I suggest you also look through the www.rcgroups.com forum where they have much more experience with various Lithium chargers, and may know of the ones you listed.
 
Last edited:
In regards to the waterproofing, the max depth of use will probably be 8 meters, since I'm only snorkeling :) I plan on buying larger o-rings to waterproof the head and tail. I plan on taking out the rubber cap of the switch and instead make a waterproof silicone cap that is glued to the body.

I've also given the pack assembly some thought and I've come up with this possible sollution (this is based on 3 x 2 18650's side by side in a compressed 3D body) :

1)
a) Shrink wrap the cells 2 by 2 facing the same direction
b) Lay three cell packs like this P +IIIIIIIII- P -IIIIIIIII+ P +IIIIIIIII- P
b*) The "P" represents a metal plate with a wire attached
c) Shrink wrap the whole thing again, and there you go.

I hope my "grapihcs" come out good in the post :D

I regards to the beam charachteristics what is needed is actually a combination of throw and flood, as the bright spot is used to see far and camouflaged fish, and the flood is used when approaching the fish to shoot them - as they are scared off if a bright spot hits them right on :)

Christian
 
In the RC world, more and more people charge all kinds of cells in parallel since I started this thread:

http://www.rcgroups.com/forums/showthread.php?t=932319

I do it most of the time, and I also discharge packs in parallel, that do not have the same capacity, as one 2000mAh pack and two 1000mAh packs.

I have never experienced any problems with this, nor do I know anyone who has.

Cheers,

Julez
 
1)
a) Shrink wrap the cells 2 by 2 facing the same direction
b) Lay three cell packs like this P +IIIIIIIII- P -IIIIIIIII+ P +IIIIIIIII- P
b*) The "P" represents a metal plate with a wire attached
c) Shrink wrap the whole thing again, and there you go.

Flat plates may not make good contact, depending on your cells. I'd recommend using something with contact springs or nubs, but the basic scheme looks sound. You're talking wiring like this, I assume?
Code:
 [COLOR="Red"]___________________________[/COLOR]
[COLOR="Red"]/[/COLOR]                           [COLOR="Red"]\[/COLOR]
P +IIIIIIIII- P -IIIIIIIII+ P +IIIIIIIII- P
              [COLOR="Blue"]\___________________________/[/COLOR]


You said the D mag-clones (BTW, they're not mechanically compatible -- if you're thinking of any Mag drop-ins, bet on serious modification) are the fattest KD/DX have, but that's not strictly true. DX carries 3 Ultrafires that have a 3x18650 in carrier configuration (think jumbo version of a 3xAAA light :D)
  • sku.20231 -- 1x halogen (maybe WA 1185?)
  • sku.20233 -- 7x SSC P5
  • sku.22573 -- 1x SSC P7
But they do cost a bit much for a host light, and since they're short, you actually lose capacity; probably not right for you, but thought I should point them out.

If you go the squash-a-mag route:
  • Consider the threads at both ends of the tube. If you're converting to fixed focus, the front threads may not be an issue, but even if you crush it with the cap screwed in, I wouldn't be surprised if you slightly oval the cap and threads such that it won't unscrew. The best thing, I think, would be to have a close-fitting steel plug in each end of the light, inboard of the threads, and crush it between the plugs, but still no guarantees...
  • You mentioned fabricating a body, sounds like you have access to a decent-sized lathe. How about using a 3-jaw chuck to squash the D body into a trilobular shape, so it takes 3x18650? Maybe fit the same 6 cells in a 2D, or put 9 in the 3D...
 
Hi Benson,

Yes, that is exactly how I plan on wiring the packs. I'll have a look at putting some nubs on those plates!

Thank you for linking to those lights, they are a bit expensive, and I don't think they look like good subjects for water proofing :)

Un fortunately I don't have access to a lathe or anything at the moment, so the fabrication idea is more or less a last resort..

In regards to the squashing, I was planning on only shaping the back part of the cylinder, by putting in a steel rod in the front as you say. I realize the tail cap wont fit, so I plan on just plugging it with silicone, and only "access" the light through the head - and if i ever have to get the pack out, I'll cut out the silicone.

Christian
 
one thing that *may* be more acceptable is having two circuits and a switch in the light..... a set of parallel 18650s and a set of parallel 14500s. You set it up to switch from one pack to the other and also charge each set seperately. This would keep capacities more balanced and also a bonus when you use up the 18650 batteries and switch to the 14500 you know you have only a certain amount of time left of light.
 
Top