DIY tripleAA in series battery holder


My entry in the OL-fromscratch contest for this year may again not be finished when the deadline arrives, but part of the build is this 3xAA 3S battery holder. I thought that it might be useful to others but me, so here's a post about it. The reason I wanted a 3xAA holder instead of a 4xAA holder is that with 3xAA's (Eneloops) it is suitable for direct driving most used leds, like XML2 and XPG2. In my flashlight build, I will use a direct drive FET-driver. I designed it to be used with a switch fixed on it and the two poles coming out at the top plate, but it can be used without switch as well, and with clever bridging some pads (I'm sure you can work it out ;-) ) the poles can also be made to come out on both sides (no copper wire in the middle) to use the holder as a 32650 battery replacement.


I made the design, sent the board designs to Mataus and after some discussion on improvements, he made the Eagle-files and sent them to Oshpark. Many thanks go to Mataus for his effort on this. This week I had finally some time to build it up. This is the original design I sent to Mataus, the final design is fine-tuned on details but in principle not different:

These are the parts I started with, a bottom plate, a top plate, a contact plate that will be fixed inside the flashlight (Mataus put my BLF-name on it, so if you use it you will have to bare with that ;-) ), some copper alloy springs from Fasttech, messing rings, an Omten switch and some 4mm brass bar.

First thing to do was to solder the springs on the negative poles, and the brass rings (I reamed the holes a bit to fit the average battery plus-button) on the positive poles (the rings are not really necessary, but they keep the batteries in position a bit better). Then cut three brass pillars to size (I chose 54mm distance between the top and bottom plate), the design was made to solder the pillars into the 3mm holes, so the end of the pillard had to be narrowed down to 3mm. I used my drill press and a small file for that, it took a while, but that was mainly because of the poor quality of the file :-( :

I guess that you can also use pillars with M3 (3mm) screw holes and use screws to fix the pillars, but you may have to use a smaller switch than I did to make room for the screw heads (there is a very decent small Omten switch that should do the job).

I soldered the pillars first from the inside, the assembly clamped lightly in a vice, using a small butane torch to heat the brass, I was careful not to make it too hot and burn the circuit boards:

With the pillars fixed in place I could take out the holder and solder the pillars from the outside of the boards as well. At last a copper wire (I stripped some common installation wire) was soldered in the center (using a solder iron) to guide the positive pole back to the top board. Here's how it came out:

It felt really nice and stiff now, like a real high end battery carrier :-)

Now the switch was soldered and the contact plate was build up with a spring and some small copper blocks, and the work was done:

I have not done extensive testing on the performance, but the resistances in this holder should be pretty low by design.

What I did do was test it on the head of my SWM D40A flashlight. Stock it has a 4xAA in series battery holder, putting out 980 lm on turbo, with this 3xAA in series holder I get 840 lm on Turbo (slowly decreasing, indicating that the voltage sag of the batteries is the limiting factor when using the lower voltage of one battery less than normal), which I think is pretty good, and a sign that the resistance is very ok. On high, the battery holder has no problem at all to keep the output at the normal 570lm for at least a while.

EDIT: in post #7 I connected this carrier with Eneloops to a XM-L2 led and measured voltage and current through the led.

Perhaps this may be useful for someone, here are the Oshpark links for the plates:

http://oshpark.com/shared_projects/3A1aMPND

http://oshpark.com/shared_projects/DgMeIHsa

http://oshpark.com/shared_projects/2T8aR4dR

It is pretty expensive to order the three of them ($16.50) but you get three of each, for three holders.

If anyone has smart ideas how to improve the design, to make it more versatile, or make it do things it can not do now, be free to post about it here, perhaps Mataus is willing to implement it?

A very nice effort guys and well executed djozz. You did an amazing job with the crappy file and drill press.

I was just thinking about those…awesome build man

Love that purple board goodness :smiley:

Thanks for the replies MRsDNF and Warhawk.

I updated the OP with the reason why I wanted a 3xAA instead of 4xAA, and I included my original drawing.

great job, it looks awesome, far better than most holders I can think of

super awesome work there, this is so nice to see just now that i was thinking on doing an 3 AA holder for my project, what a coincidence!!!
I am searching how AA behave under load and i think you can give now very detailed results if you will have time.

if it is not a problem can i post my battery section project, here too?

I need to know the dimensions in mm if it is possible too to see if it is compatible with my project just for fun. my idea is to use 3S3P (9)AA for 4.5V output hoping they will have higher V under load than 18650 (4.5V versus 4.2V)

Feel free to post anything here :-)

The length of the carrier without switch is 5.73 cm, the diameter is 31mm, with Eneloops the batteries are as good as level with the side of the boards (one is sticking out a tiny bit, less than 1/4 mm).

You only get 4.5V with 3 alkalines, much less with any serious load. NiMh's are not 1.5V, but 1.2V (a bit more when fully loaded), so the voltage is not better but a bit less than a Li-ion. But NiMh behave very good under heavy load, so who knows how well it goes...

I will do a direct drive test (no driver) with a XM-L2 tonight or tomorrow, and measure led voltage and current, this will give an idea of the possibilities of this battery set-up.

I did a small direct-drive test on a XM-L2 with this battery holder and three charged XXX Eneloops (don't have any other AA NiMh rechargables), this is the set-up:

The led was on an alu board screwed to a block of aluminium, voltage was measured at the led-board solder joints, current to the meter went through thick massive copper leads.

A few seconds after start I measured 2.87 A with 3.17 V over the led, 5 minutes later it was still 2.50 A with 3.10 V over the led.

5 minutes later:

If I read HKJ's tests of these cells correctly, after five minutes, at 2.5A load, the voltage of each cell should be about 1.24 V, so 3.72 V the total of three.

I measured voltage directly against the battery under load: 1.18 V at 2.5A (that is 3.54 V the three), batteries had run for 7/8 minutes , that is lower than HKJ's measurements, could be that my batteries are already a bit worn out. At the led-board there was 3.1 V leftover, so apparently I have lost 0.44 V underway (more than I hoped, and I don't think it is mainly in the DMM/leads), that is a 'travel resistance' of 0.18 Ohm, or a power-loss of 1.1 W. I have no idea how good or bad this is actually, I never did these type of measurements for any other battery/flashlight circuity.

Concluding: you can indeed make a serious output flashlight this way, with three Eneloops and using a direct drive FET-driver and XM-L(2) led or any other led with a low Vf. But for instance a XP-G2 in this set-up will probably already run under 2A. But it is never as good as a single high current 18650 li-ion cell: in a flashlight that can deliver over 5A with an XM-L and FET-driver. To get that performance, you need 4 NiMh cells in series, not 3, and you even need some current regulation then.

Forgot to mention, for the true DIYer: a 3xAA carrier like this fits exactly in the standard (european?) 35mm copper installation pipe (inner diameter 32mm) :-) :

For what it is worth, I made some brass dummies for AA and AAA batteries and measured the resistance of this triple AA battery holder and for comparison a cheap triple AAA-holder from Fasttech (did not have a cheap triple AA-holder to compare it with).

My DMM does not display Ohm-readings more accurate than one decimal, but... with the brass dummies in place, the cheap AAA-carrier shows 0.4-0.5 Ohm, my AA-carrier shows 0.0-0.1 Ohm. So it seems to help to make one like this :-)

On another note, and inquiry

Those small contact blocks, could you use those round 5mm metal pilars be used in lieu of those copper slugs?

I have been using these as a way to put button tops on my flattop batteries as well…nice thing…if/when batteries go kaput…I just unsolder them and re-use them

Do you think it would be possible to make a longer carrier to hold 3s2p cells, the cells end to end with no middle contact plate?

Never thought of making it longer, nice one, it would work, however, the way the boards are wired now you will end up with 6 cells in series. For a buck driver that should be a nice power source :-)

Don’t you mean 2s3p?

It’s a nice set of boards that you put together very cleanly Djozz. To make it more versatile and help justify ordering a set of 3 is it possible to design it to work either series or parallel with the use of jumpers like the Noctogon triple?

Well I have a set of top boards on the way (plus 300mm brass rods), I made a cardboard mockup and I think it should work well in the D cell maglight I’m doing. I may have to cut / add a trace or two (and I’ll have to figure out the top center positive contact) but I’m sure I’ll get it going.

On a related note I also have boards coming for a 2P conversion of my SR51 18650 carrier (thanks to WarHawk)

The 3P is made by bridging the three pads of the top boards (and solder three springs on the pads for the battery negatives), and bridging the three pads on the bottom board. You can still use the switch and middle connector back to the top board for the plus, if needed.

Hmmm, dreaming away now..., how about making it three times as long, put in 9 14500's in series and make a flashlight with one of those round 9mm diameter 35V Luxeon Cob leds (preferably the 4000K 90cri) behind an aspheric :bigsmile: