Battery info

Projects / Project Swiftlet  

AA battery - Alkaline

AA battery discharge curve from Energizer datasheet.

Lithium batteries

When it comes to power density and weight, it is hard to beat the Li battery chemistry. I want something small but easy to be replaced, so looks like Li-ion in CR123A (16340) form factor might be suited for my application. Wiki page say typical battery capacity is 500-1000mAHr for rechargeable ones.

Primary cells

There are also non-rechargeable version at 3V around 1500mAH ( e.g. Duracell, Energizer (datasheet), Rayovac. Those are expensive, but offers decent capacities for emergency replacement. Might also be worth while to have a connector for those RC heilcoptors for emergency replacement?

Digikey don't sell Li-ion batteries probably because of transportation safety issues. At the usual Chinese mail order place, I can get these batteries at around $2 a piece. For prototyping purposes, I guess they have to do.

Rechargeables

• Product name:3.7v CR123A/16340 Reachargeable Battery
• Capacity: 1000mAh
• Voltage: 3.7V
• Chemistry: Li-ion
• Weight : 16g± 1g (1pc

As with most of the claims, the Chinese battery capacities can only be taken with a gain of salt. User comment on he cheapest SKU:"Working battery, but capacity is only about 350mAh. Checked with imax charger." This is what my battery analyzer is for.

So in theory, if the power consumption is on average 2mA, a device could be operated 24/7 for a week on these batteries. 2mA * 24Hr/day * 7 day = 336mA*Hr This is a reasonable target.

Li-ion batteries have voltages between 3.7V to 4.2V, so a buck converter with low quiescent current is what I would try to use.

-------------------------------------------------------------------
The battery arrived. (1 month+1 day)    - Those were the days when mailing order was still great.  :(

I would be very surprised if the capacity is near 900mAHr as they claim.  :P

Towards the negative terminal, you'll see the case slightly tapered off. That's where the protection circuit resides. There is a thin flat strip at the middle just right at the red/black seam is the connection to the positive terminal for the protection circuit.

The gold terminal is from the protection PCB. The battery is slightly thicker.
Contrary to internet believes, you'll need a proper charger for charging the battery. The protection circuit isn't a charger.

In terms of weight:
Trust Fire Li-ion: 20.3g, Duracell Alkaline: 23.7g, Energizer NiMH: 26.3g
Soshine "700mAHr" LiFePO4: 18.3g
-----------------------------------------------------
I am a big believer of doing my detail designs up front than to waste my time and money trying random things in a frenzy in the lab. I started doing 3D modelling for my projects since my 2014 HaD contest project. They are not just pretty pictures to impress readers. The more I used it, the more I am convinced that it is a very important step for checking badly made libraries. Sometime the vendor's CAD drawing is wrong!

This following 2D CAD have missing dimensioning and wrong number. Top left hand corner, the length is actually 0.91" instead of 0.71". There is no offsets in the drawing. There are no PCB footprints and suggestions on how far the 2 halves of the connectors should be.

How likely are you to get this right in your first prototype?

Battery holder mechanical drawing

Thankfully, the vendor provided a STEP model. I used EagleUp to export my PCB to SketchUp. A few iterations later of fixing footprints and outline, this is what it looks like. The 3D models shows me contour around the battery and holder, so I can plan my component placements.

Sketchup/Eagleup modelling of battery holder

This is how much compression I think I'll need for the battery contacts. No amount of 3D modelling can tell you is how snug the fit is. I'll need to etch a PCB to test it out.

Battery holder: Negative terminal

------------------------------

"The Last Mind Bender!?" Original post date: 07/22/2015

"Do not try and bend the spoon, that's impossible. Instead, only try to realize the truth...there is no spoon. Then you will see it is not the spoon that bends, it is only yourself." ―Spoon Boy to Neo

Doing some test for the battery holder. As I have noted before, the manufacturer does not have any layout info. So I am doing some trial & error. Well mostly on the error side.

1. I measure the distance between the battery contacts at no compression force.
2. Figure out the length of the battery.
3. Now calculate the offset from 1.

The retention is good for this, but there is a bit of bending on 0.052" thick DIMM module PCB which is pretty close to the standard 0.062" (1.57mm) PCB thickness. And I was tempted to go for the 0.8mm thick PCB to shave weight. Guess that's not going to happen.

It gets worse for the Protected Li-Ion which is 0.06" (1.524mm) longer to accommodate for the protection PCB at the base.

Trying my luck again, but this time with a 2 layers PCB from the Chinese PCB I order from. It is quite a bit stiffer than the DIMM. Looking at the way their PCB cracks - only thing comes to mind is that the core is "very dry" (i.e. little epoxy). I would assume that they have a higher glass content than the old hobby PCB for etching. Could be higher TG PCB for RoHS???

The standard AA, almost no bending.

The Protected Li-Ion, slight bending on the PCB side.

I am using 0.06" (1.524mm) of compression for the Protected Li-Ion, so the Fuji Film would just touch the contacts.

A sample of "AA" batteries my collection:

So if I discard the Fuji Film battery data point, there is a smaller spread: 1.975" - 1.991" which is 0.016" (0.406mm). The trick is to target the mid point and let the very stiff springs absorb the differences.

Note: The measurements includes a piece of tape over the positive terminal so that I don't short the batteries.

Lesson learnt: Buy a different style of battery holder! The springs are too stiff.