Would 2x 1/2AA batteries in Parallel Work?

Ideas and questions about MintyBoost kits

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
Locked
User avatar
stinkbutt
 
Posts: 593
Joined: Wed Feb 17, 2010 2:40 am

Would 2x 1/2AA batteries in Parallel Work?

Post by stinkbutt »

Can't see any reason why they wouldn't. Nominal voltage for a 1/2 AA is 3.6V, well within the tolerances for the input voltage of the MintyBoost, which appears to be 3-4.5. Two of them in parallel (850-1200 mAh @3.6V each, for 3-4.5 Wh per) seem to have a comparable capacity to 2x AA in series (2700 mAh @3.0V, for 8.1 Wh total). Sure sure, I'll need to find myself a battery holder...

https://www.mouser.com/Power/Battery-Ho ... ?P=1z0y61m

But that doesn't seem to be too difficult.

Can anyone foresee a problem with this plan? I'd like to try to shoehorn 2x 1/2 AA's into an Altoids Smalls can. Because half the fun of this project is using an Altoids!

That's how people know it's an Altoids!

https://youtu.be/c6ffbVd_ff0?t=53

User avatar
adafruit_support_mike
 
Posts: 67391
Joined: Thu Feb 11, 2010 2:51 pm

Re: Would 2x 1/2AA batteries in Parallel Work?

Post by adafruit_support_mike »

It probably wouldn't work, and could easily be dangerous.

The trouble with putting energy sources in parallel is that they have to match almost perfectly. If there's any mismatch between the voltages, the one with higher voltage will try to push current backwards through the one with lower voltage.

That's a problem because batteries are designed to have as little resistance as possible, so large changes in load current produce small changes in the battery's voltage. If you flip that around so the battery is the load, a small change in voltage can draw a large amount of current. To a first approximation, every battery looks like a short circuit to GND to every other battery, and it's easy to get enough discharge to cause overheating and possibly a fire.

Only matched cells can operate safely in parallel, and the only cost-effective way to get them is to choose cells from the same production lot whose production test values match to within suitable tolerances. Trying to find matched cells from those in the market is like trying to choose two people at random and hoping they can run a 5k race within a few seconds of each other.

Energy sources that can work in parallel usually have circuits that monitor the voltage and current to make sure they stay balanced.

User avatar
stinkbutt
 
Posts: 593
Joined: Wed Feb 17, 2010 2:40 am

Re: Would 2x 1/2AA batteries in Parallel Work?

Post by stinkbutt »

Huh. Yeah, I considered that possibility but didn't take it as seriously as perhaps I should have.

What if I just diode it up? Throw a diode in front of each positive terminal only allowing current to flow in the proscribed direction. If cell 1 gets 0.1V ahead of cell 2, well cell 1 will get drawn from exclusively until it drops and we end up with balanced, at least temporarily, cells.

User avatar
adafruit_support_mike
 
Posts: 67391
Joined: Thu Feb 11, 2010 2:51 pm

Re: Would 2x 1/2AA batteries in Parallel Work?

Post by adafruit_support_mike »

Diodes will keep the LiPos from damaging each other, but will make voltage mismatch problems worse.

Diodes have an exponential voltage-to-current curve: increasing the voltage across one by about 60mV sends 10 times as much current through it. A change of 18mV doubles or halves the current.

If you have a 60mV difference between the two LiPos, the one with higher voltage will send 10x as much current to the load as the one with lower voltage. A difference of more than 100mV will essentially cut the cell with lower voltage out of the circuit.

To make matters worse, all silicon devices become more conductive as they get hotter.. the "about 60mV" estimate for 10x as much current drops by about 2mV per degree C. Diodes self-heat as current flows through them, so a diode with more current flowing through it will tend to become more conductive and steal even more current away from the other one. Under worst-case conditions, it leads to a condition called 'thermal runaway' where the hot diode keeps drawing more current until it hits the limit of the power source or the load, or burns itself up.

You can make the current balance better by putting resistors in series with each diode. It's the same principle as using current-limiting resistors with LEDs, and it works, but the quality of your current match is proportional to the ratio of resistor voltage to diode voltage. If they're equal, the currents will match to within about 5%.

At that point, you're losing about 1/3 of the power you get from a LiPo to the resistors and diodes that keep the LiPos from killing each other.

User avatar
stinkbutt
 
Posts: 593
Joined: Wed Feb 17, 2010 2:40 am

Re: Would 2x 1/2AA batteries in Parallel Work?

Post by stinkbutt »

Uh, I'm sorry, I think we've got a Lost in Translation here: Who's talking about LiPo's?

I'm talking about 1/2 AA batteries, which would be either garden-variety alkaline, or perhaps lithiums. Something like this. I'm not looking to use LiPo's. They have their own headaches.

User avatar
adafruit_support_mike
 
Posts: 67391
Joined: Thu Feb 11, 2010 2:51 pm

Re: Would 2x 1/2AA batteries in Parallel Work?

Post by adafruit_support_mike »

I used the term 'LiPo' to refer to lithum-chemistry cells in general.. rechargeable or not.

The information above applies to all power sources, whether they're chemical or entire circuits.

Locked
Please be positive and constructive with your questions and comments.

Return to “MintyBoost”