Primary batteries in parallel

Thread Starter

Ghoster

Joined Jun 5, 2017
3
Hi!

Much has been written on the subject I am about to address. But nonetheless I would like to ask for advice of the more experienced engineers on this forum on a concrete design. My problem has to do with wiring primary, i.e., non-rechargeable batteries (button/coin) in parallel. I've read about the pitfalls and dangers due to unequal voltages and resulting current flowing from one battery into the other.

After trying it out, nothing bad happened. However, since my conclusions on this matter may in the end influence the design of a commercial product, I would like to be sure the risks are next to none.

About the design:
- The desired battery type selection and parallel wiring is unfortunately the most optimal one due to physical constraints, i.e., space available for the battery.
- Currents drawn by the circuit at 3.0 V are going to be between a few uA (around 5 to 10 when device is in sleep mode) and 2 - 3 mA max. When operating, the draw will be about 1 mA on average.
- The idea is that batteries will come in a custom casing (battery pack) so there is no chance the user could insert one full and one empty battery or reverse the polarity of one or both cells.
- The parallel connection of both batteries would only be made once the battery pack would be inserted into the device.
- Adding Shcottky diodes in series with each battery would cut away valuable operating hours and should be avoided.

With above in mind, is simple parallel connection of two coin/button batteries really such a bad idea? Currents will be relatively low. These batteries have relatively high internal resistances (15 Ohms and above). Voltage differences between new batteries are expected to be under 0.3 V (given past experience with these batteries it would more likely be in the range of 0.1 to 0.15 V). This assumption, combined with high IR would mean that equalizing currents are under 20 mA and power dissipated in the more empty battery about 6 mW.

One thing I am not sure about is how much of a problem it would be that the battery wasn't made for reverse current's. In other words, is forcing current into a non-rechargeable battery a problem only due to power dissipation or is it a chemical issue and will cause problems even if dissipated power is small. Will capacity effectively decrease because of that?

Thank you in advance for all your help.
 

Rahulk70

Joined Dec 16, 2016
445
Hi,
I assume you are talking about coin Lithium cells.Always use fresh new batteries when trying to make a parallel connection especially since its primary cells. Also since the circuit draws pretty low current it should last quite long. Now using the same configuration for rechargeable cells also has a bad effect as it ruins the capacity of the first discharged cell as it over discharges, though not as bad as in series configuration.
As for the reverse current I'm not sure but from my experience, the battery with higher potential will try to dump some charge to the more discharged ones, which could result in a leak occasionally or the life of the pack will be reduced significantly.
 
Last edited:

Thread Starter

Ghoster

Joined Jun 5, 2017
3
Hi Rahulk70,
Thanks for your input. I am indeed talking about Lithium cells. Such as CR2032 for example. Or CR1632.
Do you believe that this leak is any less likely with low reverse currents as the ones mentioned in my original post?
 

Rahulk70

Joined Dec 16, 2016
445
Hi Rahulk70,
Thanks for your input. I am indeed talking about Lithium cells. Such as CR2032 for example. Or CR1632.
Do you believe that this leak is any less likely with low reverse currents as the ones mentioned in my original post?
Yes, I understood that since you mentioned its non-rechargeable. Rechargeable ones are those LIR models. From my experience, I've seen plenty of these cells and have never seen them leak. They are used in PC motherboards as CMOS backup cells. They last for more than a decade and I've yet to see any of them leak. But I've seen alkaline button cells leak. I think since these Lithium CR series cells use pure Lithium they seal it pretty good. Also Li is very reactive alkali metal. A leak could damage the circuit easily. Only in some watches that have been exposed to moisture have I seen green colored corrosion on the cells but nothing due to leak so far.Well unless its some cheap brand. Go with Maxell, Panasonic, Energizer etc or if you device is expensive Swiss Renata,Sony etc is the way to go.
 
Last edited:

dl324

Joined Mar 30, 2015
8,769
My problem has to do with wiring primary, i.e., non-rechargeable batteries (button/coin) in parallel. I've read about the pitfalls and dangers due to unequal voltages and resulting current flowing from one battery into the other.

After trying it out, nothing bad happened. However, since my conclusions on this matter may in the end influence the design of a commercial product, I would like to be sure the risks are next to none.
Your empirical data isn't useful without stating the parameters under which your experiment was conducted.

All batteries have some internal resistance. When connecting batteries of similar condition in parallel, that resistance will eventually bring the batteries into equilibrium. Problems will arise when batteries are not closely matched.

The same goes for batteries in series. This is why manufacturers of equipment that uses series and/or parallel battery configurations always advise end users to not mix and match old batteries with new batteries.
 

Thread Starter

Ghoster

Joined Jun 5, 2017
3
When connecting batteries of similar condition in parallel, that resistance will eventually bring the batteries into equilibrium. Problems will arise when batteries are not closely matched.
How will this equilibrium be achieved? Simply by discharging one battery more than the other? Or also by flowing charge from one into the other? I would imagine both will happen. And it is this latter equalizing method that I am concerned about.

How would you define closely matched? Is the 0.3 V (10%) difference at a full battery a "close match"?
 

dl324

Joined Mar 30, 2015
8,769
How will this equilibrium be achieved? Simply by discharging one battery more than the other? Or also by flowing charge from one into the other? I would imagine both will happen. And it is this latter equalizing method that I am concerned about.
Stronger batteries would charge weak ones until they reached equilibrium. That's why manufactures of devices that use multiple series and/or parallel battery always caution you to use fresh batteries and to not mix and match old and new batteries.

Primary batteries are not designed to be recharged, but slight charging will usually be tolerated. The best option is to use batteries that are closely matched. This is what battery pack manufacturers do because they know the performance and lifetime of the pack will be determined by the weakest battery.
How would you define closely matched? Is the 0.3 V (10%) difference at a full battery a "close match"?
The best you can do is fresh batteries from the same package/lot. I wouldn't consider within 10% to be closely matched.
 

ian field

Joined Oct 27, 2012
6,543
Hi Rahulk70,
Thanks for your input. I am indeed talking about Lithium cells. Such as CR2032 for example. Or CR1632.
Do you believe that this leak is any less likely with low reverse currents as the ones mentioned in my original post?
Lead acid, rechargeable lithium and maybe a few other types are OK in parallel, but with most types - especially primary; its a bad idea.

You could eliminate that problem by diode-ORing the cells, but you lose the diode Vf, Shottky barrier is better - but not much.

You get some averaging out when you parallel series stacks, but its still not the best way of doing things.
 
Top