Re: "Protected" lithium cells

Yes, or 26650 possibly.

Thanks I saw the site before but not the combined charts. The lifetime is equally important for me but that is harder to test of course.

It will definitely be protected, within each cell like those ones or else as part of a UN certified pack. Then there will be my own charging/psu circuitry.

Part of the attraction of the "protected cells" was the possibility of using standard (18650) battery holders rather than something awkward with a shrink-wrapped battery pack. Also they initially appeared cheap and high capacity but as you say that seems unlikely now.

I haven't done much with the 26650 other than build a replacement battery pack for the neighbors eBike. From my really crude testing, the 26650 has a bit more than twice the volume of an 18650 cell, but does NOT produce twice as much power in watt-hrs or joules. I have no idea why.

Do you mean lifetime, as in how many charge cycles the battery will tolerate, or runtime, which is how long it will run before your device quits?

You might want to think about a low voltage cut-out. The problem is that if your device runs on a voltage that is below the safe minimum voltage for your chosen battery chemistry, users will try to run the device until it says quit. The result is premature battery failure and a short lifetime. You'll have to chose between the bragging rights offered by a long battery runtime, versus the economy of a long battery lifetime. I suggest a compromise. Incidentally, that also applies to the maximum charge voltage, where too high a float voltage will prematurely kill the battery. You may not want to leave your batteries fully charged all the time as in many laptops. That will also reduce their lifetime. Many newer laptops have an option to cease charging at 50-60% in order to extend battery life.

Think about using a balance charger, as found in RC (radio control) airplanes, cars, drones, etc. For example: Note that the packaged batteries have designations such as 1S, 2S, 3S,

Also think about using RC standard connectors which are both cheap and can handle the power. I use XT60 for most everything: This is not a trivial issue as these batteries are capable of being charged and discharged at high currents. For example, I use this pack in a drone: It can discharge at 25C * 2.2A-hr = 55 amps and charge at 5C * 2.2A-hr = 11 amps You may not need or want to run the battery at those high current levels, but if your protection circuitry is on the other side of a connector, the connector should be able to handle the necessary current[1].

Once you eliminate the cheap junk cells, I think you'll find that packaged batteries with connectors, are cheaper or the same price as welded cylindrical assemblies. The lower cost is mostly due to these packs using prismatic (flat rectangular) cells, instead of cylindrical. Prismatic cells also save on the cost of battery holders. However, you must leave room for normal battery inflation during charging, or you'll blow the bottom out of your device, as Apple laptops are prone to doing:

Suggestion: With LiIon, forget everything you know about NiCd and NiMH batteries and start from scratch using what the RC and laptop sectors have learned from experience.

Good luck...

Charge cycles. For 18650 battery packs I was looking at the LiFePO4 types. These seem to have half the energy density but claim ~2k cycles vs ~500 for li-ion. They also seem more robust against overcharge (even without the internal protection which the ones I was looking at also have).

The RC ones do look a bit scary. I started there but now looking at this

and

These have internal protection and internal 2A current limiting (in addition to what I do externally).

Thanks!

How about using a nice metal C-clamp to short the battery? You might have to remove paint, oil, grease, etc from the jaws of the clamp first.

Matt Roberds

Done: That's really what it takes to get a connection that works to 10 amps without welding and without any contact resistance, heating, or losses. When I tried using a spring loaded 18650 battery holder at high currents, the holder melted.

That's being all fancy and using wires to relocate the short for analysis. I was talking about using something like this

to provide both contact pressure *and* the short. My guess is that you won't get the steel / cast iron of the clamp to glow, but it'll probably get fairly hot.

Matt Roberds

Methinks you may have missed my intent at suggesting the OP short the battery. The problem is not determining whether the cell will survive, but rather what happens to the connecting terminals, lugs, wires, connectors, switches, fuses, fuse holder, fusible link, protection diodes, and whatever else I forgot. Ever since I had the experience of the fuse failing to blow because of high connection and wiring resistances, I've made it a point of checking to see what happens. Besides, it's also fun. The usual is watching the insulation melt off the wires, while the fuse does nothing.

I (the OP) am looking at these ones which have 2A internal current limiting

And for another product:

I saw those links on one of your previous postings. I can't render even a bad guess if those are suitable batteries since you haven't supplied any electrical, environmental, or functional specifications for your devices. For example, how close is your load and charge currents are they to the 2A over-current protection limit? If the protection circuit trips, and has a foldback feature (as most do), your device(s) will turn off, and stay off until the power is disconnected and reconnected. I can't tell much from the "data sheet".

Interestingly, the 2nd battery pack above has a built in 10K NTC thermistor. This is very useful for initiating an alarm *AFTER* something has destroyed the battery. By the time the battery pack gets hot enough to move the thermistor, it's already damaged or dead. There's also quite a bit of thermal gradient and lag, where the battery is much hotter inside than outside, and where the outside temperature lags behind the inside temperature by several minutes.

Hi Jeff

sure, well my OP was merely asking for info about the protected cells (which I have now decided against). But I do appreciate the discussion.

One can look up the protection circuit IC too.

The 2A is the long-term maximum, it says it will supply 6-8A peaks. That one would be used in a system that takes a fairly constant 500mA.

Still, the packs with built-in protection circuit should provide an extra level of safety, compared to the bare RC modelling ones say.

LiFePO4 are regarded as safer than Li-ion even without such protection. The datasheet specifies safety parameters

So there are two independent systems of electronic protection (my own external circuity plus the built-in PCB), plus the intrinsic properties of the cells themselves. And I would be using the battery at a low fraction of its maximum current ratings.

On 9/6/2015 10:39 PM, John Devereux wrote: snip

Those are a different concept. The BATTERY PACK is protected and from a real vendor with a real factory and hopefully tests/certifications/ quality control in place. That's a different solution from PROTECTED CELLS.

Sounds like a much more rational approach.

Absent some kind of standard, what do you do two years from now when a customer wants to buy a replacement battery and the vendor no longer offers it...or is outa business???

What are your legal/third party certification options when changing the battery vendor or type?

The make or buy decision has a lot to do with your expected volumes and support lifetime...and whether, sometime in the future, Little Johnny gets an owie from a lithium battery and creates a whole new flurry of regulations...for the children...GRRRRR!!

Sure, I understand that. (I have discounted the protected cells idea now). The certified battery pack was always the plan but I was curious about these "protected cells".

Same as anything else I suppose, we look for a suitable replacement or have one made or redesign around something else that is available.

Sounds about right...

Thanks

Ah, OK. I thought whatever internal protection the cell has (or lacks) was of interest, so providing an external short with the lowest possible resistance was the goal. I didn't realize your goal was to check out the battery holder, wiring, fuse, etc.

I guess I start with the pessimistic view that if you didn't fly to Japan / Taiwan / China and watch the cells come off the production line and into your hands, then it's probably some fake cell made as cheaply as possible, and making sure *the cell itself* doesn't blow up is step

1. After that, you can start investigating the rest of the circuitry.

The closest thing I've done was to bench test an APC UPS that had a

Matt Roberds