Suggestions on Batteries with High Volume to Capacity Ratio

Electronics
#23

Unlike so many fleabay/aliexpress buyer beware brands (ala UltraFire), Vapcell tends to make claims proximate to reality - ala they don’t claim 9500mAH 18650 cells (3500mAh about max in that formfactor). That being said I can’t find any reviews of the Vapcell K62. Lgyte-info.dk has exhaustive reviews on a wide range of li-ion cells.

Note that the 18650 formfactor has undergone markedly more development in the market and thus has more off-the-shelf accessories ala chargers, holders, powerbanks.

An off the shelf solution would be to use a modular powerbank that accepts loose cells - and likely several (in parallel, running separate circuits) given the apparently significant power demands of this application.

#24

Thanks! I didn’t know there were prebuilt things like that. Now that I know to use the term “modular powerbank”, they’re way easier to find.
I’ve noticed that a lot of them are just 3D printed holders with no electronics though. So maybe I could just build my own.

I’m not sure where to get started on the protection part, though. Should every cell get it’s own chip? Do I need a single chip with terminals for every battery? Do I just wire all batteries in parallel to a single chip with a high enough current rating?
A review of that Vapcell 26650 says “These are INR batteries that are chemically different than IMR batteries”. I’m not sure if that’s relevant.
Vapcell’s data sheet says they charge at 1A normally or quick charge at 3A. So my thought now is to just find a protection chip that’s rated for 3A per battery.

My ideal solution would be to make a battery pack I can directly charge instead of having to remove every cell. That way, I could have a second pack to quickly swap between.
I’d mainly wear this at conventions, where I’ll spend like 3-4 days in a row wearing it the majority of the day.

#25

Also, I’d intend to not quick charge them at 3A, since I know that reduces the life span. That’s just to cover the worst case.
Since my new power estimate is way lower than my original, I may even try for getting enough batteries where I don’t have to charge them above 80%.

#26

I know you’re excited about building things and I hate convincing a maker to buy instead of build, but since you’re learning, it will probably be a better use of your time, money, and safety to buy a premade bank. You’re not going to be able to beat them in the month until TFF and you’ll still run into other roadblocks with your suit.

After TFF hit me up and we can build some things.

1 Like
#27

You’re right. I should stop focusing on the battery so much right now and actually work on the suit.
As a short term solution, I could just use USB banks. Maybe I’d wire a few USB plugs in parallel. I know you’re not supposed to do that, but it’s probably fine for a single con.
I did that for several years before learning I’m not supposed to and never had issues (except maybe shortening the batteries’ lifespans).

And thank you for the offer! I’ll message you after TFF, since I’d really like to learn more about battery management stuff.

1 Like
#28

Also my “fixed” power estimate is still based on the assumption I’ll need 2 rows of the densest LEDs I have everywhere LEDs are needed, and that they’re going to be full white all the time. It’s probably still 2-6 times as high as it needs to be.

1 Like
#29

Of note: many USB-C battery enclosures have two C ports, are daisy-chainable & are capable of routing power straight through a depleted enclosure. You can use adapters in the linked list below to easily access that power.

Perhaps you could put one USB-C bank hidden in/near each paw so they can easily be changed out during breaks with fresh backup battery packs? You could link all four together via USB-C to one output (100W max). You could duplicate that setup & add an A/B switch so you have a redundant battery system. Probably best to use the 20V USB-C output mode so mainboard wires can stay thinner. You can then have one big 5V regulator, or independent 5V regulators for each LED strip.

I’ve also added a 2.1 amplifier for those speakers (and a hidden sub??)

1 Like
#30

I never thought of chaining power banks together like that. That’s a cool idea!
I wonder if there’s an easy way to test for that routing feature.

My idea would be to fully charge 2 identical ones. Then measure how long it takes for one to die when chained and not chained. Since charging an almost full battery takes much more energy than gets stored, there should be a pretty noticeable difference. But that’s indirect and wouldn’t say for sure.

And thanks for the list! Yesterday, I learned that “USB C trigger” is what you call a board with a female USB C port that can handle PD negotiation. So I just started looking at some. Those cables that have it built in are pretty cool!

1 Like
#31

In that list, I included a USB-C multimeter. You could buy several & place them between battery enclosures to test if/how power is being routed. My hunch is the battery enclosure farthest from the load would be drained first, then the next-farthest, etc. Depends on the PD spec & whether it was properly implemented by the manufacturers. More info here. Don’t forget to add a fuse at the PD trigger output if you go this route. Apparently most of those don’t provide any over-current protection. For 20V, you’ll need a 5A fuse.
If the enclosure farthest from the load is drained first, that means you can probably hot-swap the most extreme drained packs in the chain for new charged ones, without losing power at the load.

If you find you need more than 100W, you could divide the system in half (etc) & add another 100W PD trigger. Alternatively, you could consider using (far less safe) RC LiPo batteries connected in series. I’ve added a few examples to the list, along with self-resetting circuit breaker options.

#32

It’s a very complicated standard and finding a supply that reasonably supports it will be more difficult than just using two or swapping wire.

Lipo charging efficiency is very efficient in the 90s percentage wise. What happens is the amount of current the lithium can absorb tapers off with charge level since the termination is a CV process instead of a CC process (to prevent over volting the lipo). That means the charging rate is slower but it’s not that it takes more energy, it actually takes less.

I would recommend you KISS (keep it stupid simple) and just work with swapping a single battery pack and not go down the very deep rabbit hole of USB Power Deliver spec and strange operation from people not following the very complicated spec.

It’s also worth noting that a “100W” USB-C charger is using it’s most favorable voltage/amperage combo to calculate the maximum wattage. If you are using a PD Trigger to set the voltage to 5V, you’ll still probably only get the 3A max that the supply supports (Maybe 5A if you’re lucky) so if it were me I’d buy more of the 15W powerbanks instead of buying a 100W one and then it operating at 15W anyway.

1 Like
#33

Good points! I’m opining out loud about features that I have not confirmed first-hand work as advertised. I’ll probably do the same proposed experiments (which shouldn’t take too long) in the future for some of my builds. The need to hot-swap tiny safe wearable power supplies in less than 10 seconds off stage left during a scene change will probably grow.

1 Like
#34

Just an update on this:
I decided to drop my goal of TFF, since I don’t want to rush it.
I already have a fursuit with some electronics in it. So I’m pivoting to upgrading that with what I’ve learned for TFF.
I have plans for Anthrocon in July. So that’ll be a more realistic goal for the new one.

1 Like