What about 25500s like these, are they really 3700 mAh? two in series would be enough and no need for pcb. Asking because some more old dead batteries surfaced that needs TLC.
From DX:
- Brand: TrustFire
- Model: 25500
- Voltage: 3.7V
- Capacity: 4000mAh (actual 3000mAh)
Skip the Trustfire cells.
The protection board only knows how many cells in series, not parallel. This does mean that one bad cell can cause certain undesired effects. In any case it also means that moving from 2s3p to 2s2p is no problem as long as the new cells can handle as much current in 2p as the old ones did in 3p.
Thanks a lot wight, think i got it all sorted out now. Will post pics after assembly and tests.
Any thoughts about these 2 PCM’s for a Efest IMR 26500 2s2p setup? Guess it’s up to how much the strobe draws? Original PCB is replaced in Ikelites new generation 2 packs because of issues with cell life.
Not balanced? 8A
They also replaced the gen1 charger which is a standard 8.4V 1A.
Alternative charger we’re thinking of using: 8.4V 2A 18W
2A might put a strain on the cells but it will charge in 3 hours.
- 0.33C charging seems reasonable for high drain cells - probably overly cautious really. Someone else can confirm or refute this, but I think that many high drain cells are rated for regular charging at 0.5C and some should be rated for at least 1C charging with reduced cell life… right?
- Unfortunately I’m in no position to recommend a particular BMS. It’s a scary concept to me actually - this is a board we are going to put a lot of trust into, so what manufacturer / part is worthy of that trust? Random AliExpress / eBay sellers seems like a stretch… I’m not saying that I wouldn’t go that route myself. I’d just strongly prefer a known-good solution if I could get one! Built-in balancing in the BMS definitely seems like a requirement when using a 2-wire charger though. In fact, this could turn into the bigger issue w/ a fast charger - if part of the pack becomes severely out of balance and the BMS has fairly slow balancing some cells could still be consistently overcharged before the balancer can take corrective action.
I’m not sure how much the strobe unit is drawing. My back of the napkin math goes like this: 22340 must be crap are verified to be crap: 3900mAh from 3p means 1.3Ah/ea. That’s terrible! Anyway supposedly the recycle time is 1.4s and it lasts for 450 flashes, so that’s 630s or 10.5min or 0.175 hours. If we are capable of consuming 3.9Ah in 0.175 hours that works out to about 22.3 amps average over this hypothetical 10.5min period of solid flash use. Peak draw charging the cap is probably a lot higher than that… I have no idea how much. Let’s call it double to triple, so maybe 60A. No wonder they’re destroying these packs! Start with an almost certainly low quality custom made cell and then add on repeated 60A peaks. Recipe for disaster. With NiCad and NiMH this would not be an issue - they just sag a lot and current drops. Upgrading the light to Li-Ion was probably a big change since they previously needed no input current limiting.
My math is pretty full of assumptions, but I consider the conclusions sound. Most non-18650 cells are potentially… lackluster for this sort of application (ultra-high-drain). Efest sounds like a decent bet, but might still suffer from shortened lifespan from this use pattern. Repeatedly hooking big, empty capacitors up to a battery is probably always going to be a mean thing to do.
Thanks a lot Wight, again and again!
First pack is finished now at least and will be interesting to see how long it lasts. First test in water was only with the built in 5W aiming/focuslight on. In 8 hrs pack went from 8.3 to 7.3V
Will go back to Ni-MH if new pack doesn’t deliver but will be a drag weight and charge wise.
Ikelite gave up again, seems to be a shortage of shitty 22430s
Maybe time to enter the 21st century and upgrade the case to fit 18650s.
“Due to long term shortages in lithium ion cells, we have discontinued the sale of lithium ion battery packs for DS-series strobes. Our equivalent NiMH battery pack continues to be available and provides virtually the same recycle time with an unrivaled durability and affordability.”
Efest 26500 has proven good on low current draws but capacity diminishes fast on high.
Looking into NiMh again and found some:
Thinking of going LiPo but it’s hard to find nice capacity cells that are not bigger than 50mm on longest side. Optimum size would be 50X25X??mm (thickness doesn’t matter).
One pack been extended 15mm to hold 8 LG 18650 HG2 in 4s2p. No BMS as none can be found that will handle the current, cells are monitored after each charge.