18650 vs spim08hp pouch cells

[lucasd05]

I've read that the average lifetime cycle count for 18650 cells is about 500, whereas I've seen 2000+ quoted as the lifetime cycle count for spim08hp pouch cells. Is this accurate? Given that, are 18650 power walls (like from salvaged modem packs, medical, or Ring packs) going to fail way sooner than equivalent capacity banks built from other types?

I ask because I'm currently looking at building a battery system for my RV. I have about 1500w of solar panels as a charge source. I'm trying to decide between building my own battery bank using recovered 18650 cells or using prebuilt spim08hp units. Both are currently available on batteryhookup.com - which I'm inclined to use because I'm within driving distance of their PA warehouse, so I could skip the shipping charges.

The per-kwh cost of salvaged 18650 cells seems like half of the cost of currently available pre-built pouch cell modules, even if I have to buy all the needed tools and supplies for such an endeavor (charger, spot welder, nickel strips, holders, etc). Is that a false economy if they are going to fail in a quarter of the time? Or are those 500 cycle estimates based on hard use scenarios that probably won't be the case in solar systems?

I'm really curious to hear what people think about the differences between these two battery types.

 

[Korishan]

average lifetime cycle count for 18650 cells is about 500

You have to take in to considering that means full charge to full discharge, which is usually 4.2V down to about 2.5V. If changing the range to say 4.0V down to 3.2V, the life cycle increases drastically, well over several thousand if the cell isn't too degraded to begin with (aka, fairly new)

are 18650 power walls (like from salvaged modem packs, medical, or Ring packs) going to fail way sooner than equivalent capacity banks built from other types?

All depends on the voltage range you select for your operation, as mentioned above. There are those here on the forum who have over a thousand cycles on "recycled" cells. I believe @hbpowerwall and @daromer both have several years worth of cells still in operation

Both are currently available on batteryhookup.com - which I'm inclined to use because I'm within driving distance of their PA warehouse, so I could skip the shipping charges.

Personally, I would almost lean towards buying the already built system. A lot less likely to fail (either due to error during construction, a rogue cell that goes thermal, bad solder/weld connections that degrade over time, etc). Being inside of an RV increases the necessity of reliability drastically. You do not want to wake up with a lithium battery fire in your RV, and well, you "do" want to wake up period.
Not trying to scare you here, but for beginners, putting something like these in an RV, or other mobile vehicle, is pretty dangerous. It's not like building them for a home, shed, workshop, etc where you can have a dedicated area away from everything to keep things safe is something goes wrong.
I would really recommend going with new, or at least almost new, prebuilt packs for a mobile application, especially one where I'd be sleeping in it.

Or are those 500 cycle estimates based on hard use scenarios that probably won't be the case in solar systems?

The 500 cycles is usually based on full charge/discharged (mentioned earlier) as well as certain amp load criteria. Some are rated at that with only 0.5C (which for a 2500mAH cell would be about 1250mA, or 1.25A). Others are rated at a full 1C. It just depends on how the "manufacturer" quantifies their results, which is why we try to get people to read the datasheet to get familiar with "their" cells that they have in their possession.

I'm really curious to hear what people think about the differences between these two battery types.

I haven't dealt with the spim08hp packs. So not sure what their specs are, or what they are capable of. I'm sure there are others that will chime in on that.
It's also possible you could look at other battery types from BatteryHookup, or other places. DavidPoz on Youtube has several videos on various different battery packs from BatteryHookup. Perhaps you can check out his videos if you haven't already.

 

[lucasd05]

You have to take in to considering that means full charge to full discharge, which is usually 4.2V down to about 2.5V. If changing the range to say 4.0V down to 3.2V, the life cycle increases drastically, well over several thousand if the cell isn't too degraded to begin with (aka, fairly new)

Tracking. That makes sense. It feels like such a simple answer, but I suppose powerbank usage allows for those gentler usage profiles to be selected and controlled by the hardware, in comparison to other settings where a user will basically drain a battery pack until a tool/toy/etc stops running before going back for a recharge.

Not trying to scare you here, but for beginners, putting something like these in an RV, or other mobile vehicle, is pretty dangerous. It's not like building them for a home, shed, workshop, etc where you can have a dedicated area away from everything to keep things safe is something goes wrong.
I would really recommend going with new, or at least almost new, prebuilt packs for a mobile application, especially one where I'd be sleeping in it.

Sure, and I take in the spirit it's intended. I do have this quirk about preferring to not burn to death when possible. Quality control is a big deal when you're sleeping a few feet away from something, and the vibration issues introduced by a mobile platform can lead to issues not encountered in other contexts.

If I do wind up assembling my own packs, it will definitely be with individual cell fusing and liberal use of other safety mechanisms. Given that I'm looking at a bank in the range of a handful of kWh rather than a massive household system, I can implement things that would be impractical on a larger scale. For that matter, I could theoretically bond a thermistor to every single cell and have the pack disconnect itself in the event something goes out of spec. I say "theoretically" because that would be a ridiculous amount of effort and an annoying amount of thermistors to wire up and multiplex into a monitoring system... but a few lost weekends vs a trip to the burn ward? Hmm. It's worth a thought.

I haven't dealt with the spim08hp packs. So not sure what their specs are, or what they are capable of. I'm sure there are others that will chime in on that.
It's also possible you could look at other battery types from BatteryHookup, or other places. DavidPoz on Youtube has several videos on various different battery packs from BatteryHookup. Perhaps you can check out his videos if you haven't already.

I have read some good things about them. The most annoying parts of using them seem to be getting the connections right and building the compression setups needed since they're soft pouches, but that's not an issue if they're available in their original mounts with all the busbars and such included. As you say, I'm sure someone on here can share an experience with them in comparison to other cell types.

 

[Korishan]

I do have this quirk about preferring to not burn to death when possible.

I think most of us have that quirk

Thermistors would be annoying, yeah. Just placing a probe every few cells would be enough to detect any abnormal heat building. Unless the packs are open design to allow air flow between all cells

I can implement things that would be impractical on a larger scale.

This is very true. Your scale is much smaller, and relatively controlled max size is still constrained. So if you spend an extra minute average per cell to add some extra safety, not that big of a deal, as long as you have the spare time available to devote to that.