Spim08hp powerwall


Solar Batteries, Panels, Chargers, Inverters & Mounts at www.batteryhookup.com Save 5% with coupon code "Powerwalls"

Jeb5573ptm

New member
Joined
Mar 5, 2023
Messages
12
Is is possible to build these in 50p configuration?

I initially had the idea of building these in 50p packs at 3.7v and then connecting in series to achieve desired voltage. Is there something I am missing? Would 50p packs become unmanageable with regards to amperage?

Would like to build a home solar storage bank using these cells to go along with the 18650 powerwall packs I am currently building.
 
There is no reason you cannot build pack in the way you are describing -- for instance, 14s50p. You'll need a BMS that can handle that setup, of course, but that's another matter.

Assuming you're looking at the 3.7v 16Ah cells, that's 59Wh per cell. So a 50p block of them will give you 800Ah at 3.7v, or just under 3kwh of storage. Then assuming you build a 14s system, for a 48v system (really 52v nominal, but "48v" is the common nomenclature), that gets you to 52v at 800Ah, or a total battery capacity of about 41kwh. (Usable capacity will be lower depending on how you decide to cycle it.)

Overall, that seems like a very usable battery. It isn't cheap. Battery hookup wants $8.50 per cell, before shipping (and shipping is high, though at least you can freight an order that large). I don't know if they would give you a discount for ordering 700 at once or not. Anyway, you're about $6k into just the cells. You'll also need to sort out bolts, bus bars, and so forth. Figure an additional 5-10% of the cell cost for these bits, if you do it yourself. I'm ballparking you're at about $150 - 160 / kwh of absolute capacity, which seems pretty middle-of-the-road for a DIY setup.

I haven't done this with the Spim08 cells, but it sounds very similar to my Nissan Leaf based powerwall. That's a fully off-grid 14s14p setup, and I have a Batrium BMS on it. (Batrium I wouldn't recommend -- again, that's another matter.) I'd have to think about the dates, but that battery bank has been in service for over two years now, I believe.

How this will interact with paralleled 18650-based packs is another question, and others here are more qualified to answer that than I am. Some feel that the capacities should be comparable, etc. I'll let them chime in with better info.

Cheers, John
 
I would add

Balance: that at 800ah / pack you could benefit from larger than normal balance (amps) capability. Batrium Watchmon + Longmons are at the upper end of the market but should just barely work - https://wiki.batrium.com/product/cell-monitor/longmon - if you're patient.
  • Bypass current 0A to 2A (typically 1A)
My packs are 260ah and my longmons balance them 20-30mv worth within a week. At 50p (800ah) that's 3x the size of mine and so I'd estimate it would take at least a month to balance up to 30mv level differences. Operationally, healthy packs don't need regular balance. In my case, I do touch-ups every 6-9 months to tighten things up by 20-30mv they 'drift'. By extension, with 800ah packs + a month within a 6-9 month window is doable.

Weight: A 260ah pack of 18650 is ~10lbs. So these packs would be ~30lbs each? 30lbs * 14s = 420lbs total. Enough weight to move installation/support considerations to the front of the planning/design list.

Personally - I'd think about staying in the <=500ah / pack range and do multiple batteries in parallel.
 
Last edited:
OGITC makes great points as usual. Even if you wanted to keep the same overall battery bank capacity, you could do 2 parallel14s25p packs. That puts each 25p cell block at 400Ah. Balancing will be a little bit faster, and won't stress a BMS as much. Of course, you'll need more BMS equipment -- cell monitors, etc. The individual cell blocks will also be about half as heavy when you are heaving them around and so forth.

My Leaf-based pack is built in a rolling steel computer server rack cabinet. For perspective, it comprises the full battery packs out of TWO Nissan Leaf vehicles, plus a couple of spare cell modules -- so you're talking about a lot of battery. After stripping off the factory Nissan casings and extraneous equipment, then reconfiguring the modules into the 14s14p setup I needed, plus adding back in the weight of the server cabinet -- I believe the pack's overall weight came in somewhere in the ~750lb range. I can roll it easily on the concrete shop floor, although I rarely have any reason to move it more than a foot or two.

There's a random lesson here... I didn't build my battery in the shop where it is deployed -- which was sort of stupid on my part. (On the other hand, I had power and air tools available in the old shop where I assembled it, and at the time I had no power in the new shop where it now resides...which was the whole point of the project.) When I needed to move the finished battery from the old shop to the new shop, I just used my tractor forks to pick up the entire server cabinet and plop it back down a half-mile or so away in its new location. With the overall weight you are talking about, let me humbly suggest that you plan ahead better than I did, or else that you happen to have the tools and equipment that I was lucky enough to have on hand!! Moving this thing by hand (without fully disassembling it) would very easily be a several-man job.

Finally, just to throw one more sub-thread into the conversation -- have you thought about capacity testing the Spim08 cells before assembling them into larger packs? I'm not sure how "new" or "used" your source would be, or how consistent the cells which you receive will be, etc. There are various schools of thought on this, but the general consensus is that each block of paralleled cells needs to have about the same capacity. (In other words, each of your 25p or 50p blocks in the 14s string needs to have about the same Ah of capacity. We're talking actual / tested capacity here, not rated capacity when new.) There are a couple ways to approach it, but they all basically come down to fully charging each cell and then discharging it, measuring its capacity during discharge. After you have an actual reading for every cell, you can build up your parallel banks so that they are all capacity-matched. Again, sorry to cloud your question, but that's something worth thinking about ahead of time -- because it will require a bit of extra planning, equipment, and considerable time to test as many as 700 cells. If you're already building an 18650-based pack, you are probably already familiar with this line of discussion.

Cheers, John
 
Finally, just to throw one more sub-thread into the conversation -- have you thought about capacity testing the Spim08 cells before assembling them into larger packs?
Just starting to do capacity testing, i have about 50(2x8ah cells) currently. The ones Ive tested are coming out very well at full or just slightly over rated capacity. Will see how they go.
Then assuming you build a 14s system, for a 48v system
This is part of my interest with this cell type. Smaller and more budget friendly than eve cells. I plan to buy them in increments. Have a spreadsheet worked out for shipping breakdown. Budget currently does not permit large quantity purchase, so I will have to see if BHU will discount larger lots.
The bussbars I am looking to make myself, i run a full on cnc machine shop. So that is where I was curious about pack size, will make copper/steel backed strips to suit. I will start with 100 in 4s25p to get to working with them and BMS etc. Ultimately looking to build it to 20 to 50kwh and most likely 48v to go with the inverter/charge controllers.
Personally - I'd think about staying in the <=500ah / pack range and do multiple batteries in parallel.
This is what I was unsure of, how much Battrium or the like could handle. Good point.

Thank you all for the replys, it will be a while but as I progress I will share my findings as it comes together.
 
There are various schools of thought on this, but the general consensus is that each block of paralleled cells needs to have about the same capacity. (In other words, each of your 25p or 50p blocks in the 14s string needs to have about the same Ah of capacity.

This has been rolling around in my head since my last post.

Is there a percentage, a formula used to determine a "balanced" build? I have used the re-packers for 18650's for smaller scale projects. But, left with the question, "would it be ok to build within 1%? 5%? difference in packs based on ah size?"

Seem to recall reading how the bigger P size packs are a little more forgiving when using mixed used cells.

Where is the limit? How do battery manufacturers determine their allowable deviation?

Also, when building a 14s100p battery bank, what are the ways to make matching replacement packs(if replacement becomes needed)? Guessing the same basic rule would apply, match capacities as close as you can.
 
I would strongly recommend that packs in series be close to the same capacity. In my case, my packs are 260ah each and vary as much as 3% with no trouble. You may be able to have more variation - I just never tried it but can confirm that <=3% at the 260ah/pack level is OK.

One way to ensure evenly distributed capacity is to sort your cells in buckets of 100mah different capacities - e.g. 2400-2499mah in one bucket, 2500-2599mah in a 2nd bucket, 2600-2699mah in a 3rd bucket and so on. Then round-robin pull cells from the buckets till they add up the desired capacity of the pack +/- 2%. If you had 3 buckets then each 100p pack would have 33 cells at 2400-2499, 33 cells at 2500-2599, and 33 cells at 2600-2699 and.... just pick one for the final 100th cell - close enough. This round-robin approach tends to average out variations within the 100mah buckets :)

IF you have a little sag after doing all you can to make sure they're all exact... and I've had this happen a few times out of 126 packs I've made... I simply do a 4p or even up to 8p mini-pack and set it on top and attach it via alligator clips wires and the sag disappears. I also leave a few extra spots in my packs so I can go back and permanently add these 4-8 cells once I'm sure it's all stabilized.

@Wolf has done a ton of research into more precise pack building that I've done - he may comment.
 
Last edited:
I know this is an old post, but I just stumbled upon it and had to share.

I've done this. I managed to swing some deals years ago on SPIM08HP cells. I wired them in 4p13s configuration to get 48V and the 52 cells fit very nicely into surplus 40mm ammo cans. I made a total of 7 cans that can run my 48vdc based server rack for a few days on battery alone. All this is fed by some second hand solar panels that can bring the whole thing back to full with one good sunny day.
 

Attachments

  • 20240426_071418.jpg
    20240426_071418.jpg
    616.8 KB · Views: 61
  • 20240426_071423.jpg
    20240426_071423.jpg
    341.6 KB · Views: 49
  • 20240426_071451.jpg
    20240426_071451.jpg
    316.9 KB · Views: 48
I know this is an old post, but I just stumbled upon it and had to share.

Nice project, nice work! I'm curious -- you're possibly the first person I have seen use the SPIM08 cells without their cell holders and associated compression. Do you feel like they are packed tightly enough in the ammo cans to provide adequate compression for the pouch cells? Or do you just figure your draw is low enough that they aren't likely to swell?

Interesting project -- thanks for sharing!

Cheers, John
 
Nice project, nice work! I'm curious -- you're possibly the first person I have seen use the SPIM08 cells without their cell holders and associated compression. Do you feel like they are packed tightly enough in the ammo cans to provide adequate compression for the pouch cells? Or do you just figure your draw is low enough that they aren't likely to swell?

Interesting project -- thanks for sharing!

Cheers, John
The cans do provide a good limit for how much the cells can swell. That said I did have one pack with a cell or two that had gone bad. Was definitely a pain to pull the whole assembly out for rebuild. If I were to build them all again I would loop a nylon cargo strap under the cells.
 
Back
Top