Do I really need to put cells under compression or just be able to withstand expansion?
- Thread starter cmburns
- Start date
featherlite
Solar Addict
When I gaze at your handwriting and locate marks scribed onto the wood, my blood pressure drops and my OCD all but vanishes.
You should see my sketch book and sheet cut plans. I hate waste. A wrong cut…. well I don’t want to talk about it
Zwy
Emperor Of Solar
Keep believing it, there also is fairy dust and magic wands that can be used.
If he was serious about actual testing, it would have 2 sets of cells, one in compression and one set loose. Both in parallel and both at the same date/ batch of manufacture plus also entered into service at the same date. Then capacity test at one year intervals.
You must have missed this comment to this video:
@upnorthandpersonal
2 years agoInstead of just randomly speculating, let's do some math. For one, the EVE datasheet states to expect a cycle life of >=2500 without fixture ('compression', yes, I know...) and >=3500 with. So we already increase our normal degradation by 28% by not having the cells in a fixture. Secondly, find some temperature discharge curves for LiFePO4 for example from Nordkyn et al. You should expect at least a drop of 2% at your current temperatures. Finally, a couple percent calendar aging. This also depends on temperature: the higher the temperature your batteries sit at, the faster the degradation - since you're in sunny hot Australia.... The literature has several models on this, but there are numerous variables I can't account for, but a couple percent is not out of the ordinary over the 4 year life since manufacturing. Based on all of this, 5%, or even 8%, doesn't sound bad at all. It would be good to do a capacity test again at 25C to at least remove that variable...
Hey, I remember writing that. And then I wrote a reply, and Youtube removed it...
Let me put it this way: I put my cells in a fixture. Whether or not it actually makes a difference I think only time will tell. That said, I'm now going into my 6th year with the first of my cells. They have been powering my house 24/7, winter and summer. I have no issues what so ever.
Let me put it this way: I put my cells in a fixture. Whether or not it actually makes a difference I think only time will tell. That said, I'm now going into my 6th year with the first of my cells. They have been powering my house 24/7, winter and summer. I have no issues what so ever.
No, I did not see that comment on the video. Thanks for pointing that out as it brings up the manufacturer's expected life cycle of the cell. It clearly indicates my current batch of uncompressed Eve304ah cells should last the next ~20years. Surely, by then, something much improved than the current best will be common place, and an environmentally friendly way to dispose of the old cells implemented.
Exactly this.
If it doesn't make a difference, great; maybe I wasted a few hours of my time having fun designing a compression fixture that meets the specification. I'm confident that I didn't do any harm by following the manufacturer's recommended procedure.
Heck, a whole lot of people still smoke and overeat as well, even though they probably know it will shorten their "calendar life." What could go wrong?
Bottom line is. Compression is better than no compression, but too much is worse than none. Also a truly rigid box is the worst of them all because as cells naturally grow in time the force will reach ludicrous numbers (thankfully these rigid cases are usually from thin sheet steel and such so they just deform).
The hotter your climate the less you gain by compression. At some point it is just not worth the risk of doing it (possibly badly) for a gain you'll not see anyway.
However, having said that it does make a difference even in stationary systems. This is why manufacturers focused on ESS like EVE have been recommending it for quite a while. There is an obvious physical principle at play here. Cells expand and contract during each cycle, but the electrode material is not very elastic. It delaminates and crumbles in time if you allow it to move like this.
However if you live in a tropical climate with routine ambient temps of 35C/95F plus. Or you have older cells that are already somewhat swollen. There is little reason to compress. In the first instance calendar aging will deteriorate the cells before you manage to use up these half a dozen thousand cycles. In the second case delamination has probably already started, by squeezing the cells forcefully you can do more harm than good.
Another situation where I'm not sure compression is worth it is in backup power. Where you might get 15 cycles per year... The balance of cost/benefit is probably towards doing the minimum possible.
If you choose to compress use something compliant. Many people use springs. Special foam is better because there is less margin of error.
The hotter your climate the less you gain by compression. At some point it is just not worth the risk of doing it (possibly badly) for a gain you'll not see anyway.
However, having said that it does make a difference even in stationary systems. This is why manufacturers focused on ESS like EVE have been recommending it for quite a while. There is an obvious physical principle at play here. Cells expand and contract during each cycle, but the electrode material is not very elastic. It delaminates and crumbles in time if you allow it to move like this.
However if you live in a tropical climate with routine ambient temps of 35C/95F plus. Or you have older cells that are already somewhat swollen. There is little reason to compress. In the first instance calendar aging will deteriorate the cells before you manage to use up these half a dozen thousand cycles. In the second case delamination has probably already started, by squeezing the cells forcefully you can do more harm than good.
Another situation where I'm not sure compression is worth it is in backup power. Where you might get 15 cycles per year... The balance of cost/benefit is probably towards doing the minimum possible.
If you choose to compress use something compliant. Many people use springs. Special foam is better because there is less margin of error.
OffGridInTheCity
Solar Wizard
I'm 18650 INR type chemistry rather than LifePo4 but I'm in year 7 and my opinion is growing that *low stress* (very low C, moderate DOD, moderate operating voltage ranges, moderate ambient temps, no physical issues) is a huge factor that's not well quantified in many of these speculations.
From my perspective, there were many 'what ifs' around 18650 4.2v chemistry when I started but they just haven't materialized for me. I'm over 2,000 cycles with 44% of my powerwall with no sign of degradation. The off-grid solar powerwall operates at <0.06C, 37% DOD within 3.5-4.0v range, 55F to 75F ambient - all low stress.
This makes me sympathetic to @timselectric posts - e.g. moderate compression won't hurt and can offer physical stability for practical reasons but perhaps moderate operations / less worry is worth considering for fixed, low stress solar operations.
At this rate it will take me another 14 years to reach 6,000 cycles at which point I'll probably be working on my DIY home fusion reactor
From my perspective, there were many 'what ifs' around 18650 4.2v chemistry when I started but they just haven't materialized for me. I'm over 2,000 cycles with 44% of my powerwall with no sign of degradation. The off-grid solar powerwall operates at <0.06C, 37% DOD within 3.5-4.0v range, 55F to 75F ambient - all low stress.
This makes me sympathetic to @timselectric posts - e.g. moderate compression won't hurt and can offer physical stability for practical reasons but perhaps moderate operations / less worry is worth considering for fixed, low stress solar operations.
At this rate it will take me another 14 years to reach 6,000 cycles at which point I'll probably be working on my DIY home fusion reactor
Last edited:
C'mon, when you're "low stress" you can do anything(within reason) and it'll be fine. The flaking and delamination I spoke of is a function of how quickly the growth happens(how quick you charge) and the temperature. Both things are affected by the current.I'm 18650 INR type chemistry rather than LifePo4 but I'm in year 7 and my opinion is growing that *low stress* (very low C, moderate DOD, moderate operating voltage ranges, moderate ambient temps, no physical issues) is a huge factor that's not well quantified in many of these speculations.
From my perspective, there were many 'what ifs' around 18650 4.2v chemistry when I started but they just haven't materialized for me. I'm over 2,000 cycles with 44% of my powerwall with no sign of degradation. The off-grid solar powerwall operates at <0.06C, 37% DOD within 3.5-4.0v range, 55F to 75F ambient - all low stress.
View attachment 248894
This makes me sympathetic to @timselectric posts - e.g. moderate compression won't hurt and can offer physical stability for practical reasons but perhaps moderate operations / less worry is worth considering for fixed, low stress solar operations.
At this rate it will take me another 14 years to reach 6,000 cycles at which point I'll probably be working on my DIY home fusion reactor
Also 18650 due to being in a circular can have their own built in "compression" so there is zero equivalent in that regard. Pouch and prismatic are very different mechanically. Also depending on exact cell nmc tolerates far more C so if you're let's say 0.06C with NMC it's probably something akin to 0.01C with LFP. No surprise it will last forever. If one can be happy with it, great, but I argue it's not what the most people want.
The thing with being "low stress" is that it only works well in certain situations. For example, now that I have a fairly large battery "low stress" just happens as a function of the size. I designed my system back when I thought I'm not going to use it for more than a whole house UPS and ~14kWh (16s 280ah) was enough for a day of work from home. But I still wanted quick recharge times from solar so I wanted 140A charging(0.5C). Now my battery is 3x the size (4th shelf still not connected, but coming soon) and I'm still happy with 140A charging... But If you told me I'll spend $14k by the end of it on my system I'd laugh you out of the room
What I'm trying to say here is that if you told me in the beginning I'd have to have a 0.15C system to achieve longevity I wouldn't even start.
Edit: Just to clarify. The original question was. "Do I have to compress". My answer is no, you don't. You're going to be fine without it. But if you do, and you don't screw it up, and your battery lives in temperatures where it's not too hot you can benefit from doing compression. Whether it is worth for you personally, that's a subjective decision. So I'm not going to fault anyone for choosing not to.
Zwy
Emperor Of Solar
One big thing for me is stress on cell terminals. Without compression, you can end up with cells slightly bloated like Andy had in the video shown in post #53.
My battery bank in my house has been in place for over 18 months, I don't have those issues shown in the video. I do have to wonder how many fires have been started by expanding cells in uncompressed fixtures. Probably more than many realize.
My battery bank in my house has been in place for over 18 months, I don't have those issues shown in the video. I do have to wonder how many fires have been started by expanding cells in uncompressed fixtures. Probably more than many realize.
Charge rate. Discharge rate. Cycle count. Calendar time. How high in SOC you max go. How low in SOC you go. Temperature. Compression force. Vibrations vs stationary.
Many factors can potentially affect life span. Hard to tell which ones matter the most. I'm just an amateur, but I'd guess temperature, charge rate, and how high in SOC you go as the top three.
Many factors can potentially affect life span. Hard to tell which ones matter the most. I'm just an amateur, but I'd guess temperature, charge rate, and how high in SOC you go as the top three.
They don't tell us. L It may very well be pressure dependant, but I doubt as much as 1.3mm would happen even with no pressure at all. (all my testing is done with a compression fixture).
The only datasheet data we do get about swelling is about pressure exerted during cell aging related swelling.
It is useful to think about two kinds of swelling. One (the one we're talking about here) is just the fact lithium takes more space when it goes into graphite (and lifepo to lesser extent). This kind of swelling, that happens during normal charging is perfectly fine and unavoidable. But it is not by a lot.
They tell us to constrain it, so the "growing" graphite (and the other side to lesser extent) doesn't flake and delaminate. But it doesn't happen fast.
Then you have another type of swelling. I read it is driven by: excessive sei film growth (normal sei growth is a part of cell formation), and to far lesser extent gas pressure caused by electrolyte getting destroyed. This is a "calendar" long term process. But the actual force of it is huge (we can check in EVE datasheet). So there is no constraining it.
It is that other type that happens when you overcharge.
Gadwall*Drake
Solar Enthusiast
Lucky you, being able to source CCA cellulose material. We just have the orange variety around here. Better than no pressure treatment, but definitely not as good.
Similar threads
