Battery types - quick comparison

[Jeremy Harris]

Thought this might interest one or two when choosing battery types.

I've been looking at a 36V 20Ah battery pack for a low peak current application and decided it was time to quickly check the current price/weight/capacity situation for readily available (read "eBay".........) battery packs before just buying a Ping. The results were interesting.

Price: The cheapest by far is to go with some relatively low maximum continuous rate (10C) LiPo cells (not a problem in this application, as they will only be discharged at around 0.5C peak, around 0.15 to 0.2C average). Next was NiMh, with the Ping LiFePO4 being by far the most expensive.

Weight: The LiPo cells are the lightest, with the Ping LiFePO4 quite a bit heavier. As expected, the NiMH cells are heavier than LiFePO4, but not by much.

Here's the actual results (sorry for the UK figures, I guess you guys in the US can multiply the price by 1.6 or so to get the $ figure and multiply the weight by 2.204 to get lbs):

£ per Wh:
LiPo £0.25
NiMH £0.48
LiFePO4 £0.72

Wh per kg:
LiPo 176.2
NiMH 66.6
LiFePO4 102.9

Given that there is some evidence that LiPo can be both reliable and safe when used carefully, I have to say the current best choice looks to be LiPo (if you're prepared to mitigate the fire risk). The fact that a 36V, 20Ah LiPo pack only weighs 4.2kg is a substantial advantage for my electric boat project, as I want to have an easily removable battery pack. If the specification for the cheap LiPos is accurate, then this pack would be good for around 200 amps. Pretty impressive for the price.

Jeremy

 

[paultrafalgar]

Thanks, Jeremy, a usefully succinct way of putting it.

 

[John in CR]

Jeremy,

If you can source the Bosch 36V fatpacks on your side of the pond for a price similar to the US Ebay price, then they're definitely worthy of consideration...High current, longer expected life than the Lipo, no BMS required, no special charger, quoted capacity is conservative not questionable, and LiMn has shown to be as safe as LiFePo4. Come join the Konion team, now with new pricing under $.50/wh.

John

 

[Jeremy Harris]

Thanks for the tip, John. I've tried hard to source power tool packs, back when Doc first started stripping hundreds of them for cells. Every service centre I approached claimed that the packs were "hazardous waste" that they weren't able to sell or give away (the daft thing was that I think they had to pay a licensed disposal service to get rid of them!).

The best price that I can find on new Bosch 36V packs here is the equivalent of around $90 plus shipping each, which makes it cheaper for us to buy from the US. Looking at some of the US eBay vendors that will ship to the UK gives an average price, inc shipping, of around $80 per pack, so a bit cheaper. This would still give a £/Wh figure of around £1/Wh, so pretty pricey when compared to the others, at least for us. The advantage of not needing a full BMS is a nice one, though.

Jeremy

 

[Tiberius]

Interesting stuff, Jeremy, thanks.
Can you post a link?

One of the reasons for the rush to LiFePO was bad experiences with the lifetime of other Lithium cells. That's the trouble with batteries; its a serious outlay and you don't know the quality of what you're getting.

I've got the test equipment and I've measured some used packs. Interestingly the only pack I've measured that gives its rated capacity was a NiMH one. The supplier says they get variations in the cells so they down rate the pack.

Nick

 

[Jeremy Harris]

Nick,

The best value NiMH cells I could find were the 3700mAh sub-C ones from http://www.component-shop.co.uk . Their 12V, 3700mAh packs work out at £21.51 each, inc shipping, if you order more than £100 worth.

The LiPo packs are 18.5V, 4000mAh ones from CyberPort888 on eBay. If you hunt around on his store, you can find a double pack deal for the equivalent of £18.26 per pack, inc shipping.

The ping pack I used for comparison was a V2.5 on eBay, I've no doubt that it would be cheaper if you dealt direct with Li Ping, but then the same might be true for the LiPo dealer if you wanted to buy several packs.

I'm not convinced about the lifetime argument with LiPo cells. I have an old laptop that still gives around 90% of it's original life, even after 6 years and several hundred charge/discharge cycles. I've no idea what type of LiPo cells it has, but my guess is that they are quite early generation ones. It's not a premium brand laptop either, it's a Benq, so is unlikely to have anything too fancy.

I think the cells that gave LiPo a bad name were mainly abused RC packs and the absolutely dreadful Thundersky cells that were being sold a few years ago. Once a product gets a tarnished reputation, then it's hard to get rid of.

Jeremy

PS: Did you watch the Red Bull racing this afternoon from Abu Dhabi? Bonhomme was pretty close, but just missed out to Arch. Lamb was miles off the pace in the final, which is a shame. Mind you, he's nearly as old as me so I suspect his reactions aren't too quick now!

 

[Tiberius]

Hi Jeremy,

I only saw a bit of Lamb and not much else - I was supposed to be fixing my camper van (the sink leaks after I fitted a bike rack).
Have you done much flying recently?

I've had two disastrous laptop batteries - one was Digital, the other Toshiba, so big names aren't a guarantee.

Nick

 

[adrian_sm]

It would be great if we had solid data to also factor in the expected life.

The way I look at it the major cost of owning an ebike is the depreciation of the batteries.
So I am not as concerned about the upfront cost, but the total cost of ownership.

Pity this type of life data is the hardest to pin down.

 

[liveforphysics]

Ive still got one of my very first lipo packs. Its a little 3cell 2200mah pack with no brand name, it just says 12C written in sharpie. It cost a premium when i got it like 5 years ago. It must have a minimum of 500 cycles in my favorite little helicopter. It takes 35amp bursts, and at least 20amps minimum when im flying. Since i started charging it at 4C, it gets 2-3 charge/discharge cycles per hour when im out flying somewhere. I keep thinking it will fail, but it seems to perform just as it did when i first got it. I dont overly discharge it though, i set the heli down when the power feels soft, which is generally around 3.4v/cell.

 

[voicecoils]

Hi Jeremy, nice quick comparison.

How did you get your Wh/kg figures? Did you look at bare cell weight, pack weight, or pack + bms + wiring weight?

For my Lifepo4 pack the bare cells should be ~105 Wh/kg but finished pack with bms, wiring, heatshrink, tabs is closer to the 80-90 Wh/kg mark.

I guess what I'm getting at is that bare cell weight allows for good direct comparison of chemistries, but for a given practical application considering the total battery system weight may be useful too. For example if lifepo4 or lipo foil pouch packs require an additional enclosure for safety/protection then it can be considered a necessary part of the overall design which in turn lowers the energy density of battery pack.

 

[paultrafalgar]

How about making Jeremy's first post (at the top of this thread) sticky for the benefit of nubes etc.? Is that possible?

 

[liveforphysics]

Voicecoils- That's another huge advantage of LiPo. They use a 1000times better setup for cell balance connectors and pack connections, and they are all wired into each pack all ready when you get them. I'm certain his weights (and specific densities) are for packs and not just bare cells. If you want to be an extra lazy guy, you can just make some daisy chains and P-groups of deans ultra connectors, and plug the packs right together to have electrically assembled the LiPo pack of your choice in a few minutes. Then, if you want to re-arrange the configuration, or expand the capacity of the pack, it only takes a few minutes to change how things are plugged together.

Also, just thought i would throw this in, the chargers designed for LiPo take a giant steaming dump on the heads of even the best chargers offered for LiFePO4 packs.

Hi Jeremy, nice quick comparison.

How did you get your Wh/kg figures? Did you look at bare cell weight, pack weight, or pack + bms + wiring weight?

For my Lifepo4 pack the bare cells should be ~105 Wh/kg but finished pack with bms, wiring, heatshrink, tabs is closer to the 80-90 Wh/kg mark.

I guess what I'm getting at is that bare cell weight allows for good direct comparison of chemistries, but for a given practical application considering the total battery system weight may be useful too. For example if lifepo4 or lipo foil pouch packs require an additional enclosure for safety/protection then it can be considered a necessary part of the overall design which in turn lowers the energy density of battery pack.

 

[John in CR]

Jeremy,

It looks like the thing to do is become a licensed disposal service, then you get paid to take perfectly good cells and put them back in service. Then you sub out the handling of the cells tested as bad. That or contact a disposal service, which probably has piles of batteries they don't have a clue about handling. That's how Marty got access to 1 or 2 tonnes of used tool packs and laptop batteries.

BTW, why does everything cost so much more on that side of the pond? Is it just taxes?

John

 

[kZs0lt]

Jeremy,

I am using some of the 4Ah cells from CyberPort. Their quality is not the best. They do deliver rated capacity, but they are very far from 10C. More like 2-3C. I tested them discharging at about 10A, but the voltage is pretty low, it is somewhere 3.4-3.5v per cell most of the time at 10A (probably high IR, as it becomes pretty hot at the end). But LiPo is really lightweight. Also note that one cell got ballooned and now it's IR is probably way higher as it's voltage drops quickly as it is loaded. Probably lost capacity too. I am not using it anymore. Don't know what caused it, as I charged it always less than 1C, and was careful not to drain it below 3.0v, but I can't be sure because I didn't have any LVC was just watching overall voltage. And they get well out of balance. So it looks they are pretty fragile, I have about 10-15 cycles on them now. But they were very cheap indeed.

Hope this short review is of help,
Zsolt

 

[liveforphysics]

I wouldn't run a lipo cell that got warm during discharge. That means you are missing out on the main advantage of super low Ri.

For 5ah lipo, I've seen cell Ri range from 3mOhm all the way to 25mOhm per cell. That's why getting decent high-C rating lipo is a good idea to realize all the benefits of LiPo, rather than just low weight and small size.

 

[Jeremy Harris]

Thanks for all the comments.

The weight comparison was for packs, so included the BMS for the LiFePO4. I assumed that the Lipo and NiMH packs would be run as they are on RC applications, with balancing/charging done externally.

I might look into what's involved in becoming a licensed disposer, it could mean getting paid twice for the same thing! I've no idea why prices here are always higher, generally speaking the price difference here is often $1 = £1 for some reason. Sales tax here is 15%, so probably not much different from sales taxes in other countries.

Thanks for the tip about those cheap Lipos from Cyberport888. I wonder if they will be better if they are only discharged at around 0.5C max? My guess is that they might be OK at this rate, in which case they'd be an economic choice for what I have in mind.

I think it is worth reviewing all the available battery options from time to time. Before doing this "quick and dirty" comparison, I was just going to buy another Ping LiFePO4 pack, as I just assumed it was best value. Now I'm not so sure, as it looks very much as if some cheap Lipos might be a better option. Certainly Lipo prices seem to have come down a lot since I last looked.

Jeremy

 

[dogman dan]

I was just thinking, hmm, maybe it would be possible to get licenced to recieve batteries for disposal, and then bulk package them and ship to the recyclers. Well, ship some of em .

 

[kZs0lt]

Thanks for the tip about those cheap Lipos from Cyberport888. I wonder if they will be better if they are only discharged at around 0.5C max? My guess is that they might be OK at this rate, in which case they'd be an economic choice for what I have in mind.

I think they are quite happy at 0.5C. Even at ~5amps they run much more cooler, at 0.5C they won't even get warm. I have posted a discharge graph, see bottom of post here. That was on approx. 10amps, or 2.5C(car H4 halogen light bulb with both low and high beam in parallel, and 0.1amp on discharger, 3 lipo cells series)

Edit: But I still consider them sensitive, as I still don't know for sure what caused the death of one cell, and hyena encountered this once also with a similar LiPo pack.

Zsolt