Battery Cell Recommendation for SAE Formula Hybrid

[arbartz]

Okay, so those of you that are familiar with the SAE Formula Hybrid competition will probably know that the rules are a tad excessive and don't really reflect those in the real world.

Specifically when we have to build our battery pack there are a bunch of rules that make it very difficult for a college team to build. I could go into all the specific rules, but in general they give us these limitations:

• Can't parallel a bunch of cells without having each one individually fused and having the BMS able to detect if any single cell fuse blows.
• Can't use pouch cell Lithium packs unless we meet very stringent regulations on limiting pouch cell expansion and make sure every material used in the pouch cell container is able to withstand 150C before melting, all while being UL94V0 listed.
• Can't really use cells that require welding tabs since we simply don't have the equipment to do so.
• Leaving us with prismatic cells and cylindrical cells that have threaded contacts.

So, given those regulations I've found it extremely difficult to find suitable Lithium cells. The specific chemistry really isn't a concern, just as long as we can meet these pack spec requirements.

We are planning on using an EMRAX 207 motor. If we go with the Medium voltage version, we'll plan around having the following specs:

• Fully charged voltage no greater than 300VDC
• Continuous discharge current capability no less than 160A
• Total pack capacity no greater than 5,561.25Wh when measured at the 2C rate (30min to full discharge)

If we go with the Low Voltage version we'll need these specs:

• Fully charged voltage no greater than 135VDC
• Continuous discharge current capability no less than 400A
• Total pack capacity no greater than 5,561.25Wh when measured at the 2C rate (30min to full discharge)

Right now this leaves me looking at the Headway 38120HP cells, in an 84S2P configuration for the MV motor. They honestly seem to be the only decent option I could find, but I can't be sure that they will hold up to their ratings. So, I had the team buy 2 of them from Headway Headquarters and have some test time scheduled at Rockwell Automation to do load testing on them. We plan on fully discharging them at a bunch of different rates all the way up to their burst rating, only stopping the test if the cell temperature exceeds the rated 60C. (I'll post the results here if you guys are interested)

So, if you guys have any other suggestions for cells to try, let me know, we'll take all the help we can get!

Thank you all for any help!

 

[amberwolf]

You're not gonna get that kind of current out of headway cells, based on results posted here on ES. (not without a lot of sag, anyway, and probably heat buildup)

I'd highly recommend you read around here and see what various cells you want to use have been tested to actually perform at.


You might look into the factory-direct new A123 pouch cells, which there are a number of bolt-together options for here on the forums, and aren't going to swell up if they're packed as the factory specifies. Just don't get the ones from other places, as they often (usually) have cut-down tabs that are going to make it harder for you to put them together into a pack (and possibly other issues).

There are also Leaf modules you could get that might do it.

EIG NMC cells may also work, and have mounting hardware available from the factory to bolt them together, and keep them contained. But you might need a large purchase to do factory direct; if that's not an option you could check with places that have used them in their vehicles and see if they have spares they can sell you, perhaps. There are also some sale threads on ES and elsewhere for them.

 

[Hillhater]

If you have the budget...look up some of the Saft VL- U cells as used by F1 teams !

 

[arbartz]

You're not gonna get that kind of current out of headway cells, based on results posted here on ES. (not without a lot of sag, anyway, and probably heat buildup)

That's sad to hear, well I guess I'll find out soon enough. :/

You might look into the factory-direct new A123 pouch cells, which there are a number of bolt-together options for here on the forums, and aren't going to swell up if they're packed as the factory specifies.

I've looked at the A123 20Ah pouch cells, but they don't have the discharge rate I'd need. I'd have to parallel at least 2 of them, and then that would eat into my Wh allowance, forcing me to drop the number of cells I series up. Aside from that, even if I get a "proven" design of a package for them, I'd still have to prove it in FEA and a bunch of other tests before the judges will accept it. Last year a team that used an Enerdel pack was denied passing tech because the judges felt that the Enerdel packs were not safe enough in terms of construction.

There are also Leaf modules you could get that might do it.

What are "Leaf" modules?

EIG NMC cells may also work, and have mounting hardware available from the factory to bolt them together, and keep them contained. But you might need a large purchase to do factory direct; if that's not an option you could check with places that have used them in their vehicles and see if they have spares they can sell you, perhaps.

I went to their website and did a comparison of their 4 different cell offerings. For our application and with our limitations, they could work (still the same concerns on packaging though), but because of their odd voltage the best configuration I could get would be 107S3P of the T006 cells. That would net us 235.4Vnom (299.6Vmax), 17.1Ah, and 4025.34Wh. While it would work, we have a lot of wasted Wh.

If you have the budget...look up some of the Saft VL- U cells as used by F1 teams !

I would love to, but I have found no places to get them from. I've directly contacted Saft and have heard nothing back. :/


Thanks!

 

[Hillhater]

I've looked at the A123 20Ah pouch cells, but they don't have the discharge rate I'd need......

..what discharge are you needing ? Those cells can output 300+ amps each. ,!

"Leaf" cells are as used in the Nissan Leaf EV .. But they are 30ahr capacity and less output than the A123's.

 

[amberwolf]

I've looked at the A123 20Ah pouch cells, but they don't have the discharge rate I'd need.

Um, if you think headways would do it, well, the A123 pouch cells should be way better than those.

I haven't tested the max discharge capability of the ones in my ebike pack, becuase the BMS on there is a low-amp unit (around 65-80A is the most I've pulled from it for short acceleration bursts of a few seconds so far on the SB Cruiser) That's 3-4C, and they're rated for a lot more than that continously.

I don't recall the exact ratings, but I expect you'd easily be able to use them 1P for either of your proposed setups. I think it's something around 350-380A for <10 seconds, and probably half or more of that continous, for the cell they make now. There's a thread discussing some of that info that wb9k here on ES was posting in.

Aside from that, even if I get a "proven" design of a package for them, I'd still have to prove it in FEA and a bunch of other tests before the judges will accept it. Last year a team that used an Enerdel pack was denied passing tech because the judges felt that the Enerdel packs were not safe enough in terms of construction.

That sounds like they need to setup an "approved design" and simply require you follow that, rather than a bunch of rules about how to go about it, and then disqualify you based on a feeling.

I went to their website and did a comparison of their 4 different cell offerings. For our application and with our limitations, they could work (still the same concerns on packaging though), but because of their odd voltage the best configuration I could get would be 107S3P of the T006 cells. That would net us 235.4Vnom (299.6Vmax), 17.1Ah, and 4025.34Wh. While it would work, we have a lot of wasted Wh.

I was talking about their NMC cells here:
http://www.eigbattery.com/_eng/designer/skin/02/01_03.asp
which I use a pack of (presently on my trike, had been on the bike the last few years till a few days ago). These are the same type of cell used on the Zero motorcycle from a couple of years ago, I think it was, and some other vehicles I can't recall ATM.

This pack I use with no BMS, and draw 65-80A with it, too, for short few-second bursts. That's 3-4C, and they're rated for 5C continous (100A), and can do <10 seconds of 10C (200A), though I don't know what the voltage sag would be at that rate off the top of my head--the spec sheet should list it though.

So for your medium voltage setup, I expect these could handle your acceleration currents, which probably wouldn't be more than 10 seconds at a time for that high a current. For the low voltage setup you'd need 2P.

 

[Punx0r]

Leaf modules would be a gift as they'd be unmodified automotive OEM-grade components.

 

[arbartz]

I've looked at the A123 20Ah pouch cells, but they don't have the discharge rate I'd need......

..what discharge are you needing ? Those cells can output 300+ amps each. ,!

"Leaf" cells are as used in the Nissan Leaf EV .. But they are 30ahr capacity and less output than the A123's.

Perhaps the specs on the A123 website aren't up to date? From what I could find the 20Ah pouch cells are only rated for 3C continuous discharge, putting them at somewhere around 60Amps for a continuous discharge. I'm basing that off of this: http://www.a123systems.com/products-modules-energy.htm

If you feel that is incorrect though, and that they could handle 160Amps continuous, do you know of a vendor that I could get 12S1P assembled packs of the A123 20Ah pouch cells from?

I've looked at the A123 20Ah pouch cells, but they don't have the discharge rate I'd need.

Um, if you think headways would do it, well, the A123 pouch cells should be way better than those.

I haven't tested the max discharge capability of the ones in my ebike pack, becuase the BMS on there is a low-amp unit (around 65-80A is the most I've pulled from it for short acceleration bursts of a few seconds so far on the SB Cruiser) That's 3-4C, and they're rated for a lot more than that continously.

I don't recall the exact ratings, but I expect you'd easily be able to use them 1P for either of your proposed setups. I think it's something around 350-380A for <10 seconds, and probably half or more of that continous, for the cell they make now. There's a thread discussing some of that info that wb9k here on ES was posting in.

Aside from that, even if I get a "proven" design of a package for them, I'd still have to prove it in FEA and a bunch of other tests before the judges will accept it. Last year a team that used an Enerdel pack was denied passing tech because the judges felt that the Enerdel packs were not safe enough in terms of construction.

That sounds like they need to setup an "approved design" and simply require you follow that, rather than a bunch of rules about how to go about it, and then disqualify you based on a feeling.

I went to their website and did a comparison of their 4 different cell offerings. For our application and with our limitations, they could work (still the same concerns on packaging though), but because of their odd voltage the best configuration I could get would be 107S3P of the T006 cells. That would net us 235.4Vnom (299.6Vmax), 17.1Ah, and 4025.34Wh. While it would work, we have a lot of wasted Wh.

I was talking about their NMC cells here:
http://www.eigbattery.com/_eng/designer/skin/02/01_03.asp
which I use a pack of (presently on my trike, had been on the bike the last few years till a few days ago). These are the same type of cell used on the Zero motorcycle from a couple of years ago, I think it was, and some other vehicles I can't recall ATM.

This pack I use with no BMS, and draw 65-80A with it, too, for short few-second bursts. That's 3-4C, and they're rated for 5C continous (100A), and can do <10 seconds of 10C (200A), though I don't know what the voltage sag would be at that rate off the top of my head--the spec sheet should list it though.

So for your medium voltage setup, I expect these could handle your acceleration currents, which probably wouldn't be more than 10 seconds at a time for that high a current. For the low voltage setup you'd need 2P.

I guess I didn't realize that the A123 pouch cells could handle that much current, I thought they were limited to 3C discharge. So if the A123 pouch cells would work, I would need to find a vendor that could make 12S pouch cells packs for me, but given the difficulty of buying A123 stuff for the consumer market I don't know if that's even out there. I could really like to get the modules I listed above as those are from A123 themselves, so there should be no problems getting through tech.

As for the C020 EIG cells, they have a discharge rate that is too low. I would have to have 2 of them in parallel, which would only allow for up to 32 in series before I hit my Wh limit. Which in that 32S2P configuration I'd only be at 116.8V.

Leaf modules would be a gift as they'd be unmodified automotive OEM-grade components.

They would be except they appear to be 60Ah, which is way too high for my purposes. I really can't go above 20Ah if I want the maximum amount of voltage allowed.

Thanks everyone for the help so far!

 

[amberwolf]

It sounds like with your limitations you should probably change your design process, so that you build your battery, get it approved, and then build your controller/motor system around the battery's capabilities so you can still reach the acceleration rates, speeds, etc you are after.

 

[okashira]

http://www.a123systems.com/products-modules-energy.htm

Can you use used cells?

 

[Hillhater]

This A123 document may be useful.
page 64 states the continuous discharge rating at 200 amps, and pulse (10sec) discharge of 600 amps.
https://www.buya123products.com/uploads/vipcase/b24d4f5b63934c59d43e93b3bb4db60a.pdf

 

[arbartz]

It sounds like with your limitations you should probably change your design process, so that you build your battery, get it approved, and then build your controller/motor system around the battery's capabilities so you can still reach the acceleration rates, speeds, etc you are after.

Maybe, but our main goal is capacity. Since everyone in the competition has the same capacity limit, it would stand to reason that in a well designed vehicle, the one with the highest capacity could get the most amount of laps on the endurance run, which is the event worth the most points.

http://www.a123systems.com/products-modules-energy.htm

Can you use used cells?

I would like to use these, but I can't find anywhere that I could buy them from in the configuration I would need.

This A123 document may be useful.
page 64 states the continuous discharge rating at 200 amps, and pulse (10sec) discharge of 600 amps.
https://www.buya123products.com/uploads/vipcase/b24d4f5b63934c59d43e93b3bb4db60a.pdf

Thanks for this document! I have the one for the A123 cylindrical cells, but I didn't have this one.

 

[amberwolf]

Maybe, but our main goal is capacity. Since everyone in the competition has the same capacity limit, it would stand to reason that in a well designed vehicle, the one with the highest capacity could get the most amount of laps on the endurance run, which is the event worth the most points.

Or...the best efficiency, perhaps? Both in drivetrain, and in the vehicle design (aero and rolling resistance, etc).