LTO Lithium Titanium Oxide (Titanate) Battery 30S (now 32S)- Seeking the Ultimate Commuter Battery

[SamRich]

Didn't find a lot of LTO batteries builds on ES (or anywhere else really) so I thought I'd share mine.

I decided to build an LTO battery mainly for its very high number of cycles.

I use my Ebike daily as a commuter and therefore run about 250 charge cycles a year and with HobbyKing Lipo packs, that's about how long they last until I get too much sag and need to replace.

LTO are rated for 5000-100000 cycles depending on manufacturer....So I'm hoping for a battery that will last 10+ years!

Edit: Jump below for detailed analysis comparing LTO, Li-ion and LIPO https://endless-sphere.com/forums/viewtopic.php?f=14&t=91976#p1341820)

Being in the North East having cells that also perform well in the cold will be great as I will be able to charge outdoors year round.

Planning to build a 72V nominal / 84V peak battery with a 100A discharge max rating (will only need 60 or so). I also need about 10Ah for a 16 mile roundtrip commute.

LTO charges up to 2.8V as opposed to 4.2V so will need 30S to repalce my 20S Lipo - works out to 84V as well - so can share charger.

Downside is weight - these things are HEAVY! This is partly due to the lower voltage so you need 50% more cells to get the same output voltage. They are about 2-3 times heavier than my LIPO hobby pack with comparable kWh, however they have 2-3X the kW rating of Multistar.

But I'll take the weight penalty for power and reliability.

I ordered 32 11ah Cells from ES member INWO on Ebay. These are NOS AltairNano cells. Here are the Specs from the Ebay page:

When they arrived all 32 cells where within 2.344 - 2.346V except one at 2.277V.
I ordered 2 extra just in case so I won't use the lower one - although I'm sure it's fine.

Here's a PIC of the 32 cells and the case I built to hold the Cells.

The case is built using 6" U channel from McMaster Carr.
Should fit the 9x5" cells snugly.

LTO BMS are much harder to find but found one on Aliexpress for 30S /100A

https://www.aliexpress.com/item/25S-60V-30S-72V-lithium-titanate-battery-BMS-PCM-protection-board-circuit-100A-150A-48V-36V/32835626111.html?spm=a2g0s.9042311.0.0.QQU7PR

However, the BMS has no balancing function (could not find any that did)

I will be balancing with super capacitor balance boards rated for 2.7V
This will require me to have a seperate balance cable that I can plug in externally for this purpose.

Hopefully the cells will stay balanced an I won't have to externally balance too often.

I'd like to hear from anyone else who's built an LTO battery!

 

[madin88]

need to follow this
i never heard about "super capacitor balancing", do you know how this works or do you have a link to this balancer?
about the 30s BMS: have you concidered about building 2 batterie packs each with a BMS and wire it up in series later? 2 x 15s or one 14s + one 16s BMS should be plenty of choices.

 

[billvon]

i never heard about "super capacitor balancing", do you know how this works or do you have a link to this balancer?

Supercapacitors often need balancers. Maxwell includes balancers in their supercapacitor banks; however, usually resistive or active op amp balancers are sufficient. From the looks of the resistors on this balancer it's a clamping, or zener-type balancer.

 

[SamRich]

need to follow this
i never heard about "super capacitor balancing", do you know how this works or do you have a link to this balancer?

The balancers are from ebay https://www.ebay.com/itm/6String-2-...d=252605256645&_trksid=p2047675.c100005.m1851
I should be getting them today so I'll see how it works out....

about the 30s BMS: have you concidered about building 2 batterie packs each with a BMS and wire it up in series later? 2 x 15s or one 14s + one 16s BMS should be plenty of choices.

LTO charges up to 2.8V (Hence the 30s as opposed to 20s) so requires a different BMS than standard Li-ion o LiFe cells.
As far as using 2 BMS in series, yes I have - I have a 14S battery with a matched 6S booster .
It works great but then you have to separate to charge and that would not work for this battery case.

 

[SamRich]

The cells come with Al tabs that require a $100,000 spot welder!
In lieu of that...I drilled and screwed metal tabs to hold the Al tabs together.
The tabs can also be used for support to solder the BMS wires.
The screws are 8-32 1/4" buttons head which gives enough clearance for folding neatly.

 

[madin88]

The balancers are from ebay https://www.ebay.com/itm/6String-2-...d=252605256645&_trksid=p2047675.c100005.m1851
I should be getting them today so I'll see how it works out....

so this is a balancer normally used for connecting supercapacitors in series and you use it for these LTO cells as they have similar max charge voltage?

 

[SamRich]

The balancers are from ebay https://www.ebay.com/itm/6String-2-...d=252605256645&_trksid=p2047675.c100005.m1851
I should be getting them today so I'll see how it works out....

so this is a balancer normally used for connecting supercapacitors in series and you use it for these LTO cells as they have similar max charge voltage?

Yes, that is the plan. It was recommended by member INWO who sold me the cells.

 

[Sunder]

I have built a 11S LTO battery, but for stationary use only. I have a BMS protected LiFePo4 solar storage battery. It holds a huge amount of energy, but in winter, its power deliver capability can leave something to be desired so I have a 30AH LTO pack which can deliver 300A. Beautiful battery, fit for purpose.

Despite being a big fan of the technology (I have drained a cell to literally 0.0v, and still had it come back full capacity, and near new internal resistance), I doubt I would ever use it in a mobile application. As you said, the weight and volume is just too impractical, and you simply don't use the benefits in a mobile application.

Do I need to charge in 6 minutes on a bike? Not really.
Do I need 10k+ cycles on a bike? Not really.
Do I need 200C+ performance on a bike? Not really.
Do I need high temperature durability. Definitely a nice to have.
Do I need sub-zero environment performance. In Australia? Are you nuts, it's 43*C tomorrow.

Turnigy Graphene have a reputation of lasting around 800 cycles by drone users - and they use them hard. Even Turnigy Multistar get a reputation of 200-300 cycles, which says to me if that's all you're getting out if it, you're using them as hard as drone users are - big long blasts of near full C rate use.

I think if you designed well, a LiPo battery would have given you similar performance in a much smaller space and weight, but if those aren't important to you, then yes, LTO would be a superior technology.

 

[SamRich]

I have built a 11S LTO battery, but for stationary use only. I have a BMS protected LiFePo4 solar storage battery. It holds a huge amount of energy, but in winter, its power deliver capability can leave something to be desired so I have a 30AH LTO pack which can deliver 300A. Beautiful battery, fit for purpose.

Despite being a big fan of the technology (I have drained a cell to literally 0.0v, and still had it come back full capacity, and near new internal resistance), I doubt I would ever use it in a mobile application. As you said, the weight and volume is just too impractical, and you simply don't use the benefits in a mobile application.

Do I need to charge in 6 minutes on a bike? Not really.
Do I need 10k+ cycles on a bike? Not really.
Do I need 200C+ performance on a bike? Not really.
Do I need high temperature durability. Definitely a nice to have.
Do I need sub-zero environment performance. In Australia? Are you nuts, it's 43*C tomorrow.

Turnigy Graphene have a reputation of lasting around 800 cycles by drone users - and they use them hard. Even Turnigy Multistar get a reputation of 200-300 cycles, which says to me if that's all you're getting out if it, you're using them as hard as drone users are - big long blasts of near full C rate use.

I think if you designed well, a LiPo battery would have given you similar performance in a much smaller space and weight, but if those aren't important to you, then yes, LTO would be a superior technology.

Those are great points and I appreciate your comments. It will be interesting to see if the weight will be prohibitive for transportation. I can always go back to Lipo if it fails. Like I said, high number of cycles with high power is important to me to reduce cost of ownership.

Battery and tires are the most expensive cost of ownership for an ebike: 4000 miles on a $400 battery is 10 cents a mile. I now plan to get over 40000 miles out of LTO for about $400 in cells and BMS...that's a huge reduction in cost by a factor of 10.

There's no way can get that usage cost with Lipo.

I will also reduce cost of tires by installing moped tires on light Al moped rims. Planning to get 10000 miles instead of 1000 miles out of $40 tires.

 

[SamRich]

Test fit of battery in the box...it's tight.

Luckily I can assemble the box around the battery. I don't think it would slide in and out easily.

I also wired the BMS

The BMS is also a very tight fit! Barely fits between the U channel...I knew it was going to be tight...but the U channel come in a bit inside so the clearance is less than specified by McMasterCarr

 

[SamRich]

I shared this elsewhere on ES but I'll include it here as well for completeness.
Had to modify my frame to fit the battery box:

Gasp!

Brazed and welded 1" square stock:

Here it is installed:
Note that the case also provides additional structure since it is ~1/4" Al plate and U channel

Note how the box also acts as a chainguide

 

[Offroader]

I use LTO 20 amp hour Toshiba SCIB cells in my Honda hybrid car to replace the NiMH batteries. These things are really good. From my testing they really don't need to be balanced. They stay perfectly balanced to .001 volts and I didn't bother with a BMS. You are not using the Toshiba cells so they may not be as good.

These LTO cells will not go up in fire no matter what you do to them. You can also discharge them to 0 volts without much damage, can overcharge them, puncture them, they won't go up in flames.

 

[Offroader]

For those who are interested watch these videos made by Toshiba when they punctured, overcharged cells, and crushed them.

http://www.scib.jp/en/about/index.htm#con1

 

[SamRich]

I use LTO 20 amp hour Toshiba SCIB cells in my Honda hybrid car to replace the NiMH batteries. These things are really good. From my testing they really don't need to be balanced. They stay perfectly balanced to .001 volts and I didn't bother with a BMS. You are not using the Toshiba cells so they may not be as good.

These LTO cells will not go up in fire no matter what you do to them. You can also discharge them to 0 volts without much damage, can overcharge them, puncture them, they won't go up in flames.

Wow that's a nice upgrade. Is that 36S?
I still like the idea of a BMS. I'm hopeful that they will stay balanced since I will otherwise have to balance "Manually" with an external board.
Thanks for sharing

 

[SamRich]

For those who are interested watch these videos made by Toshiba when they punctured, overcharged cells, and crushed them.

http://www.scib.jp/en/about/index.htm#con1

That's insane!

Where/how did you buy these? Expensive?

 

[macribs]

I have built a 11S LTO battery, but for stationary use only. I have a BMS protected LiFePo4 solar storage battery. It holds a huge amount of energy, but in winter, its power deliver capability can leave something to be desired so I have a 30AH LTO pack which can deliver 300A. Beautiful battery, fit for purpose.

Despite being a big fan of the technology (I have drained a cell to literally 0.0v, and still had it come back full capacity, and near new internal resistance), I doubt I would ever use it in a mobile application. As you said, the weight and volume is just too impractical, and you simply don't use the benefits in a mobile application.

Do I need to charge in 6 minutes on a bike? Not really.
Do I need 10k+ cycles on a bike? Not really.
Do I need 200C+ performance on a bike? Not really.
Do I need high temperature durability. Definitely a nice to have.
Do I need sub-zero environment performance. In Australia? Are you nuts, it's 43*C tomorrow.

Turnigy Graphene have a reputation of lasting around 800 cycles by drone users - and they use them hard. Even Turnigy Multistar get a reputation of 200-300 cycles, which says to me if that's all you're getting out if it, you're using them as hard as drone users are - big long blasts of near full C rate use.

I think if you designed well, a LiPo battery would have given you similar performance in a much smaller space and weight, but if those aren't important to you, then yes, LTO would be a superior technology.

Those are great points and I appreciate your comments. It will be interesting to see if the weight will be prohibitive for transportation. I can always go back to Lipo if it fails. Like I said, high number of cycles with high power is important to me to reduce cost of ownership.

Battery and tires are the most expensive cost of ownership for an ebike: 4000 miles on a $400 battery is 10 cents a mile. I now plan to get over 40000 miles out of LTO for about $400 in cells and BMS...that's a huge reduction in cost by a factor of 10.

There's no way can get that usage cost with Lipo.

I will also reduce cost of tires by installing moped tires on light Al moped rims. Planning to get 10000 miles instead of 1000 miles out of $40 tires.

There are a few people here in ES with more then 1200 cycle on Lipo batteries. From memory they build there battery pack with much more capacity then actually needed. And was careful charging and draining cells. If you stay away from hard usage, and dont have prolonged WOT sessions that pulls A close to max that will increase lifespan for pack. The same goes for cut off voltage. Don't discharge to minimum, stop before. And I guess hvc also set to cut off around 85% -90%.

Depending on your riding style and battery TLC you should without any problems at least the 2-3 times the lifespan of Lipo compared to your one year cycle today. Adding more weight by LTO will impact your ride, handling and even your consumption.

Getting motorcycle tires and rims will be a huge difference. You can put a Hookworm to an early grave within a month of hard use. Motorcycle tires are build to last even when used in high torque setting.

 

[SamRich]

There are a few people here in ES with more then 1200 cycle on Lipo batteries. From memory they build there battery pack with much more capacity then actually needed. And was careful charging and draining cells. If you stay away from hard usage, and dont have prolonged WOT sessions that pulls A close to max that will increase lifespan for pack. The same goes for cut off voltage. Don't discharge to minimum, stop before. And I guess hvc also set to cut off around 85% -90%.

Depending on your riding style and battery TLC you should without any problems at least the 2-3 times the lifespan of Lipo compared to your one year cycle today. Adding more weight by LTO will impact your ride, handling and even your consumption.

Getting motorcycle tires and rims will be a huge difference. You can put a Hookworm to an early grave within a month of hard use. Motorcycle tires are build to last even when used in high torque setting.

The issue with building a pack with much more capacity than needed is that you end up with much more weight - of course - and that's the point here. In fact, I would say that you actually do need it, not for energy capacity but for power capacity.

That's what is great about LTO. What it lacks in specific energy it makes up in specific power
It all depends on your requirements.

So I thought I'd put down a detailed analysis for my application - and please tell me if I'm missed anything.

Here are my Requirements:
72V Nominal
60 A Cont /100A peak.
> 10Ah for commute.

Consider these 4 potentialy viable options that meet these requirements on paper.

Value LIPO: Multistar
For my power requirement I would need at least 30Ah (3C)
That's 3 x 5 x 4S 10 ah pack at ~$50 each = ~$750 (on sale now usually ~$65 or ~$1000 per pack)
Weight is 15x800g = 12kg or 26.4 lbs <- heaviest
In theory as you say, if I use it nicely I could get at least 1000 cycles (11Ah per cycle) before I get too much sag at 100A, although in practice it seems to be much less - see PS.
That's 16000 miles per pack or $0.0460 per day.

Quality LIPO: 65C Graphene
For my energy requirement I would need 10Ah (that would be cutting it close - I would probably go with 15Ah, but I'll use 10Ah)
That's 2 x 4 x 5S 5 ah pack at ~$100 each = ~$800
Weight is 8x750g = 6kg or 13 lbs <- lightest
(15Ah would be 20lb and $1200)
Rated for 600 cycle (probably inflated).
That's 10000 miles for the 10ah battery or $0.080 per day. <- Most expensive

Panasonic NCR18650PF 2900mAh 10A
For my power requirement I would need 20S10P
That's 200 cells at $4 each = $800
Weight is 200 x (45.3g ) = 9kg or 20 lbs
I'll be generous and estimate 1000 charges
That's 16000 miles per pack or $0.050 per day.
Did not include $200 welder (that I don't have) and additional considerable cost of holders and strip needed to build this type of pack.

LTO:
For my energy requirement I need 11ah cells.
That's 30 1S 11ah pack @ $9 each = $270 (great deal on Ebay, also available on Aliexpress for $15 a cell)
Weight is 30 x 340g = 10Kg or 22 lbs
According to specs I should get 5000 cycles - but I'll assume 4000 since that would be 90% or 10 Ah.
That's 64000 miles per pack or $0.004 - $0.007 per day (depending on where you buy the LTO). <- Cheapest


Conclusions:
For what I need LTO is in fact comparable weight and a lot cheaper by a factor of 10X.

Panasonic is clearly a close runner up, and probably the winner for most, which is why it is so popular. I would get 3X the range! It was tempting to go that route, but was deterred by having to spot weld the pack properly with 200 cells at 100A rating and the impact on it's long term mechanical and electrical durability. Could also have bought 2X 72V 11Ah packs from Luna Cycle for $1200 but that's a bit steep both short term and long term - wise.

Added bonus 1 of LTO: I don't have to build as many pack and balance as many cells, which I know for some of us is most of the fun
Added bonus 2 of LTO: LTO can be charged outdoors in freezing weather - kinda nice not to have to carry in a 25lbs battery everyday 4 months out of the year.
Added bonus 3 of LTO: I could charge it in 20 min @ 30A - BMS limited, but will probably look at 18A (35 mins) since that's about all I can get from a 120V outlet .

PS I have tried the 30Ah Multistar Lipo combo. At first I was using 20Ah with my 50A/60A contoller and that worked pretty well for a year until I would get sometimes LVC on the way home @8Ah. But then got a new 60A/100A controller so thought I would build a fresh 30Ah pack. Right off the bat, after 5Ah I would hit LVC on hard acceleration which is about 3C. Either need 40Ah (no way!) or probably got a bad pack although I measured each 10Ah pack and IR was 10mOhm per cell on average (compared to my one year old pack that had 20mOhm IR)... anyway I'd have to figure which specific cell was bad and that's a pain... hence why I'm looking for a sustainable/reliable solution...

 

[macribs]

I would say if you feel confident that you will keep your bike much like it is today power wise, not get sucked into the more power craving madness your analysis is spot on and you found yourself a good battery pack using the LTO.

Spoiler: Rant

But as many here knows, it is easy to get hooked on power. Just a little more acceleration. A tad more top speed. A new motor. Or controller. Or even both at the same time. Not to mention we tend to get used to having the power from an e-bike and we grow to know and love to use it.

What I am saying is that great life span of a pack is of high value if you can mustard to keep bike more or less as is. The second you get the cravings for more, the savings you initial calculated goes out the window. Because you might end up using more power then the LTO can deliver. And sooner rather then later that will lead you onto the road of battery replacement to harvest more amps. If such craving comes within the first year or so after building the LTO pack, well you do the math. Then there is the fact that in the next 2-3 years we will have even better cells to choose from, most likely at a lower price point too. Higher C rate and higher Ah. So it will be even easier and cheaper to create hi performing battery pack, and if you kept self control until the next big thing in battery comes along you might find yourself struggle not to jump ship when better batteries are available. Of course you could just keep this commuter bike for its targeted use and build a specific bad ass bike for hooligan riding should such craving occur and then all my random rant is just that. Rant.

If you are the strong determined type that have self control and stick to the commute you might be the rear one that end up keeping the same pack for the next decade.

 

[SamRich]

I would say if you feel confident that you will keep your bike much like it is today power wise, not get sucked into the more power craving madness your analysis is spot on and you found yourself a good battery pack using the LTO.

Haha love the Rant

I agree - it might be hard. I've already gone from 360W to 7500W in just 4 years but at the same time, I feel like I've reached (passed?!) peak power for what is safe for commuting. I don't need to go faster than 50mph or go from 0 to 40mph in less than 4s.

And yeah the N+1 rules always applies: I already have a 1KW Fat bike BBSHD, a 2kW Leaf/winter/Commuter studded bike, 2KW Radwagon, a 4KW Leaf Cruiser/Commuter. So the I think the 7KW LTO will fit right in

The pack did not cost me much so even 2 years of daily use would be cost effective.

BTW: What I save by not having to buy a car for commuting I feel free to "invest" back into new ebike technology (and some crypto) LOL.

 

[SamRich]

Progress Update:

Added Power and Charging wires:

Total weight with case is 33lbs. A bit less than my old 20S30Ah Multistar pack.

First test run today - on stand only - not quite road ready.

It's a bit ugly but I like it

Now I got to take it all apart and paint it!

 

[Buk___]

Test fit of battery in the box...it's tight.

Luckily I can assemble the box around the battery. I don't think it would slide in and out easily.



The BMS is also a very tight fit! Barely fits between the U channel...I knew it was going to be tight...but the U channel come in a bit inside so the clearance is less than specified by McMasterCarr

That channel seems very heavy for the weight contained. Is that a requirement for LTO; or simply a reflection of the materials available/to hand?

 

[macribs]

BTW: What I save by not having to buy a car for commuting I feel free to "invest" back into new ebike technology (and some crypto) LOL.

Very nice sir. Heck I might just do the same over here.

 

[SamRich]

Test fit of battery in the box...it's tight.

Luckily I can assemble the box around the battery. I don't think it would slide in and out easily.



The BMS is also a very tight fit! Barely fits between the U channel...I knew it was going to be tight...but the U channel come in a bit inside so the clearance is less than specified by McMasterCarr

That channel seems very heavy for the weight contained. Is that a requirement for LTO; or simply a reflection of the materials available/to hand?

That's the only U-channel that I could find that was 6" wide. I could have welded a box but this was simpler with the tools at hand.
It is a bit heavy, but the thick walls allow me to tap and screws into them, making it mostly seemless both in and outside - no nuts are used to put together.

The box also provides additional structural support to the open frame

I can take it apart easily for access as opposed to something thinner that probably would have been welded.

 

[Offroader]

For those who are interested watch these videos made by Toshiba when they punctured, overcharged cells, and crushed them.

http://www.scib.jp/en/about/index.htm#con1

That's insane!

Where/how did you buy these? Expensive?

These packs are filled with 24 20 amp hour cells. They are from a Honda Fit electric car. They cost me about $250 shipped for a 24 cell pack.

I use 84 of them in series for around 210 volts in my Honda Insight Hybrid car.

I bought them off Ebay, the problem is that the Honda Fit was a lease only car so not sure if they will ever become widely available. The 20amp hour cells are also too big for an Ebike, but perfect for an electric / hybrid car.

I did lots of testing on them and all 84 cells stay perfectly balanced, I did find one bad cell in a pack and that was all. This is why I didn't bother with a BMS or balancing because they are very consistent and the cells are protected inside the plastic housing. If any of my cells needed to be balanced I would just remove the cell anyway as that would mean the cell is going bad and will get worse fast as I have discovered.

The best way to quickly check if any cells are going bad is to just compare half the pack voltage with the other half, if not exactly the same voltage then a cells is going bad. If any cell was to go bad really fast then it wouldn't catch on fire anyway, so when you eventually find it just remove it, don't waste time with a BMS or balancer it is not needed.

 

[Sunder]

I believe these are either copies, or generics of the same cell. They're certainly in the same packaging:
https://www.aliexpress.com/item/Lithium-titanate-battery-2-4V-20AH-LTO-baterie-for-diy-36V-48V-Large-power-gas-electric/32841478461.html?ws_ab_test=searchweb0_0,searchweb201602_1_10152_10065_10151_10344_10068_10130_10345_10342_10547_10343_51102_10340_10341_10548_10609_10541_10084_10083_10307_10610_10131_10132_10133_10312_10059_10313_10314_10534_100031_10604_10603_10103_10605_10594_10142_10107,searchweb201603_25,ppcSwitch_5&algo_expid=ac6cb14c-ea2a-4fab-8b29-fa445bb75da9-32&algo_pvid=ac6cb14c-ea2a-4fab-8b29-fa445bb75da9&transAbTest=ae803_5&rmStoreLevelAB=5