Bob B
Emperor Of Solar
- Joined
- Sep 21, 2019
- Messages
- 9,165
Here's one .... there is quite a bit of stuff on YouTube. https://www.caraudio.com/threads/lithium-titanate-lto-battery-bank-build.600099/
Anyone Use LTO for a Starting Battery?
- Thread starter Bob B
- Start date
Here's one .... there is quite a bit of stuff on YouTube. https://www.caraudio.com/threads/lithium-titanate-lto-battery-bank-build.600099/
I am looking to use a LTO pack without a BMS as a starter, in parallel with a lifepo4 for the RV service, but I wonder about those peak currents you mention, also about the possibility of using them at low temperatures without heating.
I don't know if the problem is the manufacturer being very conservative in his specifications.
what the manufacturer state is 10C (400A) for 10 seconds and no more than 1C below 0ºC
I don't know if the problem is the manufacturer being very conservative in his specifications.
what the manufacturer state is 10C (400A) for 10 seconds and no more than 1C below 0ºC
mcarper
New Member
- Joined
- Oct 19, 2020
- Messages
- 13
Honestly if you already have Lifepo4 cells I'd put a RV grey tank heater pad on it and just use a starting cap. Just charge the cap via a resistor otherwise it'll probably blow the fuse when you hook it up. You might need more than 1 set of starting caps depending on your lifepo4 pack already in the RV so the BMS doesn't get overloaded if the caps can't keep the voltage high. If you have something like a 200a BMS then my guess is 1 cap pack is all you'd need to keep the voltage up.
240.0US $ |16v 15v 3000f Super Capacitors Module Start Power Motor 6x 2.7v 3000f Super Farad Capacitor Dc Rectifier Car Battery - Rechargeable Batteries - AliExpress
Smarter Shopping, Better Living! Aliexpress.com
www.aliexpress.com
You could use a battery isolator and just draw current for starting from the cap bank like here
Do you leave the balancer on all the time? I guess it doesn't really matter at a max of 14.5 volts. It's more of a "middle balance" vs. top or bottom.
Yes, I use the capacitor balancer, they use less electricity. 2.4V is @ 90% of the power range
Super interesting thread. I'd be really stoked to go LTO for my vehicles, and I've got a truck that needs a replacement right now.
But after looking into LTO voltage curves I'm getting pretty confused. For example on a 6S configuration a typical alternator charge voltage of 14.2V would be 2.37V/cell. According to some sources this would be about 77% SOC:
But according to most of the other charts available it would be much lower - on the order of 5- 25%:
I have a really hard time believing people are successfully running these batteries at 25% SOC. Seems like it would lead to a lot of dead batteries if the lights got left on for a few minutes, and that 10C discharge curve has quite the drop from what open circuit would be. So are those charge curves just wrong? Is there that much variation in LTO cells? Because 6S is obviously working fine for at least one person.
Anyway, the next option might be to go to a 5S configuration but that would put the cell voltage at 2.84V when the alternator is banging away at 14.2V, and that assumes perfectly balanced cells. Most spec sheets say 2.8V as the BMS cutoff, but they range between 2.7 and 3.0V.
And I can't find anything to indicate what the consequence of slightly overvolting LTO really is. Is it significant lithium plating every moment the alternator is charging? Is it just a possible 10% loss in the absurd cycle count?
I get that LTO is a less used chemistry than LFP or LA, but it shouldn't be this hard to find info on one of the most basic failure modes for battery care, and there shouldn't be this much variation in the reported SOC voltage.
But after looking into LTO voltage curves I'm getting pretty confused. For example on a 6S configuration a typical alternator charge voltage of 14.2V would be 2.37V/cell. According to some sources this would be about 77% SOC:
But according to most of the other charts available it would be much lower - on the order of 5- 25%:
I have a really hard time believing people are successfully running these batteries at 25% SOC. Seems like it would lead to a lot of dead batteries if the lights got left on for a few minutes, and that 10C discharge curve has quite the drop from what open circuit would be. So are those charge curves just wrong? Is there that much variation in LTO cells? Because 6S is obviously working fine for at least one person.
Anyway, the next option might be to go to a 5S configuration but that would put the cell voltage at 2.84V when the alternator is banging away at 14.2V, and that assumes perfectly balanced cells. Most spec sheets say 2.8V as the BMS cutoff, but they range between 2.7 and 3.0V.
And I can't find anything to indicate what the consequence of slightly overvolting LTO really is. Is it significant lithium plating every moment the alternator is charging? Is it just a possible 10% loss in the absurd cycle count?
I get that LTO is a less used chemistry than LFP or LA, but it shouldn't be this hard to find info on one of the most basic failure modes for battery care, and there shouldn't be this much variation in the reported SOC voltage.
I don't have any real answers for you, but other than the car audio guys and jimjr's success here are some of the reasons why I'm giving it a go despite seeing the same charts as you.
There are difference s between brands, say Toshiba and Yinlong. It looks like Yinlong does better in 6s configurations.
The charge processes for testing rarely have extended float times. I figure with my normal driving and usage it will be almost all float.
I have completely oversized the batteries and have an lfp bank for "house" usage. Best I could figure I need two 6s2p batteries to get over 1000c cranking current. That puts me at 160ah for just running the headlights and dash. I even have outside scene lights running on the lfp bank.
At this point it's still tbd. I'm hopeful though. My fla batteries are on borrowed time.
There are difference s between brands, say Toshiba and Yinlong. It looks like Yinlong does better in 6s configurations.
The charge processes for testing rarely have extended float times. I figure with my normal driving and usage it will be almost all float.
I have completely oversized the batteries and have an lfp bank for "house" usage. Best I could figure I need two 6s2p batteries to get over 1000c cranking current. That puts me at 160ah for just running the headlights and dash. I even have outside scene lights running on the lfp bank.
At this point it's still tbd. I'm hopeful though. My fla batteries are on borrowed time.
A bit more research found a research paper showing 3.2V/cell as the point where damage occurs.
www.researchgate.net
Which is 16v in a 5S configuration, and typical alternators don’t come close to that. So it seems like a 5S configuration has no known disadvantage other than technically edging past a data sheet limit. Meanwhile a 6S will lose 25% capacity in the best case, and costs 20% more to boot. Seems like an easy choice.
@Will Prowse you recommend 6S above. But also mention 3.2V in one of your LTO videos. Is there anything I’m not thinking of that pushes you to 6S?
Overcharge durability of Li4Ti5O12 based lithium-ion batteries at low temperature | Request PDF
Request PDF | Overcharge durability of Li4Ti5O12 based lithium-ion batteries at low temperature | The decrease in capacity at low temperature limits further application of lithium-ion batteries. This paper introduces an approach to compensating... | Find, read and cite all the research you need...
www.researchgate.net
Which is 16v in a 5S configuration, and typical alternators don’t come close to that. So it seems like a 5S configuration has no known disadvantage other than technically edging past a data sheet limit. Meanwhile a 6S will lose 25% capacity in the best case, and costs 20% more to boot. Seems like an easy choice.
@Will Prowse you recommend 6S above. But also mention 3.2V in one of your LTO videos. Is there anything I’m not thinking of that pushes you to 6S?
Has anyone actually reported trying Yinlong in a 5S config? At this point we can only say Yinglong works well with 6S, but we don’t know if that’s the best.
So you’re using the house battery for “headlights and dash”? Or are you getting 80ah LTO cells somewhere? If not, what size cells are you using for the starting?
These 20AH are rated for 25C, so a 2P config would get you to 1000:
Edit: peak is actually 75C, so they’re rated for a full 1500A.
2.3V 20Ah 25C Prismatic LTO cells | SRIKO Batteries | Lithium ion battery manufacturers US | Top Lithium battery Companies US
TS 2.3V 20Ah 75C peak discharge Lithium Titanate Cell
They also make a 10ah in the same size rated for 75C. Interestingly Will had batter luck when he tested this style compared to yinglong. If I do go LTO those will probably be what I go with.
I haven't tested anything yet, I'm still waiting on the cells. I ordered 24 40ah Yinlong cells. It should be enough for the worst case scenario. The only thing to "leave on" would be the dash and headlights. This will be a starter battery.Has anyone actually reported trying Yinlong in a 5S config? At this point we can only say Yinglong works well with 6S, but we don’t know if that’s the best.
So you’re using the house battery for “headlights and dash”? Or are you getting 80ah LTO cells somewhere? If not, what size cells are you using for the starting?
These 20AH are rated for 25C, so a 2P config would get you to 1000:
Edit: peak is actually 75C, so they’re rated for a full 1500A.
2.3V 20Ah 25C Prismatic LTO cells | SRIKO Batteries | Lithium ion battery manufacturers US | Top Lithium battery Companies US
TS 2.3V 20Ah 75C peak discharge Lithium Titanate Cellsrikobatteries.com
They also make a 10ah in the same size rated for 75C. Interestingly Will had batter luck when he tested this style compared to yinglong. If I do go LTO those will probably be what I go with.
But what I was trying to say is that Yinlong are better suited for 6s for 12v systems than some of the higher voltage cells. And as of right now I have the cells and capacity overhead.
Some of the prismatics weren't as robust, but I didn't see the one you linked. It does look interesting though. If you try it, let us know.
mcarper
New Member
- Joined
- Oct 19, 2020
- Messages
- 13
One of the charts is a charge curve and the other is a discharge curve, thus the 25% and 75% roughly.
If LTO can do 3v per cell without damage then 5s with a big active balancer is all you'd need it seems. I'd be too chicken to do that though; 6s is the safest way and you only loose a little bit of capacity.
Jesus. When you said you oversized it you weren’t kidding. That could do 4000A.
What kind of balancing/management solution are you planning?
I have a 5L 2014 F-150 with a 30ah 6S1P battery in it. With everything on, even with the key on it supplied 40 minuets of power to 11volts @1.9v / cell and started the motor. I had it tested at a auto store and they said it was 1000 CCA . A 40ah cell will give 1600 CCA with using copper bus bars. I also have a 3a Toshiba setup that will do a v-8 Mustang.
Starting a motor only requires few watts for a few seconds, but it needs them in a hurry at the high enough voltage.
As for 3v being safe , Its like being on a edge of a cliff , do not keep there too long , the electrolyte will start gassing.
JimJr
With copper buss bars 1600 x 4 = 6400
Let's hope it works like that. If it does, I'll use the other 6s2p battery on my boat. I found a lot of conflicting spec sheets, but I think they are only rated for 10c(400a) for 10s.
I bought some bus bars from xs, they are nice but just aluminum. If they don't work out ill probably just make some out of copper stock.
I bought some bus bars from xs, they are nice but just aluminum. If they don't work out ill probably just make some out of copper stock.
The confusing part though is the charge vs discharge relationship is backwards. Normally, when the battery is charging at a given voltage it is at a lower SOC than if it is at the same voltage discharging. But the charts I found show the opposite - the charge curve is 75%, the discharge curve is 25%. Which only means the two charts are showing even more difference than at first glance.
You can stand on the edge of a cliff forever. The entire point is that going over is a sudden change. But I'd like to know why you say extended periods at 3v will gas. If true it would be important to know.
I guess the proof is in the pudding. Apparently the truck's alternator charging voltage (you wouldn't happen to know it would you?) is getting it to something like 66% SOC (assumes 120w for high beams, 100w low beams, 20w for the rest of the truck, and 12V). Which fits nicely with the first curve.
Found another source with a charge and discharge curve that seems to also say 2.4 is fine:
Fig. 1. Discharge/charge characteristics of the LTO/NCA pouch cell. a)...
Download scientific diagram | Discharge/charge characteristics of the LTO/NCA pouch cell. a) Different Crates at 20 C ambient temperature; b) 4C at different ambient temperatures; c) 0.1C at different ambient temperatures. Note the cells were first charged then discharged, both with CCCV...
www.researchgate.net
So I guess those other curves are wrong or irrelevant somehow.
My first test 6s was a 30ah with 2 cells that lost @20% capacity per month , the battery was equipped with Qemm equalizer and a led volt read out. The battery was put together with steel nuts, no nolox (to prevent oxidation) , phosphor copper buss bars and copper terminals . Oh I forgot lots of duct tape. The tractor had a 8L diesel motor (no glow plugs ) and a 35 ah alternator , Used 3 times a day and parked under a roof only for @3 years. It died when the operator left the key on until the next day and then put a "dumb charger" on Start setting (plus 40 volts) and came back too late. One of the self dis charger cells had blown its safety valve and lost most of its electrolyte.The confusing part though is the charge vs discharge relationship is backwards. Normally, when the battery is charging at a given voltage it is at a lower SOC than if it is at the same voltage discharging. But the charts I found show the opposite - the charge curve is 75%, the discharge curve is 25%. Which only means the two charts are showing even more difference than at first glance.
You can stand on the edge of a cliff forever. The entire point is that going over is a sudden change. But I'd like to know why you say extended periods at 3v will gas. If true it would be important to know.
I guess the proof is in the pudding. Apparently the truck's alternator charging voltage (you wouldn't happen to know it would you?) is getting it to something like 66% SOC (assumes 120w for high beams, 100w low beams, 20w for the rest of the truck, and 12V). Which fits nicely with the first curve.
Found another source with a charge and discharge curve that seems to also say 2.4 is fine:
View attachment 76611
Fig. 1. Discharge/charge characteristics of the LTO/NCA pouch cell. a)...
Download scientific diagram | Discharge/charge characteristics of the LTO/NCA pouch cell. a) Different Crates at 20 C ambient temperature; b) 4C at different ambient temperatures; c) 0.1C at different ambient temperatures. Note the cells were first charged then discharged, both with CCCV...
So I guess those other curves are wrong or irrelevant somehow.
I did a autopsy on the battery and found that cell that blew was at 3.5v and the rest were at 2.0v to 2.3v, Most of the steel nuts very rusty and the buss bars were extremely tarnished , even between the nuts and the cells. The weakest link was the self dis charger cell and the connections.
The 5 other cells tested to 30ah and are still good.
The watt difference 3.0v and 3.5v is only @ 1/2 watt. Why take the chance, If even that you have all equal resistance on all connections and the cells.
We still live a imperfect world, LA hasn't slid into the ocean yet
I'm all for margin if it is possible and cost effective. And it seems like the field reports for 6S show that it is so I'm increasingly leaning that way. The voltage curves are just lines on paper after all.
But your destruction occurred because of a dead battery, NOT because of alternator charge voltage. And 5S makes that circumstance less likely by providing full capacity. That's why you would "take the chance". To keep a human from putting a 40A charger on it and walking away. These are engineering tradeoffs not go/no-go judgements.
The good news is we're discussing six of one and a half a dozen of another. Both will likely work for 30 years, and both can be killed by a mistake. One costs a few bucks more in exchange for a bit of voltage margin. The other has a little less risk of a dead battery due to a stray load. Thanks for the information about your experience, and I'll post here if I get a LTO pack running someday.
BuffaloTrace
Solar Enthusiast
- Joined
- Dec 3, 2020
- Messages
- 463
I have been thinking of using these in my trusty ole saturn I cant seem to kill yet.
6 pack - Toshiba 2.3v 20ah LTO
Every battery is professionally tested between 20-21ah. Laser welded M4 studs were added for easy assembly and blue shrink wrap as well for insulation. Recommended max continuous charge rate 60a (3c)Recommended max continuous discharge rate 160a (8c)30 second Pulse discharge rate 500a (25c)Fully...
batteryhookup.com
Seven30
New Member
So my bus has ~24V (28.6 charged) dual 8D starting batteries These cost over $250 each but can deliver 1400amps of cranking current.
But I'm sick of fretting over FLA going bad which just happened again!
The question is how much what would a 2P6S 60ah pack cost considering the 8Ds run about $600 to replace.
120 x 10C = 1200amps cranking.
But I'm sick of fretting over FLA going bad which just happened again!
The question is how much what would a 2P6S 60ah pack cost considering the 8Ds run about $600 to replace.
120 x 10C = 1200amps cranking.
11S1P 40 ah is @ 1600 CCA = @ 50./ cell + 60. ballencer is @ 610.00$ , using copper buss bars
Similar threads
