First impression of LTO batteries

[BjarneK]

I recently installed an LTO (Lithium Titanate Oxide) battery for engine start and windlass use. LTO cells are nominally 2.3 V and don't have the flat section on the charge/discharge curve that LiFePO4 cells have. The full voltage range is 1.5 V to 2.8 V. The advantages of LTO cells are that they can deliver a very high current for their size and that they allegedly have a long life of 25,000 cycles. A small battery with the ability to deliver a high current can be a good choice for engine start, windlass and thruster.

The nominal voltage and voltage range for LTO cells means that you can either use five cells in series and maybe have a voltage that is a bit low for 12 V applications or use six cells in series and have a voltage that is a bit high. I went with six cells in series and only charging them to 2.5 V each. Not being charged to 2.8 V is probably not a problem even long term. I charge the battery using a Victron Orion-Tr 12/12-9 DC/DC converter which delivers around 9 Amps. I programmed a BMS to stop charging the cells at 2.5 V for a 15 V total battery voltage. The charging restarts at a cell voltage of 2.35 V.

I used 40 Ah cells that are very similar to these: https://shop.gwl.eu/LTO-technology/Y...l-B-Grade.html

The electrical system is very simple with only the following components attached to the LTO battery: BMS, DC/DC converter to charge, engine and windlass. The BMS does not do any low or high voltage cutoff, it just controls the DC/DC converter. The engine is a Nanni N4.60 and the Windlass is a Lofrans Tigres with 12 mm chain and a 30 kg Spade S140 anchor. The engine alternator is not attached to the LTO battery. Attached are the graphs of what happened during six minutes around the retrieval of 40 meters of chain at 12 meters depth.

At 0.5 minutes, the engine was started. At 1 minute, enough chain was brought in to take off the snubber. At 4 minutes, the chain was pulled in with about 15 meters left out. Then the engine was used to pull the anchor out of the muddy bottom. Finally, at around 5 minutes, the rest of the chain was brought in with the spikes at the end bringing the last bit in to have the anchor at the right place on the bow. The bottom (blue) curve shows the current and at around 4 minutes, the BMS decided to activate the DC/DC converter to charge the battery. The top curve (green) is the battery voltage.

You can see that the LTO battery has no problem handling the high loads. The total energy used during this anchor retrieval was around 2.5 Ah, so not very much considering the 40 Ah total capacity of the battery. The charging took around half an hour and the battery was not 100 % charged before this test.

All in all it seems that the LTO battery is a good choice for this application. It is quite small and the cells weigh 7.5 kg in total.

As an interesting aside, it turns out that the engine uses 250 mA when running for the instrument panel and sensors.

[CharlieJ]

Interesting. Thanks for the detailed report.

[sailorboy1]

do they catch on fire?

[Muaddib1116]

Everything catches on fire if you try hard enough.

[markcouz]

These seem like a very good technology for boats - with lots of positives and a couple of negatives (high initial cost and lower energy density than LFP).

Here is a good article (although produced by a seller of LTO batteries...)

https://zenaji.com/weve-been-buildin...ade-heres-why/

[CaptainRivet]

Originally Posted by Muaddib1116

Everything catches on fire if you try hard enough.

Wrong LTO don't, you can even drill into them while using them...overcharge you need to go very high>20v on an 1p6S for over 1h that they start to bloat...never catch fire

[CaptainRivet]

Originally Posted by BjarneK

I recently installed an LTO (Lithium Titanate Oxide) battery for engine start and windlass use. LTO cells are nominally 2.3 V and don't have the flat section on the charge/discharge curve that LiFePO4 cells have. The full voltage range is 1.5 V to 2.8 V. The advantages of LTO cells are that they can deliver a very high current for their size and that they allegedly have a long life of 25,000 cycles. A small battery with the ability to deliver a high current can be a good choice for engine start, windlass and thruster.

The nominal voltage and voltage range for LTO cells means that you can either use five cells in series and maybe have a voltage that is a bit low for 12 V applications or use six cells in series and have a voltage that is a bit high. I went with six cells in series and only charging them to 2.5 V each. Not being charged to 2.8 V is probably not a problem even long term. I charge the battery using a Victron Orion-Tr 12/12-9 DC/DC converter which delivers around 9 Amps. I programmed a BMS to stop charging the cells at 2.5 V for a 15 V total battery voltage. The charging restarts at a cell voltage of 2.35 V.

I used 40 Ah cells that are very similar to these: https://shop.gwl.eu/LTO-technology/Y...l-B-Grade.html

The electrical system is very simple with only the following components attached to the LTO battery: BMS, DC/DC converter to charge, engine and windlass. The BMS does not do any low or high voltage cutoff, it just controls the DC/DC converter. The engine is a Nanni N4.60 and the Windlass is a Lofrans Tigres with 12 mm chain and a 30 kg Spade S140 anchor. The engine alternator is not attached to the LTO battery. Attached are the graphs of what happened during six minutes around the retrieval of 40 meters of chain at 12 meters depth.

At 0.5 minutes, the engine was started. At 1 minute, enough chain was brought in to take off the snubber. At 4 minutes, the chain was pulled in with about 15 meters left out. Then the engine was used to pull the anchor out of the muddy bottom. Finally, at around 5 minutes, the rest of the chain was brought in with the spikes at the end bringing the last bit in to have the anchor at the right place on the bow. The bottom (blue) curve shows the current and at around 4 minutes, the BMS decided to activate the DC/DC converter to charge the battery. The top curve (green) is the battery voltage.

You can see that the LTO battery has no problem handling the high loads. The total energy used during this anchor retrieval was around 2.5 Ah, so not very much considering the 40 Ah total capacity of the battery. The charging took around half an hour and the battery was not 100 % charged before this test.

All in all it seems that the LTO battery is a good choice for this application. It is quite small and the cells weigh 7.5 kg in total.

As an interesting aside, it turns out that the engine uses 250 mA when running for the instrument panel and sensors.

What BMS do you use?
You accutally need an active balancer so pack stays in balance but can spare yourself a BMS, Not needed as 1p6S are very stable and the balancer will sort all imbalances at all points of SoC, you can deep discharge or charge with 18V with a faulty regulator, nothing will happen to LTO.

[BjarneK]

Originally Posted by CaptainRivet

What BMS do you use?
You accutally need an active balancer so pack stays in balance but can spare yourself a BMS, Not needed as 1p6S are very stable and the balancer will sort all imbalances at all points of SoC, you can deep discharge or charge with 18V with a faulty regulator, nothing will happen to LTO.

I use a homemade BMS. One of the cells that I got has an issue that makes it leak some current. It continually leaks around 8 mA. My BMS is capabable of discharging each cell with about 50 mA so has no problem keeping the pack balanced. Everything is still working great. With good quality cells I would think that very little balancing would be needed, so basically just the smallest balancer available.

[CaptainRivet]

Originally Posted by BjarneK

I use a homemade BMS. One of the cells that I got has an issue that makes it leak some current. It continually leaks around 8 mA. My BMS is capabable of discharging each cell with about 50 mA so has no problem keeping the pack balanced. Everything is still working great. With good quality cells I would think that very little balancing would be needed, so basically just the smallest balancer available.

LTO is different then Lifepo4 as the discharge curve is not flat you can balance it at any SOC. Because you are using high C rates for discharging will cause imbalance over time which a balancer will fix. I always use the Heltec 5A one, it doesn't cost much less taking a smaller one but they deliver the balancing current based on deviation, 60mV is 1A 100mV 2A....and only 5A total means 5A from cell to the other or 5A out and 1A per cell.
Alternative is the Neey active balancer as its at the same time a cell balancer Monitor via bluetooth with 4A on all cell simultaneously its expensive with 130Euro, too expensive for a starter.
50mA will be too small especially if the anchor got stuck and you heavily use it, the surges are high=stress on the cells. LTO need good active balancer but no BMS.

Current leaking cell i would replace, its a sign the cell was heavly overloaded or has a production defect (common on their 55AH cells but 40AH very uncommon) and has a very small internal short (most likely between 2 foils welded on a tiny spot together) then can get bigger anytime. I would just order one new cell and replace, keep the leaking one as spare. I come from car stereo competition and we tortured these LTO like there is no tommorrow...and they can withstand a lot and are extremly safe. What normally happens if you heavly overload them is they reduce capacity and they limit current delivery as getting higher resistance but no leaking. But when an LTO is leaking current its time to replace it.