A brief test of a 66160-size LTO battery

This battery was sent to me by one of my readers, and I’d like to thank him for that. It’s not exactly new and has a few quirks, but as always, you can find out more in the review.


First, a note about its history: seven batteries were bought to build a 12 V car battery. One was rejected during assembly because of its high self-discharge, then tossed on a shelf, where it sat from 2018 with a residual voltage of 0.5 V.
I was mostly interested in finding out how LTO batteries hold up to long-term storage in this condition, so I accepted the offer to put this one through a bit of a workout.

The link in the title is there as an example and to give you an idea of what batteries like this cost.

When I unpacked the parcel, my first thought was: this thing is huge... I mean, I’d seen photos and drawings with dimensions, but seeing it in pictures is one thing; holding it in your hands is quite another.


For comparison, here it is next to an 18650 battery, which practically disappears beside it.


As you’ve probably noticed, the battery has no markings, so it’s fair to call it a no-name cell. You can find various specs online, but as far as I know, cells like these have a capacity of 30 Ah, although some sellers label them as 35 or even 40 Ah.

Here are the specs I found for similar cells:
Size: 66160
Voltage: 2.3 V
Capacity: 30 Ah
Continuous discharge current: 10C (300 A)
Peak discharge current: 30C (900 A)
Maximum charging current: 10C (300 A)
Low-voltage cutoff: 1.6 V
High-voltage cutoff: 2.7 V
Discharge operating temperature: -30 to +60 ?
Charging operating temperature: -30 to +60 ?
Cycle life: 15000
Internal resistance: <0.8uOhm
Weight: 1100 g
Dimensions: 160x66x210 mm

The battery connects via screw terminals with a 12 mm thread diameter.


The battery weighs over a kilogram, is 66 mm in diameter, and measures 160 mm long without the terminals and 210 mm with them. By the way, sometimes they’re marked 66210 instead of 66160, although as far as I understand, it’s the same thing in this case.


Similar batteries are also available in other sizes: short 18 Ah versions and long 55/60 Ah versions, though I suspect that’s not the full range.


Descriptions give various figures for the impedance of these cells, but I found a photo online showing actual measurements, with the instruments reading around 0.25-0.35 mΩ.


Unfortunately, long-term storage in a discharged state took its toll: the battery’s impedance rose to 1.33 mΩ.


I used two electronic loads for the tests. The EBC-A20 was used at 0.2C and 0.5C, and the more powerful EBC-A40L at 1C and 40 A.


The person who sent me the battery had charged it before shipping, but I charged it again before testing. It took about 900 mAh.

There’s a catch here: since the battery is unbranded, its charge termination voltage is unknown. Some sources say 2.7 V, others 2.8 V. I tested it at 2.7 V first, then ran tests at 2.8 V as well. The cutoff current was 0.25 A in all cases.



Charging at 5 A (the EBC-A20’s maximum) took 5 hours and 33 minutes.


At a charging current of 1C (30 A), charging took just over an hour, but the CV phase dragged on for a long time.


I ended up running quite a few tests, so I’ll try to organize the results somehow.
All tests were run at an ambient temperature of about 19 degrees, so the results are a little worse than they could have been, though I don’t think it makes a significant difference.

First, the standard test at 0.2C, 0.5C, 1C, and 40 A. These are the results from the first run at each rate.
Unfortunately, even here you can see how bad things are. In my view, the battery is literally "dying". It delivered 26.63 Ah at 0.2C, but at 40 A its capacity was just 17.28 Ah, or 34.5 Wh—very little for a battery of this size and weight.


But the next results are more interesting: these are the consecutive tests at 0.2C (6 A), and you can see that each successive cycle performed a little better. In addition, the last one (the fourth, marked in green) was tested at a charge voltage of 2.8 V rather than 2.7 V, but this didn't help the battery much.


The same test, but at a discharge current of 0.5C (15 A). Again, there are four cycles, and the fourth was also run at a charge voltage of 2.8 V.


Three cycles at a discharge current of 40 A. For the first one (red), the charge voltage was 2.7 V; for the other two (orange and yellow), it was 2.8 V.


And now, all the test results at a charge voltage of 2.7 V on one graph.


The battery seems to get noticeably warm. Below are the results for discharge currents of 30 and 40 A. At 40 A, the temperature was about 34 °C, 15 °C above ambient, which is quite a lot.


But I wasn't done yet, and also ran a small cycle-life test. Though in this case, I mainly wanted to see whether the capacity would increase with this kind of “conditioning.” I ran 40 full charge/discharge cycles with the following parameters:
Charge at 1C (30 A) to 2.7 V, terminating when the current dropped to 0.3 A
10-minute rest
Discharge at 40 A down to 1.6 V
10-minute rest.

I could have run more tests, but didn't see much point. Besides, as it turned out, this is a fairly costly undertaking.



It may look as though the capacity gradually increased. Perhaps it did, but in reality the changes were much more likely due to fluctuations in air temperature. The jump at the end coincided with a noticeable warm spell outside: the room temperature rose from 19 °C at the start of the test to about 21 °C.


Capacity in Ah (blue) and Wh (red) on one graph.


Of course, many people are interested in what these batteries are used for. Where I live, they're most often used to replace car batteries, with 6S packs being assembled for this purpose. More often, though, these are used batteries that originally came from high-power UPS units and energy storage systems, for example, those used with solar power plants. This is due to their long service life and good charge/discharge current capability.
1. Batteries at the production stage.
2. A battery, apparently in someone's basement.
3. An assembly of batteries like these.
4. A battery for a solar power plant.


While preparing this review, I came across an interesting active balancer, probably inductive, designed for batteries like these.



Conclusions.
It turned out that the battery didn't survive being stored in a discharged state. Nevertheless, it still works, albeit with noticeably lower performance.
Storage increased the battery's internal resistance, making its available capacity highly dependent on the load current. For example, at 6 A it delivered 27.5 Ah (58 Wh), but at 40 A it delivered just 17.3 Ah (34.5 Wh).

Even so, this is a pretty interesting battery, and I think the information I've gathered may be useful as well.

That's all for now. Once again, thanks to Vladimir for providing such an interesting specimen.