CALB L173F230B LiFePO4 230 Ah batteries
To keep up with new posts on the blog, I recommend following me on Instagram. I’ll post notifications about new articles there and might also write about something interesting from time to time: link to my account
You can also subscribe to updates and new articles on the Telegram channel - https://t.me/KirichBlog
The link below the review goes to 280 Ah batteries sold by Nkon, but only because it’s the only 100% verified place I know of where you can buy quality batteries. Of course, there are other stores, but I write about what I’ve personally tested: link.
And yes, the batteries in today’s review were sent by the manufacturer, so I have no doubt that they’re genuine or that they’re new. That means the test results can certainly be used as a reference.
Shipping was handled by m3cargo. No complaints, except that delivery took an awfully long time.
The packaging is standard: a box for four cells and a tray made of expanded polyethylene. The package also included four busbars and a set of nuts.

They look pretty typical for batteries like these, although there are a few quirks, which I’ll get to later.

Dimensions: length excluding the terminals is 205 mm, width is 173 mm, thickness is 53–54 mm, total terminal length is 16 mm, the stud itself is about 10–11 mm long, and the thread is M6.
While inspecting them, I noticed that all the cells say “Norminal Voltage.” Apparently, they couldn’t decide whether to write “Normal” or “Nominal,” so they came up with a universal version :)

They’re fairly heavy, almost 4.2 kg each. I think the total weight with packaging was listed as 17.5 kg.

All the cells are clearly from the same batch, as indicated by the identical number on the bottom, but they have different serial numbers: D7A1000039, D7A1000050, D7A1000041, and D751002154. As you can see, only the last cell stands out from the group, since its serial number is clearly different from the others. That might even be a good thing—we’ll check the differences and see whether it’s really significant.
Judging by the serial numbers, the batteries were manufactured in 2023. As a reminder, the first letter indicates the year of manufacture: B = 2021, C = 2022, D = 2023. I can’t be more precise than that. The batteries were shipped at the very end of 2023.

Before testing, I numbered all the batteries. The results below are presented in the same order, which also matches the serial numbers listed above.

I’ll say right away that the tests took quite a while, because I had to literally catch the breaks between power outages—and at one point, I missed one. And since the tests were also energy-intensive, powering the UPS during charging was problematic.
The cell voltages were almost identical to begin with, which is certainly encouraging.

I found a datasheet online. The only difference, as I mentioned above, is the terminals: the datasheet shows square terminals. In listings from other stores, I also mostly came across versions with square terminals, whereas mine have round terminals.

The datasheet lists the cells' DC internal resistance.

I couldn't find the AC impedance there, but I did come across it on one of the websites: it's specified as no more than 0.25 mΩ. The same site also lists typical charging and discharging conditions.

Measuring the discharged batteries right after they arrived gave readings of 0.18–0.20 mΩ, noticeably below the specified maximum.

Next came the charging cycle. Given the difficulties mentioned above, I didn't run a “normalization” cycle; I simply charged the batteries fully, let them sit for almost two days, and then measured their capacity.
Charged at 40 A, with charging stopped when the current dropped to 4 A; charge termination voltage was 3.65 V.
The first cell took 204.4 Ah, or 690.3 Wh.
The second took 203.5 Ah, or 687.6 Wh.
Charging the third was interrupted by a power outage.
The fourth took 204 Ah.
The results are broadly similar. As you can see, charging each cell took more than 5 hours, for a total of about 21 hours just for charging.

Almost two days later, I measured the cell voltages. The maximum difference between the third and fourth cells was about 11 mV, which seems fine to me.

Next came a discharge at 40 A, which is almost the standard 0.2C test rate. Each cycle took about 6 hours. The battery tester was powered by a UPS, and even then I had to time things carefully, because the UPS couldn't run for long with the computer connected to it as well.
Well, what can I say? Yes, there is a capacity spread, but measurement error has a greater effect than the actual difference in capacity between the cells.
Cell No. 3 had the lowest capacity at 239.8 Ah, or 770.43 Wh; No. 1 had the highest at 241.6 Ah, or 777.06 Wh.
So, at a capacity of around 240 Ah, the spread is less than 2 Ah or 7 Wh.
In other words, the actual capacity the whole battery can deliver is 239.8 Ah, or 3080 Wh.

But as you can imagine, I didn't want to stop at a single test. So I assembled the batteries into a pack, measuring the total impedance along the way, which came to about 2 mΩ. Something odd happened here: the total should be around 0.8 mΩ, but I suppose some of it was lost across the busbars and contacts, although it still seems rather high.
This photo was taken right at the end, with the batteries fully charged.

After the test, I fully charged the batteries. To speed things up, I charged the whole pack at once, but since power outages could cause problems again, I did it in several stages.
I charged until the current started to drop. In total, I “put in”:
88.43+94.777+54.012=237.219 Ah
The charging process used 1200.1+1309.37+751.212=3260 Wh.

Since charging wasn't completely even due to the differences in battery capacity, I finished charging each cell individually, balancing the batteries in the process.
The batteries ended up receiving:
1. 237.219+4.7=241.919 Ah
2. 237.219+3.104=240.323 Ah
3. 237.219+3.228=240.447 Ah
4. 237.219+4.277=241.496 Ah

As you can imagine, testing these batteries at 40 A isn't very interesting. So partly for this test, partly as a “first attempt,” and partly to sell, I built a load rated for 100 A and up to 1300 W continuous.
No alligator clips this time—just terminals and 16 mm² welding cable. Hardcore.
So there were two tests here: measuring the battery capacity and, at the same time, checking how the new electronic load performed.

There are, of course, far fewer settings than on dedicated electronic loads from Atorch, for example, but I think it's enough for battery testing. The main things here are the ability to set the charge termination voltage and to calculate capacity in Ah/Wh.
The maximum voltage is 50 V, limited by the protection circuit that prevents the unit from being switched on without voltage. Yes, the load won't turn on until there's at least a small voltage at the input, and it will shut off if contact with the battery is lost, if it receives a command from the meter module, or if it overheats.
I set the actual maximum current to 103 A in case it drifts during operation.

Next, reset the capacity/energy counters and start the test.
First I set the discharge current to about 100.5 A; after a while it dropped to 100 A. Over the nearly 2.5-hour test, the current fell by a total of about 0.6–0.7 A. This was due to the power modules' current-measuring shunts warming up. Yes, I could have added a measurement circuit using a common shunt, but I didn't see the point: it would make the circuit more complicated without affecting the results at all.
At 100 A, the total voltage drop across the wires and terminals was about 0.2 V, which is pretty good in my view. This error could be cut exactly in half by adding a four-wire connection. It's not difficult, but it would be less convenient to use.
Inside are two power modules with larger heatsinks, which let me increase the power from 1000–1200 W to 1200–1400 W. As you can see in the photos, the load ran at about 1300 W for almost the entire test, without any problems.

After about 2 hours and 20 minutes, the voltage started to drop noticeably. As a reminder, the cutoff threshold was set to 10 V. The power began to fall accordingly, although the voltage had barely changed for most of the test. This is typical of the LiFePO4 discharge curve.

Since the test was run without a protection board, I had to monitor everything manually. Toward the end of the test, the voltage was dropping so quickly that there was a noticeable difference even between two consecutive photos of the same battery. Still, just to be safe, I photographed the readings on all four cells. I tried to do it as quickly as possible, so I didn't even take the time to get the probe polarity right.

The load performed perfectly throughout the test. The hottest parts on the outside of the case were the power connectors, which reached 44–46 °C. The outside of the case was practically at room temperature, though it was of course quite hot inside. I estimate the transistors were at about 90 °C.
The first two photos were taken an hour apart.

And that's it—the test is over. The battery delivered 237.28 Ah, or 2.96 kWh, which is pretty close to the results of the test at 40 A, especially considering that the discharge current here was 2.5 times higher.
In other words, the battery delivered a little less, which is normal. This also clearly shows its big advantage over lead-acid batteries, especially starter batteries, which can't be discharged at currents like this for long at all. And I'll say nothing about the size—it's like night and day...

The battery barely warmed up, partly because of its low internal resistance and partly because of its large mass. The temperature on top was about 32–37 °C. The maximum reading was masked by the hot wires to the load; right after disconnecting it, once the wires had cooled a little, I was able to find the battery's maximum temperature: about 38 °C somewhere inside, between the second and third cells.
In my view, that's pretty good for a 100 A discharge. Of course, that's less than 0.5C, but still, delivering almost 1300 W continuously for nearly 2.5 hours is pretty impressive.

This battery had already been used while preparing another article, and while I had it here I measured its voltage after a couple of weeks sitting idle. It was about 13.5 V, or 3.375 V per cell. Unfortunately, I didn't think to take a photo, so this is all I have for now.
By the way, I do have a BMS for this battery—a basic Heltec rated for 120 A. I made some fairly powerful passive balancers and might show them sometime, but I still haven't figured out the enclosure. I was thinking of making one from laminated particleboard and simply ordering it cut to size and edge-banded, but that would add a lot of weight and bulk, and is probably overkill. I have some old laminate flooring, but it's too narrow; I also have 6 mm OSB, but that's too thin. I'm not sure how to assemble a box from it at all.
If anyone can help with this, I'd be very grateful.
As for the quality of Calb batteries, I can say I'm happy with them so far, based on my experience using two of my three battery packs: 130 Ah 4S and 88 Ah 8S. Another battery, 187 Ah, is still in reserve.
They've been especially useful in the current situation.
The 8S pack runs a 1400 W (2000 VA) UPS, and that's where a problem came up—not with the battery, but with the UPS. I bought it new, but as a discounted item, and I've already managed to burn it out. You might think it failed from an overload? No, you're very wrong. It burned out after I clicked the Self-test button in its PC software. Yes, that's right: the UPS was unloaded and plugged into the mains, its AVR was set to buck by one step, and I connected it to the PC to see what it could do. I just clicked the test button, and there was an immediate loud bang. The UPS went dead. Disassembly showed that some of the transistors in one half-bridge had blown: all 4 of the upper ones (three shorted and one open-circuit) and 1 of the 4 lower ones. So apparently this kind of bug does happen—I had no idea...

That’s all for now—but only for now, because next I’ll assemble the battery and test it under real-world conditions, with power outages almost every day. To be fair, though, we’ve already gone 4 days without an outage. It feels strange to wake up in the morning and find that the power is still on.
I have no complaints about the batteries themselves. Their capacity is actually higher than specified: at 230 A, I got almost 240 Ah out of them. The battery ended up with a capacity of about 3 kWh, and it’s even smaller than a 60 Ah lead-acid car battery.
As for where to buy them, as I mentioned, Nkon is one option, but I think you can find them elsewhere too. If you’re buying on OLX, for example, be very careful: some sellers offer used batteries with the labels peeled off.
As always, I’d be glad to hear your comments, advice, and questions. I hope you found this review useful.
батареи calb отзывы, https://www.kirich.blog/obzory/akkumulyatory/1235-akkumulyatory-calb-l173f230b-lifepo4-230ach
Questions and answers
What capacity did the CALB L173F230B cells deliver at a 40 A discharge?
The cells delivered 239.8–241.6 Ah, or 770.43–777.06 Wh each. Cell 3 had the lowest measured capacity and cell 1 the highest.
What internal resistance did the author measure for the CALB L173F230B cells?
The discharged cells measured 0.18–0.20 mΩ. The assembled battery’s impedance was about 2 mΩ.
What is the review’s conclusion about the 230 Ah CALB L173F230B batteries?
In the author’s tests, the cells delivered nearly 240 Ah, exceeding the stated 230 Ah. The battery capacity was about 239.8 Ah, or 3080 Wh, and the author reports no issues with the cells themselves.
This model is included in the summary table: Battery comparison: measured capacity and resistance
The figures from this review are in the measurement database: Batteries
Discounted products
You may be interested in
Discounted products
All the content on the site and the advice are free, but we’d be grateful for your support of the project and the channel!
Comments34
Only registered readers can comment. It takes a minute to sign up, and you can log in with the same account you use on the forum.
Log inRegister



Comments are published in their original language.