It doesn’t. In fact, I’ve got that in my review notes: “no charging LED.” Turns out there is a charging LED, but it’s only on if you’re charging with the DevTerm switched on; if it’s switched off, there’s no LED but it charges anyway.
Weird!
3 Likes
Ah! I was under the impression that because the light was off, that it was not charging, I will test this out and plug it in with it off on mine now. Thanks for the information!
I was under the same impression. I was also a bit over cautious about the batteries in general given previous discussions on the forums, and wasn’t sure it was wise to leave it plugged in, with batteries, especially if the light was off and it seemed like it wasn’t actually doing anything.
It’s a shame the light doesn’t come on, and then go off when the batteries are charged. I guess that’s another question – is it safe to leave the device charging when off? Will it properly stop charging the batteries when they are full? Or will it cause problems and potentially damage the device? (And same questions for both batteries that are protected, and batteries that are unprotected, as I’ve got the latter…)
I figured I’d err on the side of caution, but if I knew it was safe to charge when the Devterm is off, then it would be a useful thing.
It is absolutely safe, yes: there’s a battery management system with under- and over-voltage protection. That’s built in to the DevTerm, and is true whether you use protected or unprotected cells; the only time you need to worry about unprotected cells is if you take 'em out of the DevTerm for some reason.
Going back to my tests with the benchtop power supply, I can confirm that the batteries charge as you’d expect: 2A while empty, then gradually dropping as they reach full charge until the DevTerm stops charging automatically. At that point, the supply still reads a 0.001A power draw - but none of that’s going into the batteries, as far as I’m aware.
I’ve left mine plugged in overnight, same as my phone and laptop and all my other battery-powered devices: haven’t had a single problem.
3 Likes
I just received two Nitecore NL1832 batteries, labeled as 3200mAh, 3.7V, 11.8Wh. They are a tight fit, but they go in fully and I can close the battery door without problem.
I charged the batteries, plugged them into the DevTerm, booted it, and let it sit on the empty desktop with an SSH session displaying battery status every minute. 2.5 hours later it shut down with the final POWER_SUPPLY_CAPACITY=21.
Is this an expected power lifetime?
That feels a little low: during my own testing, with an A-0604 running at the default four-core 1GHz setting, I got three hours and 49 minutes of video playback from a pair of no-name “3,400mAh” cells.
Try letting the batteries charge in the DevTerm, rather than using an external charger.
ghalfacree:
Try letting the batteries charge in the DevTerm
Thanks, I will do that.
Could this be due to using re-wrapped cells that have inflated advertised capacity?
Do you mean mine, or PiNYC’s? Because mine are almost certainly rewrapped and may or may not be anywhere near their labelled capacity in reality, but I’m getting a longer runtime than PiNYC’s branded cells.
If you mean PiNYC’s cells, it’s possible they’re rewraps - Nitecore cells are among the most commonly counterfeited, so depending on where they got them the cells could be fake - but I’d wait to see the results after a full charge-discharge cycle in the DevTerm itself before worrying.
Apologies, I was asking about the decreased runtime that PiNYC was getting out of their cells.
I purchased these cells directly off the Nitecore web site in USA. So, I would hope they are not counterfeit.
I also made these observations with batteries charged in an external charger:
- when I replace the batteries while the DevTerm is running, the reported capacity (from xfce4-power-manager) is much lower than the one reported in the charger
- When I replace the batteries while DevTerm is turned off, they are reported as full capacity.
However there was a significant difference between experiment 1., which was conducted shortly after the batteries were sitting on the shelve for some months because I ordered them too early and 2., which was conducted today after the batteries were charged and used regularly.
So further data may be needed 
Yeah, I got some nitecore batteries as well. They are slightly too long(I got the protected) and a very snug fit. I’m concerned by a noticeable indentation caused by the springs pushing on the bottom of the cells. They are excellent batteries but I think it might be best to search for a smaller battery. I don’t even store the devterm with the batteries inserted.
Does anyone have protected cells that aren’t a snug fit? Thinking I might just have to buy some unprotected cells to avoid long-term damage to the battery. Or maybe modify the springs so that they aren’t as long but that would be my last resort.
I did this over the weekend and discovered this:
- Charged the batteries in an external Exar charger, then put them in the DevTerm
- Charged the batteries with the DevTerm with it off and plugged in to USB-C overnight
- Charged the batteries with the DevTerm on for a few hours
- Ran my test again - idle on desktop with SSH session showing battery capacity every minute
This time I got about 4.5 hours of runtime and POWER_SUPPLY_CAPACITY ended at 6.
Again, these are for Nitecore NL1832 batteries purchased directly from them.
That sounds a lot more like what I’d expect to see. Sadly, the DevTerm’s surprisingly power-hungry - and even locking the CPU to a single underclocked core doesn’t do much for total runtime, I’ve found.
Maybe someone needs to design a carrier for a USB Type-C battery pack you can mount into the expansion bay…
1 Like
So far I’ve tried the following battery:
NCR18650B (Panasonic NCR18650B Button Top Battery — FlashlightWorld)
Unfortunately it is a very tight fit (requires pry tool to remove battery). The diameter is small enough for the back cover to close, but large enough to put enough pressure on the back cover so that the case actually starts to bow and separate from the front housing.
I ordered the Epoch 18650s just before you made this comment, and didn’t see it until after I assembled my unit and inserted the batteries. They went in, but as you say they are pretty long. They may not come back out.
Are batteries with a nominal voltage of 3.6v ok? Or should we only stick with 3.7v? I was wondering (if under load) if the DevTerm would have stability issues with a slightly lower Voltage. Or does it not matter?
shouldnt be a huge deal. thats why we use 2 cells
Ah, that’s good to know. I think I read somewhere that someone reported their DevTerm just randomly powering off when under load (using off-brand batteries). I assumed these batteries were cheaper, but probably not able to sustain heavy load or had lower voltage ratings. I was concerned that on gear 6, maybe 3.6v might not be enough.
1 Like