I have a Testo 875-1i Thermal Imaging camera that, as a daft expensive toy, had been sitting unused for a while. When I needed it, I noticed that the battery (marked 0515 1100) was no longer working at all. It would not power the camera, and the camera wouldn’t charge it.
The camera has had little use compared to the levels it was designed for, so the battery was not degraded due to discharge cycles. I suspected that the charge controller was refusing to charge it because the cell voltage had fallen too low.
The cells themselves were registering just over 1.55V each, so they were pretty much fully drained. The lower threshold for a deep-discharged cell is typically 1.5V, below which the cell should be discarded, so I figured if I could push a surface charge into the cells, then I could persuade the charging circuit to activate.
And it worked! So I don’t have to throw out a serviceable battery, or fork out £130 or so for a new one.
Warning: This article is not a recommendation that you should do this yourself. Lithium Ion batteries can be dangerous. They hold a lot of charge, and can deliver a lot of energy in a short amount of time. If you suspect the battery is actually faulty or has been depleted through use, then just replace it. Certainly do not try to revive a cell that has fallen below 1.5V, because conductive bridges may have formed internally. In my case, the cells were very likely still above the safe pre-charge level, and I took the precautions of checking for signs of overheating repeatedly during charging. If you are not confident you know what you are doing, play safe and buy a new battery.
Summary of what I did
In short, I removed the battery and opened the casing to reveal two cells with positive terminals at each end, and negative terminals in the centre (the batteries operate in parallel). The negative terminal can be reached easily on the underside of the charge controller PCB, which is accessible without removing anything, though it may need to be edged up a couple of millimetres to get a clean view.
I then took a 12V power supply that had stripped ends, and used my voltmeter to verify which wire was negative. I then held the negative wire to the negative terminal of the cells and repeatedly touched the positive wire to one of the positive terminals – just a fraction of a second each time, with half a second or so inbetween. Within a minute or two, the cells were reading 2.6V, so I reassembled the casing, powered up the camera, and then inserted the battery. The on-screen icon that had only been showing the ‘DC-Mains’ icon soon changed to a ‘battery’ icon, and within a few seconds it was showing the ‘charging bars’ animation.
It’s vitally important that the battery is inserted only when the camera is powered up! The battery only has a surface charge, not nearly enough to run the camera on its own.
So, with the camera powered up, the system reads an acceptable (but still too low) voltage from the cells, and starts to charge the battery as normal. I removed the battery after 15 seconds or so to check it wasn’t getting warm, and again after a minute, then after 5 minutes and so on. It remained cool, and charged fully (to 4.16V) within a few hours.
Some notes on the battery
The label on the battery states it’s 3.7V 5000mAh, and since it contains two cells, each cell must be 2500mAh.
To open the battery case, I first popped off the snap-on foot by holding in my hands and pushing up at either corner of the flat edge with my thumbs. From there, I could slide a watch-opener knife into the seal that runs all around the casing – it seems to be glued lightly, so the blade only needs to be a couple of millimetres into the seal.
The casing shell could then be prised gently apart to reveal the two cells and the charge-controller PCB. The battery contacts come through the PCB to the underside, which provides a useful contact point for the negative terminals.
On the charge controller’s PCB, I could see that the two outer pins are marked positive and negative, and the inner two pins are marked I2C, we can see below SDA and SCL. It might be fun to probe the address space and see what comes back.
A note on safety
The cells in my battery back have no markings, so it wasn’t possible to find a datasheet for them. This means I was taking a risk recharging them, because I don’t know what the safe thresholds are for this particular cell’s chemistry. The chemistry in Lithium Ion cells varies a lot, and therefore the data parameters vary too.
My only real concern was if short-circuits had developed inside the cells, but this would be easy to detect in the form of heat generated when a voltage was applied. Also, the fact that the cells happened to be just above the ‘typical’ lower discharge threshold of 1.5V, I guessed that this was not a coincidence and that discharge was purposefully stopped at this level by the charge controller (which would itself otherwise consume power).
So, when giving the cells the initial boost with 12V, I would check them periodically to make sure they were not getting at all warm (this would suggest there was a short circuit inside the cells). At no point did the cells feel warm. I also checked the voltage of the cells periodically to make sure it was rising as expected.
I read that cells require a pre-charge if they’re between 1.5V and 2.5V, whereby the current flowing to the battery should be limited (i.e. it should not be fast-charged). Pre-charging a cell should typically not exceed 0.5C, and for 2 cells each 2500mAh, I figured that the power-supply, which provides a maximum of 4A at 5V, would struggle to get much above that anyway when the camera is running.
In any case, the current is limited electronically by the charge controller, and the camera is not particularly fast to charge. I doubt it charges above 0.5C in normal operation. Therefore, as long as I periodically remove the battery pack to check it’s not getting warm during the first charge, then the pre-charge stage should proceed safely.
As it turned out, the battery pack was cool to the touch at all times. I have no concerns at all; the battery operates as it did when the camera was new.
The dead battery is ejected.
The foot on the battery pack has to be removed.
The case is opened by sliding a small blade carefully under the lip that joins the two halves. The first long side shown below wasn’t actually sealed on my battery pack.
The opposite long edge, shown below, had been glued but only with a small amount of soft adhesive. The blade easily slid under the lip on this edge, as well as on the short edges which were also glued.
The pads marked below are visibly connected directly to the cells we want to charge, so these are the pads I used to push 12V into the cells.
The PCB with cells can be gently prised up and removed completely if necessary to get adequate access to the metal pads.
To apply 12V, I connected the PSU negative wire to the negative terminal shown above, and then tapped the positive wire to one of the positive terminals shown. It doesn’t make any difference which positive terminal is used. I periodically checked the voltage until it got above 2.5V.
Did you glue the pack back together and what adhesive did you use for that? I’m trying to figure out how to rebuild some laptop and other old packs. Thanks.
There was enough glue on the casing that I could just push it back together, so I didn’t use extra adhesive. I had intended just using Bostik All Purpose, these general purpose glues tend to stick plastics reasonably well for home/DIY use. My other choice would have been N7000, which is often used to bond mobile phone screens to frames, hence quite strong and tolerant of different surfaces.
I’m assuming you’re not rebuilding packs for resale. If you are, then you’d have to identify the plastic used for the casing and choose the proper way to bond it – it’d seem risky otherwise, since many plastics are surprisingly difficult to get strong bonds with.
Great info! Thanks!
I choose to charge cells as mentioned with 50mA current until the reached 2.6V instead of using 12V.
Then inserting battery and connected charger before starting up.
Works fine now!
Glad you got it working! For sure, a constant current source is a lot less hacky than tapping wires from a wall-wart – thanks for sharing.
Kevin, Great write up! I measured my batteries at 1.9V at the start of the procedure. I took a slightly different approach and touched the leads from a 9V transistor radio battery for a few minutes until the battery voltage increased to 2.75V. When I put the battery pack into the camera it started charging. I’ll go ahead and order some new batteries but your fix got my camera running today. Many thanks! David.
Thanks Kevin, the battery cost an arm and a leg. I took it apart and found the same thing you did. I did not see your write up until I had taken it apart. My batterie had a very strong glue bond. Thought I was going to destroy it taking it apart. Once apart I was about to give up on it until I read your write up. I used a power supply manually increasing the voltage, limiting the current to under 50ma. Doesn’t look as good as yours but will charge now. Thanks again.
Fantastic fix! I also used a 9v radio battery as suggested by David to give it the initial boost (about 5 minutes). Snapped the case back together (no glue needed) put the foot back on the battery pack and inserted it into the already powered up camera. Icon in the top-left of the screen changed immediately from ‘plugged in’ to ‘battery charging’ (which it would not do before). After a few hours, the camera still works when unplugged. Just wanted to say thanks.
Here at work we had the same issue. Only 1.7 V left. No charging
We didn’t open the battery and just put 5 V, 40 mA on the 2 of the contacts for 2 minutes. The are paired and only of the 2 pairs will accept the load and start charging. After 2 minutes we had > 2,7 V . Exactly the same voltage on both pairs.
Put the battery in the camera and it started charging
Thanks for sharing, that’s interesting.
I had tried kicking it into life with a brief 5V, but I was too wary of burning out the charge controller to back-power it for longer.
Hi there, great post to read, indeed.
It looks like my battery is dead,
since i’m measuring 1273 mV at the cells and 1233 mV between the outer I2C_SDA and the +U_Lion contacts.
Would anyone recommend to try a reanimation, and if so how?
Hi, you should measure voltage of the cells themselves. I2C_SDA is a data-line that connects to the charge controller.
Hi Kevin, yes – 1273mV has been measured at the cells.
The cell itself bears SANYO R1122 O41B embossed into the red wrapper. Most likely that matches with these specifications for the UR18650F which has an insulator ring of the same color. https://secondlifestorage.com/index.php?threads/sanyo-ur18650f-cell-specifications.1709/ as found here:
Will go to repair and replace the cells. My question into this round is – hoot replace the cells and why not choosing for an even higher capacity such as 3500 mAh?
Yes I see now, sorry I must have skimmed your comment. So personally, I’d not choose to use those cells. Even if they took a charge, I’d be worried about bridging and short-circuits from being discharged and recharged out of spec.
For capacity, I’m sure charge controllers work on voltage and temperature to keep charging at a safe rate, so higher capacity cells should work fine, assuming they otherwise use the same chemistry.
As an aside, I’d probably be unable to resist mucking around with the old cells outdoors, just to see if they’d charge. A lot of things are made to a higher spec than is legally required, and it’s interesting to see just how tolerant it actually can be. e.g. John Ward – Cable Overload (UK mains cables)