BU-911: How to Repair a Laptop Battery
Find out the challenges and limitations of repairing “smart” batteries
Most laptop batteries are smart and consist of the “chemical battery” that is managed by the “digital battery.” A common protocol is the System Management Bus, better known as SMBus.
The typical SMBus battery has five or more battery connections consisting of positive and negative battery terminals, thermistor, clock and data. The connections are often unmarked; however, the positive and negative are commonly located at the outer edges of the connector and the inner contacts accommodate the clock and data. (The one-wire system combines clock and data.) For safety reasons, a separate thermistor wire is brought to the outside. Figure 1 illustrates a battery with six connections.
Figure 1: Terminal connection of a typical laptop battery
The positive and negative terminals are usually placed on the outside; no norm exists on the arrangement of the other contacts.
Courtesy of Cadex
Some batteries are equipped with a solid-state switch that is normally in the “off” position and no voltage is present at the battery terminals. Connecting the switch terminal to ground or pulling it up often turns the battery on. If this does not work, the pack may need a code for activation. Battery manufacturers keep these proprietary codes a well-guarded secret to which even service personnel have no access.
Use a voltmeter to locate the positive and negative battery terminals and establish the polarity. If no voltage is present, a solid-state switch may be in the “off” position and needs activating. Connect the voltmeter to the outer terminals, take a 100-Ohm resistor (other values may also work), tie one end to ground and with the other end touch each terminal while observing the voltmeter. Repeat by tying the resistor to a positive voltage potential. If there is no response, then it is possible that the battery is dead or locked by a code. The 100-Ohm resistor is low enough to engage a digital circuit and high enough to protect the battery against a possible electrical short.
Establishing the connection to the battery terminals should now enable charging. If the charge current stops after 30 seconds, an activation code may be required. Some battery manufacturers add an end-of-battery-life switch that turns the battery off when reaching a certain age or cycle count. They argue that customer satisfaction and safety can only be guaranteed by regularly replacing the battery. Mind you, such a policy also rotates inventory.
If at all possible, connect the thermistor during charging and discharging to protect the battery against possible overheating. Use an ohmmeter to locate the internal thermistor. The most common thermistors are 10 Kilo Ohm NTC, which reads 10kΩ at 20C (68F). NTC stands for negative temperature coefficient, meaning that the resistance decreases with rising temperature. In comparison, a positive temperature coefficient (PTC) causes the resistance to increase. Warming the battery with your hand is sufficient to detect a small change in resistor value when looking for the correct terminal on the battery.
After repair, the fuel gauge might not work, is inaccurate or provides wrong information. The battery may need some sort of an initialization/calibration process by fully charging and discharging the pack to reset the flags. A “flag” is a measuring point to mark and record an event. (See BU-603: How to Calibrate a “Smart” Battery).
The circuits of some smart batteries must be kept “alive” during the replacement of the cells. Disconnecting the voltage for only a fraction of a second can erase vital data in the memory. An analogy is open-heart surgery where doctors must keep all organs of the patient alive. The lost data could contain the resistor value of the digitized shunt that is responsible for the coulomb counter and other data.
To assure continued operation when changing the cells, supply a secondary voltage of same voltage level through a 100-Ohm resistor to the circuit before disconnection. Remove the outside supply only after the circuit receives voltage again from the new cells. Furthermore, some fuel gauge chips run wires to each cell. These must be reassembled in the correct sequence beginning with cell one, then two, three and so forth.
You will also need to be aware of compliance issues. Unlike other regulated standards, the SMBus allows variations and this can cause problems. The repaired SMBus battery should be checked for compatibility with the charger. Batteries for critical uses, such as heathcare, are typically replaced and not repaired. See also www.sbs-forum.org and www.acpi.info.
Simple Guidelines when Repairing Battery Packs
- Only connect cells that are matched in capacity. Do not mix cells of different chemistries.
- Never charge or discharge Li-ion batteries unattended without a working protection circuit. Each cell must be monitored individually with a protection circuit.
- Include a temperature sensor that disrupts the charge current on high heat.
- Apply a slow charge to a repaired pack to bring all cells to parity.
- Pay attention when using an unknown brand. Elevated temperature hints to an anomaly.
- Li-ion is sensitive to reverse polarization. Observe correct polarity.
- Do not charge a Li-ion battery that has physical damage, has bulged or has dwelled at a voltage of less than 1.5V/cell for some time.
- Check a repaired pack for self-discharge. Intrinsic defects often have high self-discharge.
Last updated 2016-01-29
*** Please Read Regarding Comments ***
Comments are intended for "commenting," an open discussion amongst site visitors. Battery University monitors the comments and understands the importance of expressing perspectives and opinions in a shared forum. However, all communication must be done with the use of appropriate language and the avoidance of spam and discrimination.
If you have a suggestion or would like to report an error, please use the "contact us" form or email us at: BatteryU@cadex.com. We like to hear from you but we cannot answer all inquiries. We recommend posting your question in the comment sections for the Battery University Group (BUG) to share.
Previous LessonNext Lesson
Or Jump To A Different Article
Thanks for the wisdom Syonyk.
"Warnings" might be closer. But ok.
I share your concerns about the failure mode of large banks. It's one reason that Telsa style per-cell fusing is popular (and relatively easy). That still doesn't stop a worst case failure, so using a proper enclosure, having environmental monitoring, and possibly some method of handling a critical fault are on my list to sort out.
Sorry, popular and easy? Citation needed. Who, other than Tesla, is doing per-cell fusing? I've seen some templates for laser cut strip that has fusing capability on the positive side, but it's rare to see a pack that uses it.
And I've never seen per-cell fusing in any of the packs I've pulled apart, though they're mostly ebike packs and the occasional tool pack. Standard is 10mm wide, 0.15mm thick nickel strip, though some of the folded packs use cut plates instead of just strip. The tool packs I've torn apart use 0.3mm nickel strip, generally in more of a plate form than strip.
All the DIY Powerwall rubbish I've seen is just soldered wire to the cells. I'd love to see one that's spot welding, but I haven't seen it. I'm sure it exists, maybe?
Spot welding is definitely the way to go. I haven't gone looking for a spot welder amongst my local resources, but I suspect there's one to be found pretty easily. As you noted, the Chinese models aren't bad either.
I suspect you'll have a remarkably hard time finding one locally - or, at least, a decent one. I've seen plenty of microwave oven transformer based spot welders, usually controlled by a light switch or foot pedal, where the users comment that it only occasionally blows a hole in the battery.
A good US-built capacitive discharge welder is around $3k-$5k, depending on the power - pick the thickness you want to weld, and cost is mostly proportional. You can also DIY a capacitive welder, though keeping your MOSFETs alive is a bit of a challenge (fun with flywheel diodes and the like - avalanche issues are quite real and a MOSFET shorted open is a bad deal if you have a lot of capacity).
A Chinese transformer based spot welder is around $200-ish shipped. It'll almost certainly be a Sunkko unit, and I suggest spending the coin for one that has a welding pen set. The electrodes are good for quite a few thousand welds before you have to replace them - they're a hardened copper alloy of some variety, though if you push the current up they last a long while.
The transformer units can be hard on your circuits, though - if you have access to 240V, use that. Don't be a moron and wire your house wrong to get 240V, though.
120 cells is the first pack. For me, it's a proof of concept and a good chance to see if this is enjoyable enough to really tackle. Since I'm targeting cells in the 2600-2800 mAh range, it won't be quite as pathetic as a pack running on the worn out (1200-2,000 mAh) batteries that I have in the middle of my testing range. Those are all going to a friend who has a different use for them.
So... what are you trying to do?
I get that you're trying to build a cute battery pack, but what's the end goal? You can't be doing anything actually off grid with that small a pack.
Is the end result a ~1kWh battery pack for the purpose of having a 1kWh pack? A standalone system? Grid tied (probably illegally since I doubt anyone with a license will sign off on a DIY junk 18650 based storage system)? Phone charging? Ebike?
I've been researching the right BMS, but I have some time before I'll need it. Fortunately, that means I have the time to do it right.
What voltage/amperage are you aiming for? If you say anything >1C, I'll smack you if you try and do that with laptop cells - they're about 0.5C cells, tops, and beyond that is abusing them. Especially the older ones.
The BMS will be selected based on your series group count. There are plenty of ebike-voltage-range BMS units - through about 16S is easy, beyond that starts getting tricky.
I'm fortunate enough to have a luxury coach builder as a fellow member of our local makerspace, and he's been building decent sized battery systems for coaches for years. He has a lot of experience with the general idea, although not as much with lithium ion batteries, and has been very willing to provide his input.
If he's familiar with lead acid, lithium shares reasonably little.
Lead acid is ~self balancing, if you push the voltages enough, and isn't particularly prone to thermal runaway (it can be done, but you have to try). You can't cycle them deeply, even deep cycle packs, without affecting lifespan radically.
But, really, I'm still trying to figure out what, exactly, you're planning to do with this.
It's a weird size - too big for most electric bikes (and junk laptop cells are godawful for traction power - better than lead acid, barely), but too small for anything serious. If you are serious about off grid stuff, you'd either be using a radically larger bank, or lead acid.
So... still kind of lost.
I rebuild lithium ion battery packs professionally and am quite involved in electric bike and off grid circles - I'm happy to help you not blow yourself up, but I can't make sense of your goals.