5
\$\begingroup\$

In the past, Tesla used a fuse on every cell for their car batteries. The fuse was made out of a thin metal wire that was spot welded on the cell twice as well as the busbar. Many fusible alloys melt from 50 to 90 degrees.

Because the fuse is welded onto the cell it wouldn't matter what the fuse current was, you could have a large wire capable of high currents which would melt only when the cell rises over a certain temperature.

What would be better, a lithium-ion cell welded fuse that melts at high currents, or at high temperatures?

What would be ideal fusing temperatures for both scenarios?

\$\endgroup\$
3
  • 1
    \$\begingroup\$ high temperature is simply high current with some time delay \$\endgroup\$ Jul 12, 2016 at 11:58
  • \$\begingroup\$ @ScottSeidman Note that LiIon cells have the " vent with flame" mode which a thermal fuse may be intended to be involved with. (Or may not). Also gross temperature rise under certain conditions. He notes " ... or at high temperatures" and mentions battery temperature. \$\endgroup\$
    – Russell McMahon
    Jul 12, 2016 at 14:02
  • \$\begingroup\$ Sounds like this question is not necessarily about electrical fuses per se, but thermal cutoff. (Wikipedia). As far as I understand, most lithium batteries become dangerous above somewhere around 100 deg C. A high current chunk of fusible alloy that melted all over the battery wouldn't necessarily break the circuit, would it? I guess it would depend heavily on the design. \$\endgroup\$
    – Dampmaskin
    Aug 2, 2016 at 14:28

2 Answers 2

3
+25
\$\begingroup\$

TL;DR - a current-based solution is probably the only workable one in this case, though thermal cuttoff migh be useful and is often implemented, it doesn't work exactly like a fuse.

As mentioned by the users commenting on your question, lithium-based batteries, including the cylindrical 18650 Li-Ions in the Tesla that you mention, become hugely dangerous at high temperatures. An overheated charged battery migh burst, spraying its insides everywhere, start burning ferociously, or both. Google "lipo explosion". Because of this, most responsible designs employ an electronic circuit called the Power Control Module, or PCM, which protects the battery from overcharging, overdischarging, reverse polarisation of the charger and sometimes overheating, requiring a separate thermistor. Keep in mind that this PCM module is not "welded" into the battery but usually attached with Kapton tape (Li-Po) or hidden beneath the sheath of the cylinder (Li-Ion). It is wired in series with the load and you might of it like a smart fuse.

In high-current devices, like drones, there is a risk of overheating because of the device drawing a multitude of the battery's capacity (sometimes 40x or more) from the battery bank. Therefore you sometimes find a thermistor attached to the battery control module and put between the cells - though it does not work like a circuit breaker, think of it more like of a peripheral sensor for the control module. And it's definitely not going in series with the load, at least in no application I know.

\$\endgroup\$
2
  • \$\begingroup\$ Thanks for the infos. I have been studying lithium batteries for some years and so I'm away of PCM and runaway temperature reactions that occur above 120 degrees C and explosions. The issue with Li batteries is that they can put out crazy amounts of current safely as long as they dont get hot, you can pulse 40C through some designs no problem. The real issue per battery is to isolate it as soon as it gets over 60 degrees on the terminals. I thought a temp limiting fuse wire is more versatile than a current limiting fuse because it allows power peaks while preventing overheating. \$\endgroup\$ Aug 8, 2016 at 8:43
  • \$\begingroup\$ a problem battery could technically get very hot with internal resistance error while putting out a small number of amps. \$\endgroup\$ Aug 8, 2016 at 8:44
1
\$\begingroup\$

The fusible links are there for current. They are not there for temperature. The melting temperature of the fusible links is higher than the self-ignition temperature of NMC Li-ion cells. Therefore, by the time they melted, it would be way too late to protect the cells.

Religious wars have been fought over whether a fusible link for every cell is a solution to a problem, or a solution in search for a problem.

In a few words, the two arguments go like this:

Pro fuse: if a cell shorts, the current from its parallel cells blow its fuse, isolating the shorted cell.

Against fuse: in the presence of a BMS, cells do not fail as short; a fuse doesn't protect from internal short circuits (which is the real danger), fuses add cost and reduce reliability. Also, there's no difference between a large cell 100 Ah cell and a block of twenty 5 Ah cells in parallel, yet we don't worry about the former and we do about the latter.

Witness that Tesla changed from one side to the other: they used to use a fuse per cell, now they don't anymore.

\$\endgroup\$
2
  • \$\begingroup\$ Any idea why Tesla stopped using the fusible wire connection per cell, in their batteries? \$\endgroup\$
    – Pro Backup
    Dec 28, 2023 at 19:37
  • 1
    \$\begingroup\$ This is conjecture. They saw that fuses were not preventing mishaps and were adding needless complexity and range loss. \$\endgroup\$ Dec 29, 2023 at 16:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.