I have yet to personally see one of these resistor packs that keep burning up - the dealership that sold me my TJ changed it out and tossed the bad one. With that said, has anyone thought about rebuilding one with higher wattage resistors? Is it reasonably possible?
Here are some pictures of mine which is bad according to the missing low fan speed and my multimeter:
Visually it looks fine to me. I was expecting to see some charring or delaminated PCB traces or something. All I see is the pitted and bubbly conformal coating. But the pits and bubbles aren't concentrated over just the low speed resistor trace so I'm not sure if that's related to it going bad. Maybe it has pinholes that let moisture in over time?
I dunno. All I know is my low fan speed doesn't work and my multimeter says that trace has ~15 Megaohms of resistance (versus the <5 Ohms for each of the others).
To answer your question though it would be difficult to solder anything onto this module. Maybe some sulfuric acid would soften the coating and expose some metal. But even once that's done I'd have a hard time fitting any kind of high wattage resistors onto this package.
I think the fan going full bore is roughly 10A. Of course fully on bypasses the resistor block altogether so the Medium-High speed is what passes the most current through the resistor block. So let's say the fan draws 8A on Medium-High. If I did my math right that means the Medium-High resistor would be 0.3 Ohms (which sounds sort of close to what I remember my multimeter saying). P=R*I^2=0.3*8*8=19.2W. So this little resistor block needs to make ~20W of heat on Medium-High.
I think it comes down to what kind of resistors would you use? With that kind of power in this small of a package I think the as-designed PCB traces are pretty cost effective. Yes they do make
high power resistors but they're big and way more expensive than the PCB traces that are going to have to be there anyway.
The best design I can think of is a PWM circuit.
That could definitely fit into this small of a package. And it would make negligible heat--it'd be
way more efficient than just bleeding off power as heat. But our Jeeps don't give the resistor block a dedicated ground and in this case a PWM circuit needs a dedicated ground because it's driving an inductive load (there needs to be a flyback diode). Although maybe the flyback diode could be inserted elsewhere in the loop. But a PWM circuit will always cost more than some simple PCB traces. To say nothing of certain people's aversion (I'm looking at YOU haha) to that kind of stuff.