tj90

I dont see any difference in the 1 resistor or 2 resistor solution. You can argue that the 2 resistor solution in parallel means more wire connections that are a weak link etc. If there is an opportunity to reduce the circuit complexity by substituting 1 component versus 2 - it probably would make everything more robust. Another consideration is the available space of the component. The single component will be bigger and may not fit correctly.

I looked into subbing an aftermarket power resistor versus factory a few years ago. I decided not to do it since it was unclear what the resistor value should be (I believe it was printed with one value but it was actually something else). Also, I thought about the opportunity to relocate the resistor to a less corrosive environment like under the front hood but didnt have luck with that either... I scared myself into thinking that maybe porsche thought this through and there was adequate air flow in that area.. I sheepishly installed another factory one that Im sure will die in a few years...

The germans during this vintage never liked to use relay circuits. I laugh that we have !!AMPS!! running through our toggle switches and 100W power resistors. Of course, these things are prone to failure and never understood why they chose to do it the way they did....

nile13

Garth, I'm no expert on power resistors, but I'm sure that teh single 50W 0.5 ohm heat sunk resistor will do the trick just fine.

I have a nasty feeling that the failure mode on OEM resistors is not as much heat realated, as it is moisture and possibly mechanical distruction. Unprotected porous ceramic just doesn't feel right to me.

Although I have a replacement for my A/C resistor that I bought from Sunset for some reason, I'd like to go the metal route. It's just that to make one I need to buy the resistors plus shipping, buy shrink wrap, buy the paste, buy the M3 tap and die.. you see where I'm going. If I were you, I'd put a kit together without the OEM connector, as this will be the expensive part. Let us crimp our own. Just include a bit of extra shrink wrap.

NP993

+993

Please do so, for those of us who are afraid of doing electrical work! Count me in for one. (I would call dibs on two, but got rid of the resistor for the oil cooler and am using a fan switch override.)

Paul902

Hey Garth,

glad to see the final results. Looks good. I like your choice of heatshrink colour, but wonder if it could be available in Riviera Blue, as well.

I'm planning to take my bumper cover off over Christmas break to assess the state of my resistors and do some electrical measurement. I want to see what the maximum current is across the resistor before deciding if one 50W'er is enough, or go beefy, like you.

brucec59

Excellent point! We don't want our cars looking like this one:

perlfather

Let me say how much I admire Garth for the ingenuity, however, i have seen some big problems when engineers tackle a problem in an unfamiliar environment. (I am a guilty engineer, having received 3 degrees in EE in Cambridge, MA, however, I have accumulated a bit of experience after many unintended "exciting" exothermic events.)

I would hate to see a car look like this:
https://rennlist.com/forums/showthre...ferrerid=18025
Note that this fire in the engine compartment started 5 minutes after running the car from a COLD condition.

I think there is now a bimetallic switch in the ballast resistor that opens the resistor if it overheats. If there is no such thing this would be a good idea. Have a look at the pictures before changing the design of something that has this type of potential damage.
https://rennlist.com/forums/attachme...-c4_fcover.jpg

kb

You Sir -
are a certified Rock Star. Brilliant job! We should nominate Garth for some award!

Paul902

I just ran across the TSB for oil cooler ballast replacement that states the new part should have a cooling plate (ie heatsink). From the TSB diagram it looks like a washer 'bout the diameter of the ballast resistor itself. This suggests that Porsche still feels 50W rating is sufficient, but needs a bit of heatsinking. Garth, you've certainly provided that!

nile13

Perlfather, you bring a great point about the internal bimetallic switch.

We could, conseavably heat up the resistor and see if it goes open at some temperature. I'm not sure how to do this logistically before crackig the resistor or melting the wires first, though. Cooking it in water will probably not be suficient temp What are your thoughts?

Or can we find the actual specs somewhere? Garth, you might want to look at your crumbled resistor for signs of any opening mechanism for overheating condition. I'll look at mine carefully when I get it out.

Garth S

Thanks all for the constructive dialogue: I appreciate the point regarding overheating, and in that context, stated in the opening post regarding the OEM ceramic resistor that ...
"The resistor is a short coil of nichrome wire crimped to a fusible link ( silver capsule shown), with dual leads crimped at either end - all in steel crimps: the whole assembly is potted into the ceramic cup with a porous, plaster like compound "

In that first pic, the 'silver capsule' is shown sitting on the fractured center resistor: Actually, that 0.5"x 0.125" item is more likely a thermal fuse, not a fusible link ( former reacts to temperature, latter to current) - the type of thing that is found in all coffee pots, hair dryers, etc. Like matches, they work once only. It may be a resettable bimetallic safety, but it looks an awefully like the thermal fuses that pop and send us scurring to buy new appliances .

I scraped the cement/plaster like setting compound out of a dead ceramic just now ( pictured below), and there is the thermal fuse and ~5 coils of resistive wire - another 'fuse' from the cracked resistor is sitting on top: while I have not (yet) sacrificed one of the aluminium cased power resistors, I suspect that they do not have such internal protection.

Severe oxidation/rust is evident in the three ceramic units I have, and is quite evident in the pics: my assumption was that corroding contacts increased the resistance well beyond the 0.5 ohm specified .... and played a significant role in causing the units to overheat and fail: that opinion may be incorrect.

In light of the above, I opted to double the wattage vs the ceramic unit, and add a modest heat sink: with only a brief 3 minute running test per side, the new resistors barely changed temperature .... if at all.

Regardless, risk of overheating is a serious issue; therefore, what can be suggested as a comprehensive test?? - or is there a bimetallic safety that can be sourced to attached to the module? etc. For a start, I can jumper the relays to run on low ( resistor controlled) speed , and using an IR gun, chart temps vs time until stable temps are indicated. Other ideas?

tj90

I suggest accelerated environmental stress testing. Something like 85C and 85% RH WITH bias. 1000Hr is the military gold standard. I believe I have access to a chamber over the holidays. I may be able to make this happen. We can force your resistors into Malaysian-like humidity and accelerated heat with constant power and see how they hold up. We can set power supply to current compliance and see how well they can handle the stress. If the power supply rails or fuses, youll know.

Maybe a salt spray test as well for you guys near the beach or in areas with salted roads.... I dont have access to these tests.

To be extra safe, you can put a fuseable link in series with the resistor. Youll never allow the current to reach a point to be dangerous.

nile13

Garth, if you have a thermocouple for your Fluke and can stick it into the fins of the metal resistor with some thermal compound, it'll be a very accurate temperature indicator. Then I would just jump a relay and run the fan at low speed for some time. Preferably at idle, so there's no extra air flow over the resistor and thermocouple.

Having said all that... I'm looking at the innards of that ceramic resistor vs. the finned 1 Ohm resistors... and I'd be surprised if the metal resistors will be much above 50C. At least that's my prediction.

BTW, the specs for the resistor should, in theory, provide max temp at max load without external heat sinking in ambiant temperature of 25C or so. Perlfather, is that a quoted spec for power resistors?

klr10

If you're worried about excessive temperature, then why not put one of these onto the heatsink and wire it in series with the resistors? Just a thought...

perlfather

Good idea KLR10!
Ok lets look at some of the specs on one of these wire wound high power resistor:
From the ARCOL (Mouser electronics catalog) spec sheet a 1ohm 50 watt resistor (with proper heat sinking and thermal peanut butter etc) model HS50 with heat sink at 25C shows a surface temperature of 150deg C at 50W! That is hot. If you select a 100W resistor and run it at 50W the surface temperature is specified at 50C. Much better. So keep in mind that these resistors get very hot at their maximum power rating and get hot at half their power rating. Perhaps the next step would be to determine the absolute maximum current (both steady state and in case of some failure) that would be present.

brucec59

The 50w rating is WITH heat sink (as noted above). You could use a thermal fuse like the one found inside the Porsche part. But you'd have to be sure it is physicallly coupled very tightly with the resistor(s). At this point, it becomes too much of an engineering development project, and that's one reason I went with the Porsche part. Other reasons involve resale, which will probably never happen, but I have to consider it, too.

It would have been so much simpler - not to mention more reliable - if Porsche had simply used two-coil fans.