SaabCentral Forums banner

1 - 10 of 10 Posts

·
Registered
Joined
·
2,515 Posts
Discussion Starter · #1 · (Edited)
There are some old threads with no photos on how to do this, so thought I'd refresh since my controller just failed.

Symptom is that the A/C fan runs at full speed even with keys out. You have to pull a 30A fuse 'C' in the drivers side fusebox to stop it. RHD.

Saab part number for the controller is 5045158.


Remove glovebox and get fan controller out.


Unit and new Mosfets (£4.00 for two on EBAY). Only need one. IRFP054N.


Prise off lid.


Remove blackened mosfet.






New one in, re-assemble and stay cool.
 

·
Registered
Joined
·
21,583 Posts
You could be an electronics whiz..
How were you able to determine which transistor to buy ?
Great pics, now a favorite.
 

·
Registered
Joined
·
2,515 Posts
Discussion Starter · #3 ·
Googled Saab part number.

The mosfet that came out of my controller was frazzled and I couldn't make out anything other than part of a batch number! So thanks to a member on an old thread on another forum I'd never heard of.
 

·
Registered
Joined
·
7 Posts
Great thread and pics. Just did this myself on a 2002 93 se. Some info I can add might be useful: my fan stoped altogether. I did the requisite tests (made sure the fan was still operational and that the control panel was not the culprit) and then decided to replace the mosfet. While waiting for the part I was talking with my brother who suggested I try reflowing the solder on the three pins as vibrations can take their toll on older electronic components. As my mosfet was not burnt I tried this and it did cure the problem, but only for less than a minute! I made a mistake - I tested the fan for too long WITHOUT reassembling the controler module. IT NEEDS TO BE REASSEMBLED WITH THERMAL COMPOUND and ATTACHED TO THE FIREWALL WITH THERMAL COMPOUND or else the damn thing heats up to the point of desoldering itself and baking to the point of cracking, and dropping out of the PCB :eek:. This might be obvious to many or most but I did not think 20 seconds inline would have this effect.

Anyhow, good news is I poped in the new mosfet when it arrived and everything is running fine.

So glad I did not pay the $217 + tax quoted to me by a GTA SAAB dealer for a new module. Easy fix if you can handle a soldering iron ;ol;
 

·
Registered
Joined
·
59 Posts
I can vouch this....

I just replaced the whole resister pack on my '02. I can vouch that the heat generated by this thing is crazy. I hooked up the wiring connector to the new resister pack. Ran it for like 30 seconds to verify speed controls worked. Then I pick up the resister pack to put it back on the firewall, and damn... that pack was hot!!! Highly recommend NOT running the fan if the resister is NOT mounted on the firewall with thermal paste.
 

·
Registered
Joined
·
158 Posts
who doesn't like more pictures

Stupid over priced heater


back side and heatsink


original power mosfet, cabin fan gets near 0 amps and does not spin (bad)
Interesting difference from the note above, mine is a P054, not a P054N


Old and New


Had to use my larger 375 watt soldering gun, butane did not cut it. I know it looks like .... But it works!





Got my parts from: http://www.ebay.com/itm/3857037917?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1423.l2649
 

·
Registered
Joined
·
86 Posts
OK, now I have things figured out and a bit of a warning for those who know enough to get them selves in trouble.

I replaced my bad power MOSFET with a smaller TO-220 (vs TO-247) size package. The FET I used was MORE than up to the task in terms of current rating and the like. However, I had a problem. The screw that holds the FET to the heat sink was a bit bigger than the whole in the FET. No big thing I figured and got things back together. When I put the thing in my car it "worked". The fan came on. My issue was that the fan was off when it should be blowing. However, then I realized the fan was on full even when I turned the ACC off! Turning the car off worked but the ACC controls had no affect.

That's when I looked at the circuit. Like most FETs the old one and my TO-220 replacement both connect the middle pin (the drain) to the large heat sinking back panel. I had assumed the heat sink was electrically connected to the drain pin since I saw no isolation pad between the FET and the heat sink. It turns out I was wrong. The heat sink appears to be anodized or otherwise made non-conductive on the surface that touches the FET. The TO-247 package has a plastic inside of the screw hole so it's actually electrically isolated from the screw. The non-conductive surface of the heat sink isolates the back plane. So my screw had shorted the drain and source pins and effectively bypassed the fet with the fan running full speed.

When I removed the FET I was able to run a quick test and verify things worked again. This test was all of 3 seconds since I didn't want to burn out the FET. So the up side is I proved that my problem was a bad FET. The down side was the TO-220 FET I had handy didn't work. I ordered a part off ebay ($5 because I picked a US shipper to get it hear sooner rather than later) and I feel confident that changing the FET to the TO-247 sized package will fix the issue.

Thanks for all who figured this stuff out!
 

·
Registered
Joined
·
1 Posts
Thanks for the info - my fan was on constantly so I had to remove the fuse. From the info above, I bought a replacement fet (~3 quid each from Farnell order code 9103082), but on removing the old one, it seems the old FET has no obvious issues (no shorts, by putting some charge on the gate, begins to conduct and very similar voltage drops to the new fets I bought) - any other suggestions as to what might be wrong? maybe the control panel? Any suggestion as to control panel tests? Thanks
 
1 - 10 of 10 Posts
Top