multiple blower relay/control module failures

First, let me say the collective knowledge of this sight has saved me much money and is truly appreciated.
I've had about five blower control modules go out on me in the last two years. I've researched and found many postings related to troubleshooting blowers vs blower Im but little on how to identify why my resistors continuously fail. I'm a novice in electrical troubleshooting - I replaced several relays but I don't believe they were bad. It's a 99 mounty with auto temp.

the two thoughts that come to my mind are a bad ground or perhaps a bad blower motor. where are you getting your modules from?

Thanks... the first came from O'Reilly, the next two came from Napa (the second under warranty), and the most recent from O'Reilly again. I also changed the blower last time thinking that must be the culprit. That was in Dec and the module failed by the end of Feb. They don't "appear" bad or burnt or anything - they always look identical to the new ones.

Yeah from what Ive read Explorers eat through those blowers

With auto climate control doesn't it use the transistor pack instead of resistors, like this one:?

http://www.oreillyauto.com/site/c/detail/MOT0/YH1835/02627.oap

If so it's probably the main darlington transistor on the heatsink that failed, an NPN 50A 60V TO3, part # MJ11028G

http://www.digikey.com/product-search/en?x=0&y=0&lang=en&site=us&KeyWords=MJ11028G

Personally I'd try to replace it myself if you have a beefy soldering iron, partly to lower cost and partly because if doing it yourself you can inspect and clean the heatsink mating surface with steel wool and apply plenty of heatsink thermal grease to make sure it makes the best contact with the heatsink that is possible.

I wouldn't mind if it ate 1 every year or two, but they're too expensive to replace every 2-4 months.

Yes, that's the one J_C... thanks for sending the part #... I'll order the transistor and swap out. I assume I can use some thermo grease left over from a computer build. To confirm it was the transistor that failed, do I check for a change in resistance while warming it up with a heat-gun?

To Koda's thought about a bad ground, how do I check that?

What I would do to check it is pull the module out of the firewall, leaving it connected, then adjust the climate control so it should be running the fan and use a multimeter to measure for drive current (actually it's a voltage measurement, between the pin and ground) on the base pin (pinout provided on the datasheet on that digikey link I gave, the casing itself seems to be the collector connection).

The datasheet isn't clear about which pins are #1 and 2 on the first page but you can see they are offset at the end of the datasheet under the package dimensions diagram.

Edit:This brings up an issue that I hadn't thought about previously, that if they had electrically insulated the casing from the heatsink then you have to make sure that remains insulated, although from the pictures it looks like the rivets are carrying the collector current and so they need a good, clean contact to the PCB.

If there is no base voltage then something is wrong before the transistor. If there is base voltage, and collector voltage, but no emitter voltage, then the transistor has failed. If there is variable base voltage but the emitter voltage is constant, it also signals a different kind of failure, a short rather than open fault but that would leave the fan running but not the same range of variable speed.

If it's an open fault it also won't be getting nearly as hot if it heats up at all. To be clear, either open fault or shorted fault both mean the transistor is bad.

I don't have access to a picture of other components or a schematic of this module to advise about checking other components on it. Certainly if there are resistors they should be checked, and especially check for broken solder joints since this module does heat up quite a bit.

Bad ground could be checked by measuring resistance between motor ground and a good/clean chassis ground. It should be very low like less than an ohm but you may need to zero a meter to get rid of contact and meter lead resistance if you're worrying about exact value rather than only that it is low.

The main concern when soldering in a new one is that they seem to have used quite a large amount of solder on the PCB, so take care not to excessively overheat the transistor while soldering it in. Work fairly quickly soldering one lead, then letting it cool down before proceeding to the next lead.

I'd consider using stainless bolts, nuts and lock washers instead of what looks like a rivet to secure it so you don't have to drill it out every time it needs replaced if this happens again. Then again that raises cost while rivets are dirt cheap. Rivet or bolt it before soldering so there is no stress on the solder joints.

Computer heatsink grease should be fine.

J_C

Thanks for the tip. I swapped out the transistor yesterday and the module works again. I bought three so I should be good for the next few months. Still stumped on why they're failing. But thinking that troubleshooting should be done with a working control module. Thanks again for the pointers.

Dex

^ Glad to know it's working again. That module runs hot, then they used an economized cast aluminum heatsink. It probably worked fine in their laboratory tests but they don't care what happens after the warranty period.

As for the aftermarket parts, they might have been inferior quality or poor heatsink interface or something more insidious like a parts supplier that was using counterfeit transistors. I consider Digikey a trustworthy source so hopefully it will last longer this time.


It's a bit of a shame they opted to use a TO-3 packaged transistor because TO-247 is much easier to find or make a better heatsink for.