Blower motor controller died again, the fan kept running a maximum speed if with the ignition off and the HVAC off, quite worrying really. This time it was the Power MOSFET dead short (brown out) its now be replaced and all is working again.

Please note that the unit is NOT a resistor pack, at least not on the 2005 H2, not sure where this terminology came from?

I've reversed engineered the entire design, only 1 datasheet eludes me the AA539B 20 pin SOP device, if anyone has it it would be useful.

Most of us in the group will have issues with this unit at some stage so I can help save you the cost of a replacement, replacing the MOSFET is quite easy really and very cheap, not more than a few dollars.

Hope members find this helpful....

 

Yes - I've seen others also mentioning a resistor pack but my understanding the the GM trucks higher end air setups with the digital control panel like our Hummers have is that the blower is actually PWM controlled (PWM signal from BCM to a FET amp to the blower). Many other vehicles including the GM trucks with the basic HVAC do use a resistor pack for the blower though.

That being said - good luck finding a datasheet for that. Just like some of the devices in the BCM - there are no datasheets to be found. Probably would have to carefully scope a known working unit to try and determine exactly what it doing (frequencies, pins, etc).

 

You are correct it is PWM, there is just a single input which is a 34Hz, 0.5V to 5.5V signal with varying "Off" periods, the longer the period the greater the fan speed (see chart)

The fan output voltage I've yet to check..

Controller Inputs are Black=0V, Red =12V, PurpleWhite =Speed Control Signal
Ouputs are Red=12V, Black = Drive Signal

The design is simpe but wierd, the MOSFET effectively "grounds" the blower motor, so if it goes short-circuit there's is nothing to stop the fan running continuosly i.e. no fail-safe

Technology has moved on and there may be a better ways of doing this.

 

What might be a good idea would be to change the blower 12V to be a switched source (key run position using a relay if needed) instead of constant so that if the thing does short at least it will only run when the ignition is on. One of the catches with the many ground triggered systems that GM vehicles use. You might also find more info on other vehicles forums. GM used the same digital climate control on uplevel trim versions of the full size trucks and SUV's as well as the mid size suv's like the Trailblazer and Envoy. I've run across posts before where people have had the exact same short and fan runs all the time problem. Only quick solution is to yank a fuse out to kill it until its fixed.

 

That would be an effective, fail-safe solution.

When I first repaired my unit (damp got in) I wasn't sure if the MOSFET case connected directly to the heatsink. As a caution I added a heatsink isolator + silicon compound.

This time when the FET blew I simply replaced it and made a better job of insulating it, but alas only a few hours ago the FET blew again!

Repeated the process and again at max fan speed it blew again.

So now I've mounted the FET directly to the heatsink and "touch-wood" it seems OK.

As the controller mounts inside an all plastic vent it should not present a problem but not the heatsink is "live" i.e. directly connected to the fan's black cable. the question is " Is this normal?"

If anyone has a controller to hand and can measure this it would be useful to know.

Also I scoped the fan output and it is DC not PWM (see revised graph)

 

Odd way for them to set it up that way but excellent info regardless. Could just about make your own knowing that it is taking the PWM signal and using that to control a variable DC voltage output. That is getting above my head though.

 

Thanks for posting this detailed info!
I have the same controller and the original heat sink compound is non-conductive. Thus, the heat sink is not "live".

Can you tell me what the MOSFET part number is?

 

The P/No for the MOSFET is IRFBA1405P and you will find it on eBay (at least) or from RS Components (USA) for $15 for a pack of 5-off (depends what country you're in)


Heatsink compound is used to improve thermal transfer between surfaces, though it is insulating (has to be) it cannot be relied on. If you measure between the heatsink and the controller's black wire you'll probably find they are connected (lucky if not)

Hope you find some MOSFETs :-)

 

https://patents.google.com/patent/US20030063900A1/en

 

Very interesting find, well done for fining that. The article provides clues to the internal working of the final chip, though the final chip pinout is completely different.

All that complexity yet when the MOSFET fails the result is the worst possible outcome, the motor won't switch-off !!