Teardown: SAAB electronic heater fan speed controller.

[Refrigerator]

While pulling my old crashed SAAB apart i found out that the heater fan speed was controlled electronically instead of a resistor pack.
I thought it was quite modern for a mid 90's car and so i decided to take the module apart and reverse engineer it. But as i traced the circuit out i noticed something strange, that being that there was no freewheeling diode. Weird i thought this module worked with PWM, efficiency and all that right?
Nope it's actually just a huge linear constant current regulator, shame.
There's only one IC inside which is an LM2904 dual op-amp. The muscle is an IRFP054 MOSFET, 60V and 64A continuous at 100^oC is quite nice.
At first i thought that i could use this to control the motor on my homemade electric scooter but since it's linear that's not really possible.
But i could definitely use the MOSFET to make a speed controller. 

[Refrigerator]

And a couple more pics.

[Dabbot]

Interesting! The values for "R3" and "R3?" are a strange choice. The same could be achieved with 8K2 and 10K.
Maybe double-check that bit?

Edit: On second thought, the 16K5 resistor looks to be hand soldered in. Manually trimmed?

[Refrigerator]

Interesting! The values for "R3" and "R3?" are a strange choice. The same could be achieved with 8K2 and 10K.
Maybe double-check that bit?

Edit: On second thought, the 16K5 resistor looks to be hand soldered in. Manually trimmed?

I might have read the colors incorrectly, as the resistors are super small and telling the difference between blue and purple was hard.
But i also think it's manually trimmed, kind of weird that they didn't use a trim pot. Perhaps trim pots aren't good enough for this application? 

[GlennSprigg]

While pulling my old crashed SAAB apart i found out that the heater fan speed was controlled electronically instead of a resistor pack.

Yea, I just had flash-backs of 'old' (older) cars...  To keep the 'resistors' cool, the wiring went to a Block with the resistors, mounted
inside the main fan/conditioner unit body, so that the air-flow cooled it!  Was a locating trap for the uninitiated. 

[tom66]

It's interesting that they still went for a linear regulator despite having all of that circuitry there which could have easily implemented PWM.  Perhaps they were worried that PWM would be audible, or possibly visible in the lighting.   I recall the headlights and interior lighting on my Peugeot would dim occasionally when the wipers swiped the window, if the engine was at a low RPM.

On older Opel/Vauxhall vehicles the resistor pack has been known to cause the vehicle to burn down when the thermal fuse does not correctly interrupt power due to a stalled motor condition.  If the fan motor stalls and the fan is set to a low or medium position, the resistor pack has no cooling and begins to get very hot indeed.

[Refrigerator]

It's interesting that they still went for a linear regulator despite having all of that circuitry there which could have easily implemented PWM.  Perhaps they were worried that PWM would be audible, or possibly visible in the lighting.   I recall the headlights and interior lighting on my Peugeot would dim occasionally when the wipers swiped the window, if the engine was at a low RPM.

On older Opel/Vauxhall vehicles the resistor pack has been known to cause the vehicle to burn down when the thermal fuse does not correctly interrupt power due to a stalled motor condition.  If the fan motor stalls and the fan is set to a low or medium position, the resistor pack has no cooling and begins to get very hot indeed.

The interior lights in my SAAB were PWM controlled and you could hear the slightly audible noise when they dim slowly.
I actually thought this feature was super cool. But then again why not PWM for the fan?
Does switching high currents with PWM have EMI implications?