Heater blower transistor failure mystery...

Post in this forum for topics relating to the interior.

Moderators: GRNSHRK, ron, bfons

northNH
Posts: 32
Joined: Tue Dec 12, 2017 11:29 pm
Location: New Hampshire USA

Heater blower transistor failure mystery...

Post by northNH »

Greetings from northern New Hampshire,
I’m stumped and hoping for some insights...

Over the 20 years I’ve enjoyed my ‘86 CSI, I have replaced the heater blower motor transistor every ~5-7 years with the regular old 3055s.
Always been straightforward, always successful without drama or unexpected surprises.

I recently took the car out of ~3 year storage, and after several weeks of normal blower operation (i.e. fan motor smooth, quiet and powerful for both variable and high speed operation), I lost variable speed control.

As I said, no big deal, BTDT. Everything else that should work does i.e. variable AC fan control, no blown fuses, etc.

So I bought two 3055s, installed one, straightforward as always.
After ~30 seconds it “blew” and fan stopped, back to pre-installation non-variable operation. Transistor did get hot, no fuses blown...

Huh? Removed/replaced the transistor, same result.

Got two higher-powered 3771s, and same results, both blew within 30 seconds.
And dammit, everything checks out correctly with the transistor once soldered to it’s heat sink prior to installation.
And not possible to connect the 3 wires to the transistor board incorrectly, even at my advanced age.

Although I haven’t measured the current draw of the motor (on high), I find it hard to imagine it’s the problem, as well-functioning as it’s always been.
It’s acting like the main power supply to the transistor is somehow grounding through the plastic blower motor case, but that ain’t possible.

Suggestions most appreciated.
9D42F778-7D0A-4101-A198-816026B40619.jpeg
9D42F778-7D0A-4101-A198-816026B40619.jpeg (801.57 KiB) Viewed 13472 times
'86 635 5spd 175K Blk/blk orig paint/int
‘88 M6 24K time capsule Black/gray
'92 MBz 300CE Sportlined 175K Blk/wine orig paint/int
‘94 MBz E320 wagon
'62 Volvo Amazon 65K mile time capsule

One lucky guy...
User avatar
hornhospital
Platinum Member
Platinum Member
Posts: 2931
Joined: Sun Jul 11, 2010 6:46 pm
Location: Silverhill, AL
Contact:

Re: Heater blower transistor failure mystery...

Post by hornhospital »

Pretty much has to be excessive current draw by the blower motor.
Ken Kanne
'84 633CSi "Sylvia"; '85 635CSi "Katja";'85 325e "Hazel Ann"; '95 M3 "Ashlyn"
User avatar
86_6series
Platinum Member
Platinum Member
Posts: 1069
Joined: Mon Apr 02, 2007 10:08 pm
Location: West Nyack, NY

Re: Heater blower transistor failure mystery...

Post by 86_6series »

hornhospital wrote: Sat Jun 20, 2020 4:53 am Pretty much has to be excessive current draw by the blower motor.
+1
86-635CSi
96-MB C220 SOLD
98-740i
71-Continental MK3
Pod
Platinum Member
Platinum Member
Posts: 4492
Joined: Tue Mar 16, 2010 8:32 pm
Location: usually in the barn, Kent, England.

Re: Heater blower transistor failure mystery...

Post by Pod »

A few drops of penetrating fluid, followed by some light oil on the blower motor bearings could work wonders.
emac
Posts: 178
Joined: Thu Feb 06, 2020 2:57 am
Location: SC

Re: Heater blower transistor failure mystery...

Post by emac »

I was having a fuse blow every time I turned the car on. Swapped out resistors, cleaned and lubed the fans and everything was fine....until the fuse blew again. It turned out to be the connector to the resistor under the dash. I tightened the female connectors on the plug and it has worked perfect since then. Dont ask me why a loose connection would cause the fuse to blow. No clue. I found it when I had the fan running and jiggled the connector and it popped and blew the fuse.

Just for grins, you may want to check the electrical connector.
northNH
Posts: 32
Joined: Tue Dec 12, 2017 11:29 pm
Location: New Hampshire USA

Re: Heater blower transistor failure mystery...

Post by northNH »

Thanks for all the reply’s; however they have not turned out to be useful...

I decided to replace the (original) blower motor with a new Bosch unit, very reasonable at $62 delivered (FCP).
Although the old motor ran well at full speed, swapping around fuses showed it would blow a 7.5Amp fuse, heat up a 10Amp one, and run fine with a 15Amp one (normally supplied by 30Amp, which has never blown).
Oil to shaft bushings without benefit...

New motor installed, new 2N3771(15Amp) transistor installed.
Entirely as before, motor ran a total ~10 seconds, correctly increasing speed as I slowly increased knob rotation until 2/3 full, whereupon motor stopped just like when I started with my first recent transistor replacement.

Repeated once again with my last transistor, same result. Motor correctly responded to increasing speed control knob until ~2/3 full travel, then motor stopped exactly as before. Transistor warm but not hot when it failed.

To repeat of significance, all throughout, when AC selector switch is engaged, AC blower motor works entirely correctly from fan switch, both variably and at full speed, eliminating a bunch of possible faults.

Wondering if the AC selector switch, which runs a GN/VI wire to the heater blower housing for speed control (GN/BK from the switch for the AC motor speed control) could be malfunctioning/shorting and sending full voltage rather than regulated voltage to the transistor...
Or an issue with the (hidden) blower housing connector?

If AC switch is OK, what should the voltage on this wire look like coming to the transistor when the fan speed control is advanced?

Not ready to spring for a new complete transistor unit yet (as I have no good reason to believe it will be OK), but I’m getting real irritated that I can’t get to the bottom of this simple circuit which worked fine for years with a simple transistor replacement.

TIA
'86 635 5spd 175K Blk/blk orig paint/int
‘88 M6 24K time capsule Black/gray
'92 MBz 300CE Sportlined 175K Blk/wine orig paint/int
‘94 MBz E320 wagon
'62 Volvo Amazon 65K mile time capsule

One lucky guy...
emac
Posts: 178
Joined: Thu Feb 06, 2020 2:57 am
Location: SC

Re: Heater blower transistor failure mystery...

Post by emac »

Did you check the electrical terminals in the connectors? If you have removed them a lot, maybe one of them isnt making good contact and causing resistance in the circuit. I would take a small probe and try to tighten each terminal in the connector. I guess you could put your hand on the wires to see if any of them are getting hot.
User avatar
86_6series
Platinum Member
Platinum Member
Posts: 1069
Joined: Mon Apr 02, 2007 10:08 pm
Location: West Nyack, NY

Re: Heater blower transistor failure mystery...

Post by 86_6series »

The Bentley manual has a complete section on trouble shooting HVAC.
Including the voltage specs.

If you don't have one, I could scan that section and e-mail you.
86-635CSi
96-MB C220 SOLD
98-740i
71-Continental MK3
BenM635i
Posts: 167
Joined: Thu Dec 10, 2015 7:23 pm
Location: St Albans, UK

Re: Heater blower transistor failure mystery...

Post by BenM635i »

86_6series wrote: Fri Jul 10, 2020 11:34 am The Bentley manual has a complete section on trouble shooting HVAC.
Including the voltage specs.

If you don't have one, I could scan that section and e-mail you.
Jumping in on the thread... would be really useful if you could scan the bentley blower fault diagnosis section and send to me... not the same problem but my fan speed switch blew (again) after replacement 2 years ago.

Have previously replaced the big transistor and blower motor spins freely so want to check everything out before i sink another wedge on a new variable switch.
'85 M635 #160
northNH
Posts: 32
Joined: Tue Dec 12, 2017 11:29 pm
Location: New Hampshire USA

Re: Heater blower transistor failure mystery...

Post by northNH »

@86_6series...
Thanks for your kind offer; I do have the Bentley manual, and an original paper copy of BMW’s “1986 635CSi ETM” (complete detailed schematics).

Nothing in either is helpful; no voltage specs for the speed-controlling wire for the transistor from the AC switch...

I haven’t measured this voltage yet (I believe it’s “pulsed”), or better yet, compared it’s value(s) to the other identical-function wire from the AC switch to the AC blower motor transistor, which works perfectly.

Too freakin’ hot to do more than just think about it, even in northern NH...
'86 635 5spd 175K Blk/blk orig paint/int
‘88 M6 24K time capsule Black/gray
'92 MBz 300CE Sportlined 175K Blk/wine orig paint/int
‘94 MBz E320 wagon
'62 Volvo Amazon 65K mile time capsule

One lucky guy...
User avatar
anchored
Posts: 165
Joined: Fri Aug 09, 2019 7:15 pm
Location: Atlanta

Re: Heater blower transistor failure mystery...

Post by anchored »

I am having the exact same problem. I started with the same transistor and moved to the 2n3771 and blew two of them. the seem to blow right around the time I move past 2/3 just like northNH. According to my multimeter its definitely the transistor that is going out.

I have checked the relay and put a new one in with no change.

The motor is well oiled up and running well at full speed.
SHRKATK - Red 87' 635csi
northNH
Posts: 32
Joined: Tue Dec 12, 2017 11:29 pm
Location: New Hampshire USA

Re: Heater blower transistor failure mystery...

Post by northNH »

@anchored...

I have not gotten any further in solving this repeated transistor failure; sounds like you’re in exactly in the same place.

There is also a similar configuration 2N3771G transistor, rated at 30 amps (vs 15A for the 3771, 10A for the 3055) which MAY solve our problem, but doesn’t really explain why these previously suitable lower amp transistors are failing.
Though I doubt it, perhaps we’re dealing with a batch of defective 3rd world crap transistors? Mine were made in Malaysia I believe.

(It appears the 3771G is now used in the “OEM” replacement piece from BMW @~$125; F**K that)

Still a piece of this picture is missing...

If you give the 3771G a try, let me know if it does the trick,
Michael
'86 635 5spd 175K Blk/blk orig paint/int
‘88 M6 24K time capsule Black/gray
'92 MBz 300CE Sportlined 175K Blk/wine orig paint/int
‘94 MBz E320 wagon
'62 Volvo Amazon 65K mile time capsule

One lucky guy...
User avatar
anchored
Posts: 165
Joined: Fri Aug 09, 2019 7:15 pm
Location: Atlanta

Re: Heater blower transistor failure mystery...

Post by anchored »

Also found people using these from some googling.
2n3772 20a 60v

2n5302 30a 60v ~200w This seems like the way I am going to go next.

2n5303 20a 80v
I have yet to replace my motor, but that's my next plan along with the new transistor.
SHRKATK - Red 87' 635csi
northNH
Posts: 32
Joined: Tue Dec 12, 2017 11:29 pm
Location: New Hampshire USA

Re: Heater blower transistor failure mystery...

Post by northNH »

I did not see any difference i.e. blowing transistors before vs. after I replaced my seemingly-good-old motor with a new one...NO difference.

Unless your motor is blowing a 7.5 or 10 amp fuse while running on high speed as I described testing above, do me a favor and don’t replace your motor till you/we get to the bottom of the blowing transistor. It’s probably not your motor, it wasn’t mine...

Why should this simple system/circuit, perfectly happy for years at a time (until most likely vibration necessitates regular transistor replacement) using the same-spec cheap 10 amp transistor over and over again, suddenly need an uprated 30 amp or 20 amp or even a 15 amp rated transistor to survive??
This inquiring mind wants/needs to know...

I’ve not finished examining/testing all the (scant) relevant associated wiring, but believe that when I finally do our problem won’t be your old motor or my new one, or a too-low rating of the transistor.

Happy to hear your further thoughts🤔
'86 635 5spd 175K Blk/blk orig paint/int
‘88 M6 24K time capsule Black/gray
'92 MBz 300CE Sportlined 175K Blk/wine orig paint/int
‘94 MBz E320 wagon
'62 Volvo Amazon 65K mile time capsule

One lucky guy...
User avatar
hornhospital
Platinum Member
Platinum Member
Posts: 2931
Joined: Sun Jul 11, 2010 6:46 pm
Location: Silverhill, AL
Contact:

Re: Heater blower transistor failure mystery...

Post by hornhospital »

It may be helpful to get an amperage draw reading of the motor running at various speeds. Running drag isn't the only thing that kills motors and control circuits. The windings could be failing although the motor spins freely by hand. Just a thought.....
Ken Kanne
'84 633CSi "Sylvia"; '85 635CSi "Katja";'85 325e "Hazel Ann"; '95 M3 "Ashlyn"
User avatar
anchored
Posts: 165
Joined: Fri Aug 09, 2019 7:15 pm
Location: Atlanta

Re: Heater blower transistor failure mystery...

Post by anchored »

I have gone ahead an ordered a 2n5302, mainly because it was cheap.

Will be holding off on ordering the motor. Based on the electronic trouble shooting guide, the only things behind this in the circuit are the transistor and variable resistor for the fan speed controller.

both of those would be strange because the AC fan works fine.
SHRKATK - Red 87' 635csi
northNH
Posts: 32
Joined: Tue Dec 12, 2017 11:29 pm
Location: New Hampshire USA

Re: Heater blower transistor failure mystery...

Post by northNH »

Having tested and confirmed normalcy of all the pertinent wiring and function up to the critical AC switch, the only part of the circuit that remains potentially responsible is the connector (“C1” terminals 1,2,5,6) where the 3 wires for the transistor (power in, power out to motor, and the term #2 GN/VI control wire from the AC switch) enter the heater box to the left of the glove box.

I have not yet laid my hands on this connector, which I believe someone earlier noted this connector as a potential trouble spot.
We’ll see.

Most interested in the result of your strongarm 5302 attempt 😍
'86 635 5spd 175K Blk/blk orig paint/int
‘88 M6 24K time capsule Black/gray
'92 MBz 300CE Sportlined 175K Blk/wine orig paint/int
‘94 MBz E320 wagon
'62 Volvo Amazon 65K mile time capsule

One lucky guy...
emac
Posts: 178
Joined: Thu Feb 06, 2020 2:57 am
Location: SC

Re: Heater blower transistor failure mystery...

Post by emac »

northNH wrote: Thu Jul 30, 2020 10:04 pm Having tested and confirmed normalcy of all the pertinent wiring and function up to the critical AC switch, the only part of the circuit that remains potentially responsible is the connector (“C1” terminals 1,2,5,6) where the 3 wires for the transistor (power in, power out to motor, and the term #2 GN/VI control wire from the AC switch) enter the heater box to the left of the glove box.

I have not yet laid my hands on this connector, which I believe someone earlier noted this connector as a potential trouble spot.
We’ll see.

Most interested in the result of your strongarm 5302 attempt 😍
That was where my problem was. I took some small picks and tools and tightened all the terminals at that connector and haven't had a problem since. Going back through the previous owners records, he had worked on the resistors several times over the years and never really solved the issue. I just stumbled across the solution luckily. I still dont understand why a loose connection was blowing the fuse....but it was. Fingers crossed for you.
User avatar
hornhospital
Platinum Member
Platinum Member
Posts: 2931
Joined: Sun Jul 11, 2010 6:46 pm
Location: Silverhill, AL
Contact:

Re: Heater blower transistor failure mystery...

Post by hornhospital »

Loose connection = high resistance. High resistance where it shouldn't be leads to all sorts of trouble, including blown fuses, and maybe blown transistors.
Ken Kanne
'84 633CSi "Sylvia"; '85 635CSi "Katja";'85 325e "Hazel Ann"; '95 M3 "Ashlyn"
User avatar
anchored
Posts: 165
Joined: Fri Aug 09, 2019 7:15 pm
Location: Atlanta

Re: Heater blower transistor failure mystery...

Post by anchored »

I have 2 extra AC switches laying around, but they are from the e28 and seem to have one less pin. I will give tightening the connections a shot.
SHRKATK - Red 87' 635csi
User avatar
sansouci
Platinum Member
Platinum Member
Posts: 1850
Joined: Mon Jul 08, 2013 10:52 pm
Location: Southampton NY

Re: Heater blower transistor failure mystery...

Post by sansouci »

hornhospital wrote: Thu Jul 30, 2020 11:13 pm Loose connection = high resistance. High resistance where it shouldn't be leads to all sorts of trouble, including blown fuses, and maybe blown transistors.
Think of a loose connection like a spark plug that can develop oxides on the contact surfaces. Oxides tend to be highly resistive and take significant energy to overcome.
Sansouci
84 E24 633Csi Auto, Bronzit/PearlBeige 6997510
93 E32 740il M60 Auto, Alpenweis/Ultramarine
60 528i M30 5-speed Green/Beige (crushed)
71 240Z 4-speed White/Blue (rusty & sold)
65 396 Chevelle 4-speed, Marina Blue/Black (stolen)
User avatar
jps635
Platinum Member
Platinum Member
Posts: 872
Joined: Sat Feb 27, 2010 10:50 pm
Location: NZ

Re: Heater blower transistor failure mystery...

Post by jps635 »

I recall reading discussions about all this when I changed mine years ago. I believe Chris Wright detailed the various transistors but can't seem to search on his posts now, however came across this, which may be of help.


Post by ScottAndrews » Mon Mar 14, 2005 5:33 am

There is a lot of confusion about this aspect og the 6ewr HVAC system..., and a lot of broken fans out there.

This circuit works as follows:

First of all, a transistor is a device that has three terminals. These are called the Emitter, Base and Collector. The transistor is basically a current amplifieing device. The current that flows from the Base to the Emitter controls the current that flows from the Collector to the Emitter. So you put a small current through the Base to the Emitter, and a larger current flows between the Collector and the Emitter.

Inside the heater control assembly (the complex thing on the dash that has the heat and fan dials), there is a small transistor. This is connected to the variable resistor that is controlled by the fan dial (usually a ring around the clock on the older cars). When you move the ring, you are changing the current flowing through the base this smaller transistor. This in turn creates a current flowing to the input to the big transistor in the heater. As you change this current, the current through the big transistor changes and this makes the fan run at different speeds. The reason they use these transisotrs is because the fan draws a lot of current, and the variable resisotr woudl get to hot if you tried to control the fan directly.

When the fan dial is in the MAX position, it closes a switch that bypasses the big transistor. This is to avoid passing all the current through the transistor (which would make it get too hot).

So, there are several reasons why the fan only runs on HIGH.

First, and most common is that the big transistor is open circuited. If this happens the variable signal from the control unit doesn't do anything, so the fan only gets current when the MAX switch is closed.

If you change the transistor, you MUST be certain the new one is hooked up correctly. If by chance you cross up the emitter and base it the current won't flow. The emitter and base look pretty much the same, so this is a real possibility. The link below shows a diagram of the transistor case. As you can see, for the TO-3 case the case itself is the Collector. When viewed from the pin side, with the pins offset toward the top, the right pin is the Emitter,and the left pin is the Base.

To check your wiring, the Blue wire to the heater should connect to the Emitter, and the Brown/Green wire should conect to the Collector. For some reason my ETM does not show the color for the Base wire.

So check this before you do anything else. While you are in there, turn on the ignition, and measure the voltage on the Blue wire. This should range from zero to about 11 volts as you turn the fan dial (Wh already KNOW it doesn't, but it SHOULD). When the fan dial reraches max it should jump a bit to 12 volts. Now turn the fan dial and measure the voltage on the base of the big transistor as you turn the dial. Probably you will see little happening. CHange your DVM to Amps (you may need to move the test lead connections too). Now connect the red lead the to base of the transistor, and the black lead to ground. Measure the current between the base of the big transistor and ground as you move the dial.

If the controller is OK, and the big transistor is bad, you should see this current change as you change the dial position (you may have to fiddle with the current range on the meter to see it, since I am not sure how much current is supposed to flow... my guess is it is about 0 to 0.2 amps)

If the current changes with this test, then your problem is the big transistor. If, however it doesn't, then your problem is the smaller transistor or the variable resistor.

You can test the resistor by measuring the resistance between the small transistor base and collector as you move the dial (with the unit out and disconnected from the car). If this resistance changes smoothly as you rotate the dial, the variable resistor is OK.

On my car the circuit board around the transistor was burned.. a good sign that the transistor is bad...To fix the smaller transistor, you have to remove the heater control unit, and partially disassemble the harness on the back. This will expose a small flat transistor that looks something like the TO-220 style shown in the picture. It is held to the circuit board with a small screw. I found a similar (NPN power transistor) at the local electronics shop, snipped the leads and soldered the new one in to the original leads, and held it in place using the screw.

There is a good primer on transistors at this site:

http://www.kpsec.freeuk.com/components/tran.htm
'85 635csi JPS (RA2-66)
northNH
Posts: 32
Joined: Tue Dec 12, 2017 11:29 pm
Location: New Hampshire USA

Re: Heater blower transistor failure mystery...

Post by northNH »

Thanks for your effort, but we’re way beyond that basic primer...

The fact that the fan controller works as it should for the AC blower and the AC switch is working as it should eliminates any fan controller issues before the AC switch.

The fact that the upgraded transistors fry almost immediately (even in my case with a new fan that only draws <10Amps) says wiring, or a really large batch of bad transistors 😏:

It is virtually impossible to mis-connect the 3 wires at the heater blower transistor.

My understanding is the output voltage from the fan control thru the AC switch for both the AC and heater blowers is 12 volts variably pulsed, rather than variably reduced, tho’ I’ve not measured this yet.

Tomorrow I’ll get my hot little hands on the “C1” connector that COULD explain the problem.
'86 635 5spd 175K Blk/blk orig paint/int
‘88 M6 24K time capsule Black/gray
'92 MBz 300CE Sportlined 175K Blk/wine orig paint/int
‘94 MBz E320 wagon
'62 Volvo Amazon 65K mile time capsule

One lucky guy...
User avatar
DesktopDave
Posts: 66
Joined: Tue Sep 30, 2014 12:36 pm
Location: Pittsburgh, PA, USA
Contact:

Re: Heater blower transistor failure mystery...

Post by DesktopDave »

Did you check the fan speed switch transistor inside the fan switch housing? Or that delicate little quarter-turn variable pot? It's made by a German stereo company called Ruwido. They're still in business, but only produce universal remotes today.

I'm told it has dual resistance tracks, but I haven't really dug into it to understand how it works. There's been a few in-depth discussions about that little potentiometer. I recall a detailed R&R writeup somewhere too, but I only have a PDF. There's also been a brave attempt here to replace it with a variable PWM power supply.
'86 635CSi, currently running a B32 with Motronic 1.3
User avatar
anchored
Posts: 165
Joined: Fri Aug 09, 2019 7:15 pm
Location: Atlanta

Re: Heater blower transistor failure mystery...

Post by anchored »

I thought about the variable resistor, but the AC fan works perfectly across all speeds.
SHRKATK - Red 87' 635csi
Post Reply