By on May 23, 2012

 Geek alert!  This one is tech-heavy!

TTAC Commentator Skor writes:


Not a question, but it’s good to know that as long as there are cars there are people who will come up with hacks, no matter how complicated things get.

I have an 1992 Cadillac Seville and the blower motor fan disintegrated and jammed up the blower motor, which in turn shorted out the blower resistor. The squirrel cage disintegrated. I checked inside the blower housing to see if there was anything loose, but I could find nothing except pieces of the fan. All of this stuff was OE, no one has been in there since the car came from the factory. I’m sure someone at GM thought that saving a nickel on plastic was a good idea. Well, I bought a new motor/fan assembly and installed it. Wasn’t too bad, about $40. When I tried it, it was a no-go. Next thing I do is check the fuse (30 amp) it was fine. Then I checked the blower motor controller. It was blown. Crap! Apparently when the fan blew up, it jammed the motor, which then took out the motor controller. I’m thinking to myself:

“Nice engineering. The components on this circuit blow before the fuse does. WTF is the fuse for?”

So I call the Caddy stealer and he tells me $225 for the control module. Crap! I Check Autozone and they want $180. Crap! So I get my Google-Fu on and I end up at a Caddy enthusiast forum. A dude on the Caddy forum is an electrical engineer and he was so pissed when the same thing happened to his Caddy that he came up with this circuit that you can build from cheap discrete components.

I got my google-fu on and found a dude who hacked a new circuit together for about $10. I bought the parts but haven’t soldered the thing together yet. The link appears to be dead, and I can’t find the original thread anywhere. Fortunately, I took pictures of this hack, and still have them on my hard drive. You are welcome to use them. Please note: I DID NOT DESIGN THIS CIRCUIT. I took a course in basic electronics when I was in high school. If you give me a schematic, and parts list, I can build it, but I can’t design it. This circuit was designed by an electrical engineer, and I don’t know his name.

The connector block on the controller has 6 pins, but only 4 are used. There are only 4 wires on the harness plug.


NC————————–Ground———————Input(control signal)

The battery and No-Connect on the left are spaced far from the other connections you have no problem telling left from right.

  • The FET source is connected to the battery through a 30 amp fuse….it’s always hot, even with the key off.
    The FET drain goes straight to the blower motor.
    The TIP3055 base is connected to the control signal coming from the climate control computer. I don’t know what this signal looks like since I don’t have a scope. I put the VOM on it and when the climate control is switched on it can vary from .5V to 8+V.
    TIP3055 emitter goes straight to chassis ground.

The cases of the semiconductors I used are TO-247. Normally I would mount stuff like this on mica (the transistor and FET’s net to be electrically insulated from the sink) with heat sink grease, but radio shack only sells mica for TO-220….too small. I ordered sil-pad from Digi-Key. The sil-pad cost $.24 a piece…..if you use sil-pad, you don’t need heat sink grease. The resistors are ½ watt, 5%, carbon film. After I soldered it together, I coated the bare solder connections with JB weld.

This is what the motor controller looked like when I pulled it out of the car.

I cut the original circuit board off the heat sink; one of those “thick film” things which is great for mass production but can’t be repaired…it would be like tying to unbake a cake. I bolted the discrete bits to the original sink and soldered it all together. Does it work? You betcha ya. I ran it for a half hour outside the blower housing. The transistor stays cool, the 3 FETs get a bit warm but nothing that could be described as hot. This is what the DIY version looks like.

When there is the will, there is a way.

Sajeev answers:

As the owner of several unloved Detroit techno-wonders from the 1980s, I have one thing to say about your last remark:  So true.

Wait a second! I remember your amazing advice to me about my Fleetwood 75 Limo:  you also know your TPS reports!

Look, most components are not that unique, anyone with basic knowledge of circuit boards and possession of factory wiring diagrams can fix just about any electronic bit on a car from the 1970s-1990s.  I remember reading a C/D road test of a loaded out Fox body Ford LTD where the author remarked how difficult it will be to keep the electronics working decades from now. Whatever!  I’ve learned how to fix many a Fox Ford electronic dashboard component, and I don’t even own a soldering gun!

The newer stuff is admittedly much harder, but that will change over time.  I suspect someone within the B&B is already quite savvy with i-Drive interface rebuilding, or similar.

Once again: when there is a will, there is way.  Hats off to you, my good man! The 1992 Seville was/is a wonderful design, and far easier to keep alive without that head bolt munching Northstar motor. You got one of my favorite GM products of the era, and that’s a compliment. So let’s show off your handiwork once more.

Send your queries to [email protected]. Spare no details and ask for a speedy resolution if you’re in a hurry.

Get the latest TTAC e-Newsletter!

23 Comments on “Piston Slap: The Cadillac Hack...”

  • avatar

    Well, my hat’s off to you, Skor, for repairing your car. I pretty much fail anything electrical and the only thing I have been able to repair that has any comparison were the rear window lift components on a 1992 Chrysler LeBaron convertible we used to own.

    The back glass regulators on these cars failed on a regular basis because of cheap components and the motor constantly getting wet and rusting out/shorting out. When mine went out shortly after buying the car 1n 1999, I removed the components, saw the problems and fixed them by hacking cable anchors and brackets from stuff lying around in my tool box and hanging on the walls in the garage.

    My efforts were successful, as those windows never failed as long as I owned the car. The motor? That’s a different tale of woe, hence we got rid of the car in 2007.

    Electrical stuff fascinates me and back in the days when you could buy any electrical thingy at Radio Shack, my buddy and I tried all sorts of stuff, mostly unsuccessful…

    • 0 avatar

      Thanks. I seem to have forgotten the link to the schematic diagram. This was not designed by me, and I don’t take any credit for it.

  • avatar
    Educator(of teachers)Dan

    On just about any car you’ll find interesting examples of strange wiring. On my 2004 F150 Heritage the air conditioning compressor and the reverse lights share a fuse.

    When the body shop guys who installed my flatbed arced the reverse light wires they blew the fuse. They couldn’t figure out why the lights didn’t work until I got in the truck and realized I had no air conditioning.

    The only thing I can figure is that the electrical engineer who designed the circut thought of it as a “safety feature”. Plenty of idiots would drive around without reverse lights but how many will tollerate no air conditioning?

    • 0 avatar
      Roberto Esponja

      On my father’s (POS) 1977 Ford Granada, the A/C and power windows apparently shared a fuse, something we came to find out to our chagrin after a hot summer day at the beach. That 1/2 hour drive back home back in 1978 was NOT fun…

  • avatar

    Nice Job Skor!!

  • avatar

    I worked on a Ford Taurus climate control system with a blower that failed. the motor control had a Motorola transistor to control the blower speed, should be a simple repair.
    Except the transistor could not be bought. Only sold to Ford, the part was unobtainium. A trip to a junkyard found one in a car where the blower was removed already, exposing the part for easy retrieval. 20 dollars later all was done.

  • avatar

    I once replaced the blower motor in my ’83 Civic from one that came out of my friend’s then recently died ’80 Civic (it blew out it’s 4th and 5th gears in the manny tranny) as my blower would only work in one speed.

    Getting TO it was the biggest thing, but fortunately, it was down near the bottom of the dash, not buried way up in it so it could’ve been MUCH worse. That was back in the late 90’s, I think either 96 or 97.

    The most recent “wiring” job was wiring up the pigtails for my new head unit to the Mazda connector that was to plug into the one on the wiring harness for the original radio last Saturday.

    This DID entail soldering and yes, I have the tools, took a while as the wire strippers were a semi rusty pair of needle nosed pliers with cutters as I’d forgotten my wire stripper tool (and zip ties).

    I was installing a new double DIN head unit to update it to Bluetooth, USB and Aux capabilities. The previous radio was a single DIN Alpine with only the iPod cable, which then replaced the factory unit, I think in ’06 or ’07.

  • avatar

    That is some sick handiwork. Good job.

    Electrical problems are the most rewarding to solve. I think it has something to do with all the EE’s at college that went out of their way to make me feel like a simpleton. Whenever I trouble shoot a wiring problem, I feel like I just cured AIDS.

    • 0 avatar

      I bought the car for $1,500 with only 52K on the clock. The main reason the car was so cheap was because the instrument cluster was dead, and the dealer wanted a LOT of money for the repair. I did some research and found out the the power supply in the cluster was defective…..they all failed eventually. I went to the local U-Pull-It place….now out of biz…..and bought a used cluster I knew was good… had a re-manufactured sicker on the back of it, which meant it too failed at some point, and was returned for a rebuild. I pulled the board that contained the power supply from junkyard cluster, and installed it in my original cluster. Voila, been working ever since….total cost $40.

  • avatar

    Old engineering axiom:
    A $50 circuit board will fry itself to protect a 50-cent fuse.

    Certain automobile components seem to be designed never to be serviced, let alone replaced.

    • 0 avatar

      Electronic components are designed to be made as cheaply as possible, which means on automated equipment…there is VERY little direct labor involved. Integrated components like that can not be repaired…….generally speaking. Ever try to replace surface mount components?

      With the exception of VLSI, like computer CPUs, it is possible to build circuits out of discrete components that will do the same thing….as long as you have the space.

      As for really complex stuff, like the i-Drive interface, it’s only a matter of time before someone figures out how to hack an iPad to replace a defective i-Drive interface. In that case, you would need to know coding, as well as how to build the hardware bits needed to put the two together.

      What really pissed me off about this was the fact that the jammed motor killed the motor control module. I mean, a jammed motor armature is not unheard of. Again I would not have been angry if the module was reasonably priced…like $15 or $20, but it was $200!

  • avatar
    schmitt trigger

    The image you have of the original module shows a Sanken Electric module (STK plus a number).
    Sanken specialized in the design of application specific hybrid modules. Not hybrid in the automotive meaning of the phrase, but a module that used both monolithic and discrete components in a substrate which was molded in solid epoxy.

    That technology is so early 80’s…and expensive too, but it was a way to accomodate several functions prior to the modern ULSI chips.
    However, I fail to see what complex functions require to be incorporated to a simple fan controller, why did they use it??

    My theory, based on harrowing experience from a previous job for an electronics tier-1 supplier to the Big 3, is as follows:

    Maybe GM had a large inventory of those modules left over from previous model years, and in those days where accountants ruled supreme, the engineering group was asked to utilize the surplus.

    Now, when the fan assembly’s bill-of-materials was costed by a separate group of accountants, they screamed bloody murder, the thing is expensive!!!, and ordered the engineers to reduce cost. The engineers faced with the conflicting requirements, decided to reduce cost by replacing the steel fan with one made out of papier-mache. Ah! since they were also using a fuse box from a previous project and the only spare fuse position was the 30 amp one, let’s go for that instead of designing a new fuse box with the proper fuse rating.

    Just my theory, but I’m sure it is fairly close to the truth.

    • 0 avatar

      I agree with most of your analysis, with the exception of the fan material. I haven’t seen a steel fan in decades. The problem isn’t that the fan was made out of plastic, the problem started because the fan was made out of the WRONG kind of plastic. The OE fan plastic was as brittle as glass when I took it out. You could easily crush it with your hand. I’m guessing that a more durable grade of plastic would have cost 5 cents more.

      Yes the original module was a hybrid. I cut the module cover off with a Dremel to see what was inside. Unfortunately, I did not take a picture. The circuit board was not potted in epoxy, but in some soft gooey stuff that I assume keeps out moisture and acts to suck up vibrations. The board contained some capacitors, resistors and a couple of integrated components that were not marked, but looked to be the same thing. I assume one, or both of those integrated bits was fried. I had no intention of trying to figure out what they were, or make an attempt to repair the thing.

      • 0 avatar

        Yes, that gooey stuff is the shizzle for keeping moisture away from the electron-carrying parts and has other excellent properties (see below).

        I have worked at several companies where we potted electronic modules, and the hard stuff doesn’t work nearly as well as you might think. Why? Well, it has to do with the thermal coefficient of expansion of the various components in the assembly. If the components on the board and the hard potting compound expand at different rates as the unit heats up and cools down, it causes mechanical stresses, and can actually pull components off of the circuit board (such as large electrolytic capacitors which even today are still through-hole mounted). And forget about using it if you have delicate gold lead wires which connect the electronics to the leadframe (the stamped metal part which includes the external terminations).

        And the hard stuff doesn’t bond as well to the surface of the components as you might think, allowing moisture to creep in and cause electrical leakage and corrosion. I have done failure analysis on a number of parts and have seen this first-hand.


        I always wondered why they used that tacky, gooey sealant, and now I know why. It can handle the widest temperature extremes (-40 deg. F to +200 deg. F for something mounted underhood)

      • 0 avatar

        “Yes, that gooey stuff is the shizzle for keeping moisture away from the electron-carrying parts…”

        A horrifyingly awesome use of the word shizzle.

  • avatar

    I am very familiar with that blower control module; GM started using it on their FWD cars that had the automatic climate controls sometime in the mid-to-late 1980s. I had two cars with it and I also replaced it on my neighbor’s Cadillac.

    You have identified a key issue – no overload (as opposed to short-circuit) protection! One insidious failure mode that my neighbor’s Cadillac had (I had to get onto IATN which is a subscription-only automotive technician forum to solve this one) was that the blower motor PARTIALLY shorted, meaning that it was drawing maybe 30-50% higher than its full-load current but still operating fine otherwise. As you have discovered, the module is not protected from this, and will eventually fail. BUT, most technicians fail to solve the root-cause issue, since the blower motor will still be functional (you have to check the motor current which almost nobody does). So the module is replaced, and it fails again 6 months later (“well, it must have been a bad part”). That gets expensive real quick!

    And the failure mode (constant-on) will suck your battery stone cold flat dead in short order. GM also really screwed up by not putting an accessory power relay in this circuit!

    I salute you for repairing this! I was just like you 20 years ago as a young EE that wanted to repair everything myself, and have found out that it’s getting much more difficult to do so (plus I have two little kids now so I don’t have the time). Now I would just go to my local pick-n-pull, get another module, and make sure they mark it so I can bring it back within 30 days if it doesn’t function properly.

    You’re lucky that your car is a 1992; my 2001 Buick has a multiplexed electrical system, and every module is hard-coded with the vehicle’s VIN. Substituting in another module with a non-matching VIN causes that module not to function, and this can only be rectified by going to your friendly dealer who has the TECH2000 scantool and (for a hefty fee) will reprogram the module so you can use it in your car. This makes swapping in junkyard parts economically unfeasable. As an EE, I am distressed to see the direction that automotive electronics are taking. Cars are becoming like computers where your best bet is to drive one for 3-5 years and then throw it away and get a new one, and that is incredibly destructive to our environment and resources.

    My dad had a 1993 STS that he bought used with 23K miles on it, and he kept it for 12 years – thank God we never had the head bolt issue with it (it was a Northstar engine, you are very fortunate if yours has the 4.9 instead). Beautiful car, wonderful to drive, great fuel economy, but repairs can get expensive and very difficult as well with the Northstar underhood (“Where’s the starter? I don’t see it.”).

    • 0 avatar

      I had no idea Buicks of that era had such complex, badly designed electrical systems; I always thought of them as having dated but durable technology. Guess I was wrong!

      The Seville was certainly a beautiful car with lots of quirks!

      • 0 avatar

        Multiplexed systems are NOT a bad thing, they are a good thing. If you compare the wiring harness of a pre-multiplex car with that of a modern car, you will be amazed at the reduction in complexity. Most components only need power and bus signal connections, ground is through the body. Much simpler, and MUCH more reliable. Compare the MASS of wires that has to go into the door of an older car with power locks and windows to the pair of wires in a new car. But, you need different tools to work on them (some sort of computer). You can buy the tools, and for at least VW and BMW, there are aftermarket equivalents to what the dealer uses.

        People have been whining about increased complexity since the Model A replaced the Model T (my God, that thing has a WATER PUMP!), but yesterdays rocket science is todays normal.

    • 0 avatar

      “You’re lucky that your car is a 1992; my 2001 Buick has a multiplexed electrical system, and every module is hard-coded with the vehicle’s VIN.”

      I’m not very articulate, but here goes.

      Actually, every module is calibrated to work with your particular model and manufacture year….the VIN identifies the calibration or function required of that particular module. Years ago, before modules could be re-flashed, you needed to keep separate modules in stock for all the various calibrations, now the modules are “universal” and are differentiated only by the software that’s installed.

      This is a good thing, because it holds down cost. Fewer parts to make and stock. The problems occur when the auto makers take the attitude that all their code is proprietary, and will not release it to anyone except their dealers. This is a major problem for independent repair shops or DIY guys. This is why we need a national right to repair law.

  • avatar

    Fuses tend to be slower to react to over current conditions compared to a semiconductor device. Also the durability of semiconductors to carry over current has greatly improved over the years. For example, early transistor audio amplifiers would blow out their output transistors if the speaker wires even touched for a brief instant. Today’s amplifiers are very robust to faults in the output wiring.

  • avatar

    A company I used to work for used small (2-5 hp) DC motor drives in their machines. The smaller ones, having no overload circuits, would frequently get overloaded, resulting in blown power output devices. The fuses would always be sitting there looking perfect, while the solid state devices looked like an overcooked dinner. Yes, the fuses are too slow to do any good. They protect the wiring.

Read all comments

Recent Comments

  • dartman: I bought a 7.5kw solar system in 2015 for $15k (after 30% tax credit) I was averaging $225/mo for...
  • dartman: I bought a 7.5kw solar system in 2015 for $15k (after 30% tax credit) I was averaging $225/mo for...
  • Jeff S: Paid about $4.09 for regular at Sam’s Club in NKY a few days ago which isn’t bad especially when...
  • ToolGuy: 13 city / 17 highway / 14 combined “I approve” – Vladimir Vladimirovich Putin (To...
  • ToolGuy: Hey kids, check it out. 1879 (the year) Edison incandescent bulb takes a little while to come up to full...

New Car Research

Get a Free Dealer Quote

Who We Are

  • Adam Tonge
  • Bozi Tatarevic
  • Corey Lewis
  • Jo Borras
  • Mark Baruth
  • Ronnie Schreiber