Blower Motor Resistor Fix: PWM

I have a 2012 Tacoma that has eaten several blower motor resistors. After replacing the resistor connection as well as the blower, it ate another one in 2 months. So I began researching other solutions. The result is to use a pulse wave modulator (PWM). $20 on Amazon. Now I have variable speed on the blower motor! If there’s any interest in seeing my setup, I will post some pics.

 

I'm interested...

 

Please post part # etc.
TOMB

 

The part # from Amazon is 2001712013.

The first thing I had to do was understand the factory wiring system. As this took a couple of hours with a multimeter, the drawing below is how I understand it:

Next is the part itself:

This is a pulse wave modulator (PWM). It creates square wave form dc voltage that the blower (and LEDs by the way) recognizes as an average voltage. In other words, the voltage is effectively varried with the potentiometer with very little heat generation, unlike the blower resistor. This PWM is way over-engineered for the given amperage and should last many years. The hookups are plus and minus for the battery and plus and minus for the blower. The Potentiometer knob clips into the board’s connections, but the wire for it is only 5” long so some cat5 wire might be needed to extend this.

Next is how to power the PWM. Since the factory system utilizes a relay toggled positive wire (black with white stripe) that runs all the way to the blower, I didn’t bother any of the original wiring to the blower switch. This allows the relay to still function and give the new system fused power. To do this I first located the wire harness that runs to the blower motor. I chose a place under the dash to cut both wires in that harness and laid the connection end of it aside for now.

I soldered a 12 gauge wire to the black with white stripe wire (harness side, not connection side) and ran it to the PWM battery positive terminal. The PWM battery negative terminal gets grounded to the chassis. This takes care of the powering issue.

Note: I put a yellow wire nut over the white wire w/ blue stripe that I cut, also visible in the above photo. It’s not needed anymore. This effectively keeps the old hvac system intact and ready to hook back up if ever needed.

Next is the simple task of running new 12 gauge wires from the PWM to the blower motor and soldering them into the original connection wire harness from above. Black w/ white stripe is positive and white w/ blue stripe is negative. I used red shrink tubing as well.


The last step is to decide how to mount the PWM and knob. I put mine behind the area where the knob is below:
Mounting the knob in the accessory slot took a lot of dremelling and drilling. But it will fit! So the PWM is just sitting in behind there on the plastic garnish. I might have to improve that later if overheating becomes an issue.

The operation of the system is simple. Just turn on the original blower knob and that powers up the PWM. Then you can adjust the knob to any speed you want. It’s that simple. Even though there is an off position on the knob, there’s no need to ever turn it off as the hvac system will power it off automatically.

I hope this helps!

 

Good job. I would for the immediate future keep an eye on the heating when running at a lower speed. Unused energy has to be expended somewhere. There is a reason for all of those heat sinks on the board.

I agree there will be reduced heating, but having this PWM unit laying inside the dash on plastic would be a concern for me.

 

Indeed it is a concern. I may well go back and put the PWM inside a project box. There’s just not a lot a room up in there. Maybe I will relocate it and put a PC cooling fan on it.

 

The "W" in PWM is for "Width".
The reason this is important is in explaining how PWM works;
Consider an imaginary PWM that operates at a frequency of 1/60 Hz -- 1 cycle per 60 seconds, i.e., 1 cycle per minute.
When the PWM is set for a 10% duty cycle, it means that it is on for 10% of the cycle and off for 90%. That means that it will be on for 6 seconds, and off for 54 seconds.
20%: On 12 seconds, off 48 seconds.
30%: On 18, off 42
40%: On 24, off 36
...
100%: On 60, off 0.

So the width you are looking at, is the width of the on pulse within the cycle.

I'm not particularly fond of that setup. Works in a pinch with an off the shelf "kit", but it leaves a lot to be desired.
The capacity of that board is also massively overkill, and the control interface is quite limited.

Not sure why they need so many transistors on it, unless they used BJT's?
Take for example FQP30N06L... that N-channel FET will handle 32 amps continuous, or 128 pulsed. In other words, for a 60 amp capacity, two of those in parallel will easily do the job. Not 12. For a blower motor, you're probably looking around 5A, so one of those FET's will be more than enough, and won't even need a heat sink.

I also would, for my own implementation, use a microcontroller to manage the PWM duty cycle. With an MCU, you could maintain the factory switch, but without the resistor. Take the MCU's 3.3v to the factory switch's input, and the various outputs from the switch to digital inputs on the MCU. Then you take another digital pin on the MCU as a PWM output to the transistor's gate pin.

Basically.... Adafruit trinket M0 $9 + FQP30N06L $2 = $11. Costs less and more flexible than the drok board.

 

In the case of an FET, there really isn't any "unused energy" unless you're operating it at below threshold voltage -- which in a PWM design, you aren't.

There are a couple of reasons for all those heat sinks on the board. The first is probably that they're using BJT's rather than FET's. That's the only reason I can come up with to explain why they need so damned many of them -- that board is non-reversing, so there is no H-bridge requiring lots of transistors. The second is that they are running the PWM at 15 kHz, which is quite high for motor control -- should be fine at 2-500.

Transistors don't make much heat when they're "fully" switched on, but when the gate is under the threshold voltage (or the base in the case of a BJT), they become partially conductive and end up producing a lot of heat. So when you're building a PWM amplifier with an FET, there are two things you do to minimize heat; (1) limit the number of times that the gate voltage transitions between high and low, and (2) reduce the time that the gate voltage sits between 0 and threshold.

(1) is done by selecting a lower frequency.
(2) is done by minimizing capacitance on the line and maximizing the driving current.

Sometimes it helps to use a smaller transistor to drive the larger one's gate, however, some modern MCU's are still able to do a pretty good job.

 

It has eight NCE7190A MOSFETs and 4 schottky rectifiers (MBR20100CTG) to protect from induction loading. They use all those heat sinks for passive cooling at 50% duty cycle. As tested, at 200 watts of load, the board uses 215 watts. I can live with that. The factory blower resistor consumes a lot more than 15 watts at lower speeds! The heat sinks have no noticeable heat production at full load for over an hour. Overkill? Yes I admit that. Can it be done better? Yes I believe so. Maybe my application is just a starting point. I wanted something that wasn’t going to burn out quickly and for $20 this fit the bill. I did research this board ahead of time. Here’s a test video I found:
https://youtu.be/a6LLbJtnPPA
Either way it’s a lot cheaper than $35 blower resistors.

 

8 NCE7190A? That's enough to take 720 amps continuous. Holy crap that's overkill.

 

I guess better overkill than underkill!

 

Yeah but for an 11 amp draw, I probably went over the top.

As a side note, the head lights no longer dim momentarily when I flip the fan switch to high.

So I’ve read a lot of the threads on here about the 2nd gen Tacoma blower problems. I've been considering how this problem started in the first place. Obviously there are weak points in the system such as the connectors. I suspect though that the culprit here is not the capacity of the system itself but rather the lack of dealing with the voltage spikes from the inductive loading of the blower. This can easily cause arking across switches and connectors as well as burn out resistors. This is another reason the PWM makes sense here. Those Schottky Rectifiers keep inductive loading from backing up into the hvac wiring and components.

 

I wonder why the existing resister fits up into the blower air stream... heat dissipation perhaps?

 

That is the only reason. Dissipate unused power as heat on the lower speed settings.

 

Remember the old style resistors that were a resistive coil in the air stream? I'm sure you do,

 

Maybe

 

Update on the project:

After running the PWM for a couple of days, I decided to relocate the board to someplace I could keep a watchful eye on it. At first it didn’t seem to be making any heat, but after getting it out in the open I discovered it was. Just not a lot. I got a laser thermometer on it and a volt meter. Here’s what I got:

PWM Testing @ 86° F ambient.


12 volts 118° F

9.5 volts 112° F

8.0 volts 96° F

6 volts 87°

Those voltages corespond to the factory switch settings closely. I may end up putting a micro pc fan on the enclosure.

 

but what about the wire nuts?

 

With a few resistors to build voltage dividers it would be really really easy to reuse the stock fan knob. No microcontroller required.


Also that pwm board is so overkill. I love it.

This might be another one to consider:

$10 RioRand Upgraded 6V-90V 15A DC Motor PWM Speed Controller 1PC

https://www.amazon.com/RioRandTM-Up...?keywords=pwm&qid=1552186669&s=gateway&sr=8-7

 

Wait ..as in analog speed control?