looking for a compound potentiometer - PeachParts Mercedes-Benz Forum

sixto · #1 11-04-2009, 04:03 AM

Where does someone go for extra special pot?

I'm thinking of how to add manual blower control, or more precisely blower speed adjustability to my 87 300D similar to the blower control in an early W140. The early W140 has a thumbwheel with detents at min, auto and max, corresponding to the min, auto and max blower settings in the W124 ACC. What's cool in the W140 is you can turn the thumbwheel so you get a little more or less air than the auto setting. In the W124 it's gale force, auto or butterfly f@rt.

To accomplish this, I need a special kind of potentiometer which I imagine looks/works something like this -

I see nothing of the sort in the Digikey catalog and I can't do the custom work required to two basic pots to roll my own. I was thinking an audio fader comes close; but again requires some custom work. Would be nice to put that idle fader thumbwheel to use. There are C4C W140s at the PnP but all the PBUs are gone.

Does anyone know of an off-the-shelf pot like I described?

One concern is the ACC might continuously challenge the adjusted blower setting putting me in a never ending battle for comfort. I hope it's tolerant of small adjustments.

Thanks,
Sixto
87 300D

KarTek · #2 11-04-2009, 06:09 AM

I think you just need a "4 wire" pot.

However, if it's any kind of standard electronics pot, it probably doesn't have the capacity to soak up the load from the fan motor. Does the w124 have electronically controlled variable fan speed? If so, you could probably use a small pot and splice it into the electronics to bias the speed up or down as desired. You'd have to actually go into the CCU though and be pretty savvy with circuits to determine where to put it.

If the car doesn't have electronic controls, you'll have to build a chopper for the DC with a variable frequency to run the fan.

Failing both of those, you'd need a giant pot with ceramic insulation to handle the heat, spliced into the leads for the fan.

mpolli · #3 11-04-2009, 04:07 PM

Quote:

Originally Posted by KarTek

if it's any kind of standard electronics pot, it probably doesn't have the capacity to soak up the load from the fan motor.

Correct

mpolli · #4 11-04-2009, 04:10 PM

Quote:

Originally Posted by sixto

Does anyone know of an off-the-shelf pot like I described?

Doesn't exist. There are center-tapped pots but they are pretty rare. There are no power center-tapped pots that I have ever heard of. That is not to say that a person with money could not have one made. You just will not find it in any catalog on or off the shelf.

Billybob · #5 11-04-2009, 04:33 PM

Both the 140 chassis and the 124 chassis cars use solid state PWM blower motor controllers AKA "porcupine". So there would not be a power issue as it relates to the blower motor itself. I would suspect that the potentiometer in this application is simply a control signal for the CCU processor, the signal is an input and it is at some point compared to one or more other inputs, likely sensors. The CCU processor uses these various inputs to determine necessary output controls i.e. blower motor speed, monovalve frequency, AC compressor clutch, etc.

If the application in question 87 124 blower motor control updated to a stand alone infinitely variable speed control is the goal there are many PWM DC motor controls that should be available to do that. Disconnect the CCU control wire at the three pole bulkhead connector and connect the stand alone controller output in its place. Adjust the blower motor speed via a control pot located in the cabin.

The original question seems to assume a resistance based voltage di

If the desire is to integrate such a function into the stock controls while maintaining stock form and function it would be very difficult/expensive to do so.

I am somewhat surprised that someone in the aftermarket industry has not yet developed a completely programmable microprocessor CCU box to replace the 20+ year old OEM controls. This is the perfect application for the modern micro controller, reading sensor states, processing according to an algorithm, and then outputting states to necessary driver circuits as needed. I suppose the development and manufacturing costs can't be supported by the limited market demand.

It appears the original question diagram illustrates a resistance based voltage divider speed control, this is completely impracticle. Even 123/early 126 chassis cars use a switched resistor ladder controller because a high power pot would have been prohibitively expensive.

al76slc · #6 11-04-2009, 04:49 PM

My company imports cheap pots. This would not be something in our product line.

You may try these people (no relation with us)

http://www.etisystems.com/singleselect.asp

http://www.vishay.com/resistors-variable/potentiometers/

Hope this helps.

Al

Billybob · #7 11-04-2009, 04:50 PM

Quote:

Originally Posted by sixto

The early W140 has a thumbwheel with detents at min, auto and max, corresponding to the min, auto and max blower settings in the W124 ACC. What's cool in the W140 is you can turn the thumbwheel so you get a little more or less air than the auto setting. In the W124 it's gale force, auto or butterfly f@rt.

This function is without question a design feature of the CCU processor and programming and not a function of the input itself. If the CCU has not been designed to accept/recognise and process to a suitable output state this type of input signal it will make no difference having a different input.

mpolli · #8 11-04-2009, 06:07 PM

Quote:

Originally Posted by Billybob

a high power pot would have been prohibitively expensive.

Yes, one that does not burn up...

sixto · #9 11-06-2009, 01:37 AM

Thanks for the responses.

To be clear, I'm not trying to alter the PWM output of the blower controller/porcupine to the blower motor, I'm trying to alter the voltage signal output of the PBU to the blower controller. It's a signal that varies between 2V and 7V. I don't know what current DC signals typically generate but I imagine it's a tenth or an amp or less.

I'll put the context aside and focus on center tap pots such as the model P271 described in http://www.bitechnologies.com/pdfs/p270.pdf This clipping looks like the inspiration for my diagram, though I've honestly never seen it before:

I don't know what CW and CCW mean other than the physical terminals of rotation but I'll treat them as terminal designations.

Question - what voltage will S see if CW has 2V, CT has 5V and CCW has 7V? Assume S is 20% of the distance between CT and CCW and resistance is linear across the sweep range.

Thanks,
Sixto
87 300D

Billybob · #10 11-06-2009, 02:36 AM

Quote:

Originally Posted by sixto

Thanks for the responses.

To be clear, I'm not trying to alter the PWM output of the blower controller/porcupine to the blower motor, I'm trying to alter the voltage signal output of the PBU to the blower controller. It's a signal that varies between 2V and 7V. I don't know what current DC signals typically generate but I imagine it's a tenth or an amp or less.

The "controller/porcupine" is nothing more than a high current semiconductor. it is dumb it has no intellegence function. To use a water analogy it's simply a large valve. The CCU in dash controller sends it a smaller lower voltage/amp control signal, the controller/porcupine essentially amplifies that signal. The CCU signal does not vary in voltage it is a Pulse Width Modulated signal, the pulse widths vary and as a result the greater the pulse's width the greater amount of time the controller/porcupine is turned on and the greater amount of power the blower motor receives and the faster it turns. Pulse width determines the percentage of time the current flows, greater percentage faster blower speed.

If your goal is to control the speed of the blower motor the only way is to vary the percentage of the pulse's width. The controller/porcupine is only on or off, it will require a threshold voltage and a nominal current to turn it on, if you try to reduce the voltage you will get below the threshold and it will not turn on. Too high a signal voltage can damage the controller/porcupine, although in this automotive syste the Vin control voltage and the Vmax of the semiconductor are probably the same. I'm not sure what the semiconductor exactly is, could be a MOSFET, SCR, Thyristor or something else.

I'm going to presume you have not looked at the control signal from the CCU to the controller/porcupine with any type of oscilloscope? That is what you need to see what the signal actually is, if you attempt to measure the CCU signal with an analog meter or digital VOM you will only see the averaged power of the signal represented as an unsteady voltage value.

If you hadn't already seen this, here's an old write-up done by Steve Brotherton which shows what the PWM signal to the monovalve looks like. The signal from the CCU to the controller/porcupine will be similar. The output of the controller/porcupine will have the same frequency and pulse width but will likely have an amplitude of +12 volts and you can't see it but it will be flowing a much larger current. The blower motor uses a 30 amp fuse if I'm not mistaken, so that means the motor can pull up to 360 watts of power at full speed. The electrics and electronics used to control the ACC stuff are high frequency digital On-Off rather than the old fashioned linear ramped style of doing things.

http://www.peachparts.com/Wikka/AcDiags

sixto · #11 11-06-2009, 02:40 AM

Quote:

Originally Posted by Billybob

I'm going to presume you have not looked at the control signal from the CCU to the controller/porcupine with any type of oscilloscope? That is what you need to see what the signal actually is, if you attempt to measure the CCU signal with an analog meter or digital VOM you will only see the averaged power of the signal represented as an unsteady voltage value.

A lot of what you said went over my head but this statement says all I need to know. Thanks!

Sixto
87 300D ... stuck with MB's blower speed until I find another 140

mpolli · #12 11-06-2009, 01:59 PM

Quote:

Originally Posted by sixto

Question - what voltage will S see if CW has 2V, CT has 5V and CCW has 7V? Assume S is 20% of the distance between CT and CCW and resistance is linear across the sweep range.

Depends on the source impedance from the device supplying the CT signal, vs the load presented by whatever value of pot you choose. (and of course that can change as you turn the pot). Also we would need to know the source impedance of your 2V and 7V references. Essentially it as a kind of electrical tug-of-war and we need to know how strong everybody is.

Now I am a little confused since your first drawing showed the pot supplying the motor directly and now you say it only provides a signal to a controller.

I just read BillBobs post and if this signal you have is PWM (which I have every reason to believe he is right) then your plan will not work.

If you need to receive one PWM signal and then create another different one to replace it, that can be done, but you would need to design a circuit to do that which would be somewhat complicated...

mplafleur · #13 04-19-2010, 08:24 AM

Quote:

Originally Posted by Billybob

I'm going to presume you have not looked at the control signal from the CCU to the controller/porcupine with any type of oscilloscope?

I actually looked at the signal yesterday with an oscilloscope while diagnosing another member's fan problems.

The signal coming from the CCU ('86 and upward models only) is NOT a PWM signal, but a DC voltage. There are 6 fan speeds that I know of, low = 1, auto = 2-5, high = 6.

The voltages I measured were:

Off = 0v
1 = 1.2v
2 = 2.0v
6 = 7.0v

I did not measure fan speeds 3, 4, and 5.

When I get my clamp-on current meter back, I will be able to measure the current on the control signal. My guess is that it is very low. It is probably going to a high impedance input.

To do what you want, a simple 7v regulator connected to a single turn pot should suffice. Maybe a 5k ohm pot. Connect the regulator to the full CW terminal and ground to the other end. The center tap goes to the fan controller.

The 5k pot is just a guess. I'll know more when I know how many milliamps the factory circuit draws.

mplafleur · #14 04-21-2010, 09:33 AM

Quote:

Originally Posted by Billybob

I am somewhat surprised that someone in the aftermarket industry has not yet developed a completely programmable microprocessor CCU box to replace the 20+ year old OEM controls. This is the perfect application for the modern micro controller, reading sensor states, processing according to an algorithm, and then outputting states to necessary driver circuits as needed. I suppose the development and manufacturing costs can't be supported by the limited market demand.

I've been think of doing exactly that. I'm trying to conceptualize what it would look like to the user. I would have to make the front end fit the same holes in the center console.

Fully automatic like the original?
Fan speed override?
Keep the same buttons?
Digital readout?

whunter · #15 01-17-2014, 01:11 AM

Quote:

Originally Posted by mplafleur

I've been think of doing exactly that. I'm trying to conceptualize what it would look like to the user. I would have to make the front end fit the same holes in the center console.

Fully automatic like the original?
Fan speed override?
Keep the same buttons?
Digital readout?

Did you get a chance to breadboard this?

.