Upgrading antique automotive heater components

[Fingolfin]

Hello,
I hope this isn't too far outside this forum's usual fare.

My pride and joy is a 1959 Morris Minor. As you can imagine, it has a pretty simple electrical system composed of mechanical relays and such, and usually I have no trouble working on it. However, I'm homebrewing some upgrades to the heater, and I've run into a quandary.

From the factory, the Minor's heater uses a simple continuously-variable wire-wound rotary wiper rheostat/potentiometer to control motor speed, with only two prongs - 12V power in, and variable power out. The motor also has two wires, variable power in and ground.

I've replaced the original motor with the blower motor from a mid-'80s Jeep (PN **broken link removed** from O'Reilly Auto Parts). I selected this motor because it had the right dimensions to fit into the original heater case.

The Jeep motor is much more powerful than the old one, so it draws more current (I do not have figures at the moment but can test). Wiring the Jeep motor into the original circuit and actuating the rheostat, the first "full on" (12 o'clock) position gives roughly 3/4 motor speed, and rotating the rheostat further cuts speed down to about 1/2 by the 3 o'clock position, and then zero by 5 o'clock. Any further overloads the rheostat (smoke!).

The Jeep motor was originally powered through a four-position switch (counting "off") rather than a continuously-variable rheostat, and each position was wired to a metal-coil resistor (so three resistors of decreasing resistances). It seems to me that this resistor would not work with the continuously-variable rheostat in place of the four-position switch, since only one of the three resistors could be wired in. Common four-pin relays won't work, because they're on/off rather than continuously variable.

I have considered transistors, 555 timer with pulse-width modulation, etc., but do not have the technical know-how to choose or implement a solution. Do you have any suggestions?

 

[alec_t]

Welcome to ETO!
There are many motor controllers available from e.g. robotics parts suppliers. Something like these, for example should suit.
Since the starting /stall current of a motor is much higher than the full-speed running current (the normal rated current) you should allow for this when choosing a controller.

 

[JimB]

Why not use the switch and resistor unit from a Jeep?
They are designed to work with the motor that you have chosen.

I had a Morris Minor myself, about 50 years ago. Where does time go to when you are enjoying yourself?

It met a sad end, when I wrapped it around a tree near Rugby one Saturday night.

JimB

 

[Fingolfin]

Thank you both for your quick replies!
Alec, do I understand you correctly that I should measure the amperage the motor draws when close to stalling (i.e. at lowest speeds), and then choose a motor driver board that exceeds that amperage? Would such a controller be wired in line between rheostat and motor?
Jim, I have considered that - and indeed have already bought the resistor to experiment with. I don't want to use the Jeep switch 1) because it won't fit into the Minor heater frontplate without modification, and wouldn't look original and 2) I really like the smooth continuous action of the original rheostat, as opposed to the discrete three speeds of the Jeep switch. I may end up having to use it anyway, but if it's possible to stick with the old rheostat, I want to.
Sorry to hear about your Minor! They're wonderful, characterful little cars. Mine, named Mog, was my first car (I'm 26), and I'd be driving it today if someone hadn't broken the windshield in early July!

 

[alec_t]

The controllers shown wouldn't require the heavy-duty rheostat.

Alec, do I understand you correctly that I should measure the amperage the motor draws when close to stalling (i.e. at lowest speeds)

That technique wouldn't work if you still had the rheostat in series with the motor, because the rheostat would be limiting the current. Is there a plate/label on the motor which states the running current or wattage? Failing that, do you have a digital multimeter to measure the DC resistance of the (disconnected) motor?

 

[Fingolfin]

The motor has no markings except the part number, but I do have access to a multimeter.

The controllers shown wouldn't require the heavy-duty rheostat.

I'm afraid I don't follow you here - is the controller board replacing my rheostat? I hope to use the original rheostat to control the motor speed (through whatever intermediate wizardry necessary).

 

[alec_t]

is the controller board replacing my rheostat?

Yes. Your rheostat passes all the motor current (several Amps) so needs to be heavy duty and wastes a lot of power. Those controllers make use of a much lighter duty 'rheostat' (actually a potentiometer) which only needs to pass a few mA.

 

[crutschow]

Can you replace the rheostat with a small potentiometer and still make it look original?
The post then controls the module which regulates the motor current.

 

[KeepItSimpleStupid]

I played with this https://www.jameco.com/z/K8004-Vell...h-Modulator-Kit-Control-DC-Motors_120539.html on an old car. It's probably not the best solution. It does have "soft start". The motor did "sing". The windings were loose and it made audible noise.

Now suppose, you can get a PWM solution to work. The use of a variable voltage that changes the RPM usually provided by a potentiometer.

It may be possible to change the rheostat to a larger value or it may be possible to use the low value rheostat and create a variable voltage. The OEM resistance value may be hard to measure.

It might be worthwhile asking if the rheostat goes from high to low speed or vice versa,

I've fixed car heater motors by replacing the brushes from a vacuum cleaner repair place. I did a shim and cut to fit.

Fun problem: Motor was held by friction on a rubber ring. Cold temperatures would cause the motor to fall and thus hit.
Every once in a while I had to push it up. I found out the fix too late when it was no longer available.

 

[Fingolfin]

Hmm. Do I understand the consensus correctly that there's no way to make use of my original rheostat? My thought was I could somehow isolate the old rheostat from the motor's current pull and use it to actuate a beefier potentiometer sort of device that would then control the motor speed - or, maybe in clearer terms, that a bit of circuitry could sense the rheostat's resistance (as it changes by the knob), and then alter the motor's power supply in proportion, without the motor current passing through (and harming) the old rheostat. Perhaps that's the pulse-width modulation KeepItSimple mentioned.

The old rheostat has an off position at 11 o'clock. At 12 o'clock it goes full on, and progressively winds down to a very low speed around 9 o'clock, at which point it hits a stop and won't turn further. Luckily both old and new motor are bolt-in, rather than rubber-mounted!

I apologize if I'm not making a lot of sense - my experience with electrical work is all hands-on from (old) automotive and house wiring, so the terminology is often over my head.

 

[KeepItSimpleStupid]

Without quoting what you said, essentially yes:

that a bit of circuitry could sense the rheostat's resistance (as it changes by the knob), and then alter the motor's power supply in proportion

If the rheostat was, say 10 ohms, passing a constant current through the rehostat of 0.5 Amps would give a 0-5 V signal. I asked about high speed first, because that allows the motor to start moving.

10 ohms is difficult to measure without a meter capable of being zeroed or a 4 terminal resistance meter. You can measure the current through the resistor and the voltage across it to get the value, but I would not use the "new motor" to do so. An automotive incadesent bulb would work, but you have to use the ammeter and voltmeter at the same time or use a high wattage low-value resistor and a voltmeter. The voltage across the resistor is proportional to current and voltage is voltage. potentiometer voltage and resistor voltage will give you the resistance of the potentiometer.

PWM is Pulse Width Modulation. The ON time or duty cycle is varied so the motor has the same torque at all speeds.
A non audible frequency is used so you can't hear it. With this approach, the MOSFET does not get as hot as a resistor.

In a first order approach, you can add a power resistor in series with your current rheostat.

With some measurements like;

The speed of the new motor is about 50% of the old motor with the potentiometer at 20% of (10 ohms), The new motor is drawing 5 Amps with 12 input and 8V measured across the motor.

We might be able to make an intelligent guess for the resistor. It will be a power resistor and it will get hot.

Lots of options for PWM. But you need current and voltage.

As a "Side note" the resistance of the motor will be slightly dependent on brush position and it would essentially be the wire resistance. The motor does initially draw that amount of current briefly on turn on. The steady state current does depend on load (amount of cold air). A PWM unit has to be from 2 to 5x the measured steady state value. The 5x may just be a surge specification.

Info:
**broken link removed**


it's probably better to buy the control that accepts an analog input and for US to design the way to get from the Automotive pot to an analog signal.

 

[KeepItSimpleStupid]

I did some hunting and came up with this: https://www.pololu.com/product/1383/resources which may be way to complicated.

Handles 23 Amp with out a heat sink and up to a 40 V motor AND it takes a 0-3.3 V analog input. 18 V is too tight of a limit. You would need to reverse polarity protect and add a few transient supressors to work in an older automotive environment. It's "likely" to work. I did not look at every detail.

I'm thinking an LM317 type of circuit e.g. https://www.bristolwatch.com/ccs/LM317.htm to generate a constant current source so, you could get a 0-3.3 V output signal. Assuming I needed is less than 1.5 Amps. A heat sink will likely be required.

 

[crutschow]

Can you measure the resistance of the rheostat?

 

[Fingolfin]

Crutschow,
The resistance in the rheostat is as follows:
- First "on" position (12:00): 0.1 ohms
- 3:00: 1.3
- 6:00: 3.1
- 9:00: 4.9
- Stop (10:00): 5.6

 

[Nigel Goodwin]

You might look at salvaging a controller module from another car, a number of Continental European cars (such as Citroen) have used fully variable controllers on some of their models, you could probably source one (along with the dashboard potentiometer) at low cost from a scrap yard?.

 

[crutschow]

Deleted.

 

[KeepItSimpleStupid]

So, it you put a 1A constant current through the rheostat you would end up with a 0-5 V signal, but in the wrong direction. With the 0-3.3 controller, you;s have to flip the direction UNLESS you had access to all three terminals of the rheostat. So, a 0-3.3 V is doable. A 3.3 (ccw) to 0 (cw) is easy if you can flip the CW and CCW positions at the rheostat.

crutschow A 5v IA should be able to invert that. It just makes the project messier. Pick one: https://www.analog.com/media/en/technical-documentation/data-sheets/AD8237.pdf

That's something worthwhile to check.

 

[alec_t]

you would end up with a 0-5 V signal, but in the wrong direction.

Could be a workable option, since the fan speed would then increase with clockwise rotation of the rheostat. That seems more logical to me than the original arrangement in which the fan kicks in at full speed and is then slowed with clockwise rotation.

 

[KeepItSimpleStupid]

hat seems more logical to me than the original arrangement in which the fan kicks in at full speed and is then slowed with clockwise rotation.

yes and no. The likelyness of high speed first is to get the motor moving and make sure it doesn't stall. The standard resistor method does use low, med and high, but the low speed resistor is selected to the motor will start.

Flipping maintains the OEM labels.

In any event, it looks pretty easy to find a PWM module with a voltage input. It might be possible to change the voltage representation by configuration. e.g. make 3.3 v 0% and 0 = 100%

The other good part is 1 Amp across 5 ohms is only 5 W. ------> I <= 3.3 V/5.6 ohms < 5 W

 

[KeepItSimpleStupid]

from the pdf of the motor controller:

If mixing mode is disabled, only channel 1 affects motor speed. If mixing mode is set to “right” or “left”, channel 1 is considered the “throttle” input and channel 2 is considered the “steering” input. Left mixing mode obtains motor speed by summing the throttle and steering channels (CH1+CH2) while right mixing mode obtains motor speed by taking the difference of the throttle and steering channels (CH1-CH2).

So, it looks like you can backwards it easily.

The OP/TS could be up and running with an external potentiometer very quickly until we coax him how to build the OEM pot interface. He would heat/defrost and defog quickly. I won't have time to look it over right now.

The designed circuit should likely have either a 3.3. v clamp or one that clamped on the 3.3 V supply. The analog input is not tolerant on overvoltage.