Questions about: Resistors in series with motors = lousy performance

[mmagliaro]

Please bear with the diatribe.  The questions come at the end.

Let me start by saying that I am aware of the pitfalls of putting a resistor in series with a motor, to allow
using 3v or 6v motors in a 12 loco.  The main problem is that the resistor will drop more voltage as the load
on the motor rises, so the motor speed will be erratic under load (i.e. it climbs a hill and slows much more than
it normally would because the resistor drops more voltage).

I have had good luck using resistors in series with coreless gearmotors in the past for 2 reasons:

1. With a gearhead, the coreless motor draws very little current (perhaps 50 mA) and that current doesn't change much under load.
2. I typically am only using a low-ohm resistor (i.e. 33 ohm or so) just to "knock of a few volts".  I am not trying to limit
12 volts down to 6. 

-----------------
Now, I have a 3v coreless gearmotor.  I thought I might be skating on thin ice using 180 ohms or so to keep
the current limited.  I figured I might see bad slow-downs on hills and such.
Well, it's worse than that.  The low-speed performance and start-up of the motor are terrible.

At first, I thought it was the engine chassis.  Then I took the motor out and observed how it runs on
the bench with and without the resistor.  There's absolutely no denying it.  I can get the motor to start and run
*much* slower without the resistor in there.  And this is free-running on the bench with no load on it!

I am puzzled by this.  It draws about 20 mA either way.  And there is no load on the motor.
I could even allow that just before it starts, it draws a little more current, so the resistors drop more,
playing this vicious cat-and-mouse game until finally the motor starts to run.
But then, I would expect that I could back off, slow it some, and see it run slowly.  No way.

Without the resistor in there, I can bloody well run it so slow that I can count the revolutions with my eye.
No way can I come close to that with the resistor in there.

What is going on here?

-------
BTW,
I am going to ditch the resistors and use a pair of back-to-back zener diodes.  That creates a stable voltage drop
and should work much better anyway.  And yes, testing with those on the bench, the motor doesn't exhibit this weird
behavior.  A 6.2v zener causes the motor to start at just above 6.2, and it can crawl like it should.
And with a 50mA max load, 6.2 x .050 = only about 300 mW, so a 1 watt zener can handle this easily and it's not very large.

[peteski]

There is no magic or mystery.  Resistor is a linear device, its behavior is predictable (and based on the resistance).  It is the motor/mechanism which is the oddball here.

You have the means to measure voltage and/or current of this circuit.  If you plot some voltage/current graphs of that motor/mechanism you will see that the resistor just doing its thing.  How does the motor/resistor combo behave if the motor is free-spinning (not driving the mechanism)? I'm actually surprised that you haven't tried that yet.

The other big question is whether the motor is different (brand/model) from the other motors which exhibit better behavior when used with resistors?  Also, what are its ratings vs. the better behaving motors?  How about static resistance of the windings?

In the end, you are well aware that using resistors is asking for trouble. You were getting away with doing that and now when a motor behaves in a way which is expected you act surprised.   

[mmagliaro]

There is no magic or mystery.  Resistor is a linear device, its behavior is predictable (and based on the resistance).  It is the motor/mechanism which is the oddball here.

You have the means to measure voltage and/or current of this circuit.  If you plot some voltage/current graphs of that motor/mechanism you will see that the resistor just doing its thing.  How does the motor/resistor combo behave if the motor is free-spinning (not driving the mechanism)? I'm actually surprised that you haven't tried that yet.

The other big question is whether the motor is different (brand/model) from the other motors which exhibit better behavior when used with resistors?  Also, what are its ratings vs. the better behaving motors?  How about static resistance of the windings?

In the end, you are well aware that using resistors is asking for trouble. You were getting away with doing that and now when a motor behaves in a way which is expected you act surprised.   

Pete, you missed, or misunderstood the part, where I said I tested it on the bench.  I meant I tested the bare motor, out of the mechanism, on the bench, just free-spinning with no load on it.  And under those conditions, it still exhibits this bad behavior with the resistor in there.  The current is 20 mA in both cases (with/without resistor).  The only difference, as you would expect, is that the supply voltage has to be higher before the motor moves.
The motor can start on 0.24 volt on its own.  With the resistor in there, is starts at about 2.9, and the speed is much higher than it is without the resistor at 0.24.

Pretty weird.  With the zener in there, it works beautifully.   With a 6.2v zener, it starts at just about 0.25 above the zener voltage.

As an aside, the zener solves another problem - the one I talked about in the other thread about the pulses riding on top of the smooth DC.   When the voltage is, say, 5.2, the half-wave pulses have an amplitude of about 2 volts.
So the peaks are hitting 7.2, and the zener is effectively chopping the supply on and off as the pulses rise and fall over the zener break-down point.  So the motor gets these nice, round, half-wave pulses of about 1 volt and it just
*crawls*. 

[peteski]

I did misunderstand. When you said you bench tested it I thought it was still installed in the locomotive.

Still, the problem can be easy explained by Ohm's Law. Resistor is a linear device. Motor is not a linear device and in this case this results in the poor running characteristics. Zener Diode is highly non-linear and it will regulate the voltage much better, regardless of the current draw.   Your 3V motor for whatever reason must exhibit current draw curve which is different (or the changes are more pronounced) than with the 6V motors).

I still think that you might want to take some voltage/current readings of the 3V and 6V motor setups and graph them to see if you can maybe see the different power draw characteristics of those motors.