What's new
What's new

Advise on motor speed control


Jun 3, 2008
Milan NY
This will be a long post

First off machining is a hobby for me. I am an electronics technician for a living and hope to fabricate a new circuit to fit into the old motor controller housing

The long story

I am looking for information for a motor speed control for the table feed for my Rockwell milling machine. The machine uses a Robbins and Meyers universal motor (see photo) and when I received the machine it was connected to a {dc-1} motor controller housing that had been gutted and some sort of home-brew circuitry (a bridge rectifier on a heat sink and a lamp dimer module rated for 150 watt bulbs).

The lamp dimmer module was blown (on/off worked but no control) and the bridge rectifier was making about 160 volts (normal for a bridge rectifier but a little high for the motor in my opinion)

The motor works fine with 4 leads coming out (2 for the brushes and 2 for the field coils) and reverses with swapping the correct wires with a double pole double throw switch.

My first thought was to use a router motor speed controller from harbor freight which does work but is going to require me to play with the potentiometer value as the lower third of the scale results in no movement thus giving me a compressed range for the speed control, also the speed changes quite a bit from no load to loaded when cutting .

My new thought is to reuse the original controller box and install a full wave rectifier (not a bridge) and use it to power the field coils as a permanent magnet (similar to the magnets in cordless power tools) and only changing the voltage and polarity on the armature for direction and speed. My hope is by having a constant magnetic field in the field coils I will have better speed regulation between load and no load.


1) Does this sound correct or am I missing something ?

2) Without the field coils in series with the armature I suspect I will have to add a resistor to act as a current limiter through the armature ? And if so how do I calculate value and wattage ?

Unfortunately my Audels electric motor book doesn’t cover a lot on universal motors.

Any thoughts and advice will be greatly appreciated as I work on more milling projects hand cranking the table is wearing thin



  • DSCN0367.JPG
    903.2 KB · Views: 5
  • INBOX 410836.jpg
    INBOX 410836.jpg
    909.7 KB · Views: 5
I'm going to run it on dc and can easily convert the output of the router speed control to dc
My complaint is the difference in speed between load and no load
After looking at the original rockwell speed control schematic i suspect it has always fluctuated speeds under load and i am trying to come up with something better
As a "series" motor, it inherently has a large speed range from no load to full load. That just goes with the territory.

You are really not able to do much about that unless you rewind the field to operate on high voltage/low current, because series motors have a high current field that drops relatively little voltage.

The full motor current goes through the field windings, so the field current varies, as do some basic motor parameters on account of it. A series motor's highest speed is dependent on the residual friction in the motor and due to "windage". It's lowest speed is directly affected by load, but the "natural" speed varies with field current, which is the same as motor current. At a heavy load, the motor's natural speed decreases, because the field is stronger. Series motors are very good as traction motors, and in any application where there may be a heavy load.

I suppose you could step down the mains voltage with a transformer, rectify that, and run the field with that output.

A constant field would turn the motor into a form of "shunt" DC motor, which inherently has a much more stable speed from no load to full load. But the full load torque is going to change (be less).

Most speed controls will not be any better. Only one with speed feedback can maintain a steady speed as the load changes. There are simple ones with a crude form of speed feedback, and there are ones with actual rpm feedback, which can be quite good.
My complaint is the difference in speed between load and no load
In theory you can make a an external PID controller to make an negative droop compensation so that when the load goes up you increase the reference. Open loop controller but better than nothing. I bet it could work quite well with some experimentation. Toughest part is getting current feedback into the controller.

But that is really no different than IR comp - you just need a whole bunch more of it. If you had the drive schematic you might be able to modify the drive for more IR comp.
My thoughts about my suggested wiring scheme is my experience with reusing treadmill motors and battery operated tools i don't get the big shift from load to no load
Those motors have permanent magnate fields and only change the voltage to the armature
I thought by having a constant voltage on the field coils i could simulate the permanent magnets in the handheld tools
Remember this is only a 1/7 horsepower motor for the table feed
Yes, constant field current gives a constant magnetic field, just like the magnets. That is a "shunt-wound" motor.
They have a much more stable speed, because the magnetic field is stable, not changing with load, as with a series motor.

Converting a series motor to a shunt would motor is not easy, and there are likely much better ways to attack the problem.

One is to pretty well ignore it.... it's not the end of the world. And motor speed is self-limiting to a degree, since faster feed leads to a bigger load on the motor, which slows it down.
This is the circuit hardinge uses for their HLVH feed motor. I believe there's some inherent feedback to keep torque constant over varying speed.


  • schematic_1.jpg
    492.3 KB · Views: 7
Yes, that has feedback from the back EMF of the motor.

A good idea, but not equal to an rpm-based feedback, which can be very steady indeed.

You may note that it is using a shunt wound motor, which is pretty stable to begin with. The performance with a series motor will be better than nothing, but may not be as good. The series motor tends to multiply the effect of small speed changes, something between doubling and squaring.
It might be possible to measure the voltage across the field windings under a range of conditions, then connect them to an external power supply. The armature can then be connected to the KB drive linked above, which I believe include back-EMF compensation.

Given the motor is 1/7HP, it's about 105W and therefore probably rated at a touch over an amp. I expect that something like a 5V or 9V 1.5A switchmode supply will be fine, but do measure.

The motor doesn't particularly care about if you slightly exceed the armature supply voltage; what matters is that the current remains within the rating, and most good motor drives have protection to ensure that.