Rays logo Timing a motor.

Introduction

I have an old 15x3 electric motor made by MRI. It has some problems in that it has burned away parts of the comm. It is a good motor, and I wanted to resurect it so I hit on the idea of buying a cheap motor with a similar armature and swapping the armatures over.
I found a Yokomo Pro Stock 2 - 15x2, at a reasonable price and set about swapping the armatures over.
Being the sort of person I am, I wanted to know if I had made an improvement by swapping the armatures over, or could I have simply thrown the MRI away and used the Yokomo instead.

Tucked away at the back of the shed, I have a " Buds Racing Products dyno" with which to do some before and after tests.

Buds Racing Products Dyno

This is a handy little tool which appears to be discontinued now.
It comprises a box with some resistive loads in it. You put power into the dyno, from a nicad pack or other power source. You can then take power out of the dyno and run the motor to be tested. This drives a slave motor via a piece of stiff rubber tubing. I have found automotive brake line tube to be ideal.
The slave motor is effectively being used as a generator, power from the terminals is fed back into the dyno across the loads. Readings are taken from a Digital VoltMeter, (DVM), plugged into the dyno box, I use the 20V DC setting.
The dyno and accessories required

This small box used to cost in the region of 80 US dollars. This seems a bit steep for what is in it, so I was quite glad to have only paid 8 UK Pounds on an Ebay auction. At that price I am quite happy with what I have.
So what does it give you.
There are three resistance settings simulating different loads on the motor.

Whats inside the BRP Dyno

It doesn't matter what they are as we are only going to use the dyno for comparison purposes.
The outputs to the DVM are amps drawn by the motor on test, and RPM. The amps figure is the actual amperage drawn. on the 20V scale, multiply the DVM reading by 10 to get the actual amps.
The RPM figure is an arbitrary reading useful for comparisons.
Its usefulness stems from the ability to measure a motor, make changes then measure again. You can see a rise or fall in current drawn and/or rpm achieved.
This is exactly what I intend to do here.

The measurements

Firstly, it is important to minimise all of the possible variations in the measurements. To that end I will not be using a nicad cell pack as a power source.
Things that can affect a nicad pack include ... Without having a degree in chemistry or physics, I resigned myself, not to be able to control the state of a Nicad pack with any accuracy.
I did do some measurements, but all this proved to me was that readings taken using a Nicad pack for a power source are not in any way repeatable. Even by repeaking the pack between dyno runs.
Instead I used a PC Power supply. This is an ATX power supply I had already modified to power my peak detect nicad charger. I had the foresight to put a 5V output on it.
While lower than the 7.2 volts of a chaged Nicad race pack, it will still give meaningfull relative results.

The readings from the dyno will be dependent on which slave motor you use. As it is running in reverse, idealy it needs to be zero timed. It should also not suffer from going out of adjustment.
I picked out an ancient mabuchi motor, which used to come with Tamiya kits many years ago. It is a zero timed, non rebuildable unit.
I figured that if I just used it without cleaning or adjusting it at all, then after all this time and use, it was not going to vary much.

I wanted to have a full picture of the pros and cons of the armature swap, so I took measurements from the MRI motor before and after the swap.
It was also an opportunity to see what effect the timing had on a motor. The MRI has a scale marked on the can like so.

Scale markings on the MRI motor

The results below are based around this scale.

What is the scale

My first assumption was that the 0 on the scale would correspond with 0 degrees of advance/retard.
i.e. exactly 90 degrees to the motor mounting holes, which are between the magnets in the can.
This wasn't the case.
I suppose it is inevitable in mass produced items such as this, but there are minor variations. Measuring with a pencil and protractor, showed that
A) The magnets are not mounted centraly in the can.
B) the scale is not applied such that the zero aligns with the mounting holes.

Mechanical measurement of the zero point

Look at the motor by holding it up to the eye with the endbell towards you. Imagine a line drawn between the two pairs of motor mounting holes, through the central axis of the motor.
This is your hypothetical line of zero timing. Zero is normally marked on the can here.
Rotating the endbell anti-clockwise in the can will advance the timing, Clockwise retards the timimg.
Retarded timimg on an electric motor is considered to be a "Bad Thing". I'd be interested as to why that is.

Magnetic measurement of the zero point

Due to slight variations in manufacture, the mechanical zero point is not always accurate. As I found with my MRI motor, the magnets may not be central in the can, The sticker with the scale marks on may be misapplied, The magnets themselves may have developed variations in field strength, The armature and commutator might be slightly mismatched.
Thus, the only way of truly establishing where zero should be is to measure it while the motor is running.
The zero timing point is where the motor runs forwards, and draws the least current into a constant load
I guess you can do this with a Whatmeter or similar, measuring current draw from a power source.
I have the BRP Dyno and a DVM which I used in conjunction with the 5V supply from my PSU.
Undo the screws that hold the endbell in place. Rotate the endbell to where zero is marked, (or where the motor mounting screw holes are, if you dont have markings).
Hold the endbell in place, and apply power to the motor.
Gently rotate the endbell back and forth, until you find the place where the least amperage is being drawn.
Mark both the endbell and the can.

Measuring the MRI motor

A motor can has an external diameter of 36mm. Application of Pi d gives us the external diameter of 113mm. Dividing that result into 360 degres gives us approx. 3.2 degree for each millimeter of case diameter.
Measuring the markings on the MRI scale showed that the 3 and 6 markings were 4 and 8 millimeters away from the 0 mark respectively.
Using the dynamic/magnetic measurement method detailed above showed me that the magnetic or true zero point was at the -3 mark on the scale.
Thus, when timed to 0 on the scale, the motor was actually running with 12.8 degrees of advance.

Now I could measure against the scale, knowing what was really advanced and retarded.
I measured both rpm and current draw into the three loads provided on the BRP dyno for all points on the timing scale.

        RPM     AMPS  
Motor timed to scale mark True degrees Load 1 Load 2 Load 3 Load 1 Load 2 Load 3
MRI 15x3-6 -6 -12.8 5.53 4.44 3.88 0.54 0.77 0.91
MRI 15x3-3 -3 0 5.86 4.83 4.21 0.43 0.62 0.75
MRI 15x3 - 0 0 12.8 5.92 4.88 4.18 0.44 0.61 0.75
MRI 15x3+3 +3 25.6 5.98 5.00 4.26 0.50 0.69 0.83
MRI 15x3+6 +6 38.4 6.35 5.18 4.46 0.59 0.80 0.97

Then I swapped the armature for the Yokomo 15x2 wind and Re-calculated the true zero point, which this time was at the +3 point on the scale.
I could now repeat the measurements.

        RPM     AMPS  
Motor timed to scale mark True degrees Load 1 Load 2 Load 3 Load 1 Load 2 Load 3
MRI (Yok)15x2 - -6 -6 -38.4 4.06 3.30 2.83 1.29 1.43 1.54
MRI (Yok)15x2 - -3 -3 -25.6 5.25 4.33 3.76 0.74 0.93 1.08
MRI (Yok)15x2 - 0 0 -12.8 5.59 4.56 4.00 0.63 0.81 0.97
MRI (Yok)15x2 - +3 +3 0 6.14 5.10 4.46 0.49 0.68 0.84
MRI (Yok)15x2 - +6 +6 12.8 6.43 5.31 4.59 0.57 0.77 0.92

I finally settled on about 6 degrees of advance so that I minimised my current draw while still operating with some advance.

        RPM     AMPS  
Motor timed to scale mark True degrees Load 1 Load 2 Load 3 Load 1 Load 2 Load 3
MRI (Yok)15x2 - measured at 6 degrees advance +4.5 6 6.24 5.15 4.11 0.44 0.65 0.81

Interestingly, while the rpm figure is about right for 6 degrees, the current draw figure looks to be low.

Measuring the Yokomo

The Yokomo Pro Stock 2 15x2 motor is a non-rebuildable design.
i.e. It cannot have the timing changed.
I did measure the timing statically, by measuring the can and brush hoods. This gave me an approx. figure of 14 degrees of advance.
Putting it on the dyno before I removed the armature gave me the following...

        RPM     AMPS  
Motor timed to scale mark True degrees Load 1 Load 2 Load 3 Load 1 Load 2 Load 3
Yokomo Pro Stock 2 - 15x2     6.15 5.05 4.40 0.59 0.78 0.92

So what have I learned - and was it worth it

Well I know a lot more about motors and timing than I did yesterday.
Ultimately, it was worth changing the armature, because the comm on the MRI was burned and damaged, making the motor stutter at low revs.
Looking at the figures above, shows that I have gained by moving the 15x2 armature to the MRI can.
I can time it to 6 degrees of advance rather than the 14 (ish) which it was running in the Yokomo can.
At 6 degrees, I am getting more RPM for less current draw.
I also have the option of changing the timing. Increasing the advance to 13 degrees would give me about the same current draw as it had in it's original can... but with many more revs.

And how does the new hybrid compare with the original 15x3 wind that the MRI had.
Well it is always going to be difficult to say, because the comm was damaged anyway.
However, inspection of the figures above shows that at an equivalent power draw, I am getting more revs with the new armature.
A sucess all round !

Last updated 29th December 2002

Back to my Home Page