Tesla's Charging Circuit and it's Application to Pulse Motors

[TinselKoala]

The mechanical power dissipation of a small rotor on good bearings can be very low. Here is how to measure it with fair accuracy and precision.
First, you need to know the rotational Moment of Inertia of your rotor. The MoI can be calculated by accurately measuring and weighing the individual components of the rotor and doing a little computation.
http://www.youtube.com/watch?v=lpsFcV13uY8
Next you need accurate means of measuring the angular velocity (ie RPM) and time. A chart recorder comes in very handy here, but isn't really necessary.
The energy of a rotating object like a flywheel or motor rotor in Joules is given by
E = (Iw²)/2
where I is the MoI from part 1 and w is the angular velocity from your RPM measurement (translated into radians/sec).
So now you know how many Joules your rotor is storing when it's turning at any given RPM.
So you run your rotor up to a given RPM however you like, remove the power and start the clock. (Actually you would power to something above your start RPM, remove power and start the clock as the rotor slows through it.)  Allow the rotor to run down to another known RPM and stop the clock. The rotor will have dissipated the difference in energy, during the difference in start and stop times. The figure you get is the average power dissipation during the interval. Make the interval shorter and shorter and you approach the true instantaneous power dissipation but you will find it harder and harder to measure precisely.
dE/dt = power in Watts
Often the power dissipation vs RPM will be nearly linear over much of the rotor's speed range, but since aerodynamic drag goes as some higher power of RPM the faster you go the harder it gets and the more power the rotor will dissipate, and departures from linear will happen. SO for fast and/or very light rotors you would want to take a lot of data to make a good plot of power vs RPM.

So.... if your rotor is turning at a constant RPM under power, you look on your graph from the above, determine the mechanical power dissipation at that RPM, and then you know with fair certainty that the rotor is receiving that much power from your source. The difference between what the rotor receives and what you are inputting will give you an efficiency figure.

[MileHigh]

TK:

Wow that clip had real integrals!  (X-squared becomes 2X or X-squared becomes X-cubed over three.  That's your survival pack!) lol

I have mentioned some ways to measure the moment of inertia of a pulse motor rotor before but it's a bit complicated.  Just tonight I finally came up with a friendly way to do it, or at least a sketch of how to do it.

Imagine you have your pulse motor on a table.  On the rotor itself there is a tiny extra wheel that acts as a guide for some thread.  The thread connects to a small pulley wheel so that it takes a 90-degree bend from horizontal to vertical.  You attach a weight to the end of the thread.

So imagine your weight is 25 cm above the floor and the thread is under tension.  You are holding the rotor to prevent it from turning.  You release the rotor and measure the RPM of your rotor with your optical tach when the weight hits the floor.  Or perhaps you do it with more precision, perhaps using a microcontroller as an example.

And that's it!!!  The only measurement to make during the actual experiment is the tach RPM reading and then crunch the numbers!  Easy as pie!

The energy supplied by the falling weight is MGh.  That becomes the energy put into the rotor.  Since the weight hits the floor, you have very sharp "thread tension on" and "thread tension off" "signals" that torque the rotor up to speed and put energy into it.

I think that's a real damn good moment of inertia test that most pulse motor makers should be able to do.

MileHigh

[Farmhand]

Thank you very much Tinsel, I think you might have tried to help me understand what you're saying before,  I can only apologize I haven't soaked it in yet. However with your last post I will study it and see what I can do.

Couple of specific questions. Should I also stop the motor switches from switching during the run down or should I just cut the power (to the coils) ? Also should I allow the motor coils to charge the capacitors or not ? because in reality when the motor is running the transformer action is likely not happening because of the applied potential to the caps from the supply. Which could mean the rotor is seeing more load when running down than what it must overcome to run while powered. Which is kind of confusing me, I could conceivably fudge the result one way or the other in a couple of ways, if you know what I mean. I want to take those things into consideration, but was not thinking totally along the lines of investigating it till now.

I'll study what I need to do in your post and have a think about it while taking the circuit behavior into account, then I might be better able to explain what I "might" see as a way of the result being fudged. At first glance there would seem to be some things might need taking into account. Maybe I'm just over thinking it.

I'm certain you or others could then tell me if it is a consideration or not when I get to that point.

Cheers

P.S. Ahah, maybe if I cut the power to the switches and the coils promptly then the capacitors will remain charged and so rule out any transformer action form the motor coils themselves. I can test that roughly and fairly easily to see. Maybe.

..

[Farmhand]

TK:

Wow that clip had real integrals!  (X-squared becomes 2X or X-squared becomes X-cubed over three.  That's your survival pack!) lol

I have mentioned some ways to measure the moment of inertia of a pulse motor rotor before but it's a bit complicated.  Just tonight I finally came up with a friendly way to do it, or at least a sketch of how to do it.

Imagine you have your pulse motor on a table.  On the rotor itself there is a tiny extra wheel that acts as a guide for some thread.  The thread connects to a small pulley wheel so that it takes a 90-degree bend from horizontal to vertical.  You attach a weight to the end of the thread.

So imagine your weight is 25 cm above the floor and the thread is under tension.  You are holding the rotor to prevent it from turning.  You release the rotor and measure the RPM of your rotor with your optical tach when the weight hits the floor.  Or perhaps you do it with more precision, perhaps using a microcontroller as an example.

And that's it!!!  The only measurement to make during the actual experiment is the tach RPM reading and then crunch the numbers!  Easy as pie!

The energy supplied by the falling weight is MGh.  That becomes the energy put into the rotor.  Since the weight hits the floor, you have very sharp "thread tension on" and "thread tension off" "signals" that torque the rotor up to speed and put energy into it.

I think that's a real damn good moment of inertia test that most pulse motor makers should be able to do.

MileHigh

I like it very much.  If I do the test can someone help with the calculations ? I think I've got pulleys and everything I need, also then it can be compared to the other method.

Cheers

[MileHigh]

Sure and I will do a quick rundown on the calculations.

You know from the previous postings that MGh = Iw²/2

Working it.....  2MGh = Iw²

Then.....   I = 2MGh/w²

You can see from the formula that increasing the size of the weight M or the drop distance h will be met with an increase in angular-velocity-squared in the denominator.

As a suggestion for all, if you took the average of several measurements you can expect to have a more accurate measurement.  For example, you might want to average out three runs of two different weights for a total of six measurement runs.  I only mention this because it's a practice that you don't see very much on the forums.

MileHigh

P.S.:  TK, thanks for the "w^{2"  I am a font illiterate see!}