Re: calculating bps for sync rsg (confounded and confused)

To: tesla-at-pupman-dot-com
Subject: Re: calculating bps for sync rsg (confounded and confused)
From: Rscopper-at-aol-dot-com (by way of Terry Fritz <twftesla-at-uswest-dot-net>)
Date: Sun, 05 Mar 2000 11:38:45 -0700
Approved: twftesla-at-uswest-dot-net
Delivered-To: fixup-tesla-at-pupman-dot-com-at-fixme

In theory, the answer would be 1800/60*4, however, I have found from 
oscilloscope traces and many hours drawing time lines that this is not always 
the case.  The real frequency is dependant on the time each rotating 
electrode is in proximity to each stationary electrode per revolution.

For example:

Given LE = thickness (diameter) of largest electrode, R = radius of rotating 
electrode to motor shaft center, P = number of gap presentations per 
revolution.

For my RSG LE = 0.25in, R = 2.125in, and P = 16 since I have two sets of 
stationary electrodes (LE) tied together 120 degrees apart and 8 rotating 
electrodes in sets of 2 - 11.25 degrees apart.

My motor is 1800 rpm, so as theory goes 1800/60*16, my frequency should be 
480 hz, but I measure 565 hz on the scope. Why?  Because the thickness of the 
largest electrode effects the frequency.   

If you take one circumference length and lay it out flat as a timing chart 
and divide the chart by the number of presentations P, you get G = 
2*PI*2.125/16 = 0.8345 ( the time one electrode can fire).  Subtract LE/2 
(for my RSG geometry) and divide by G and you get the percentage of time each 
rotating electrode is in front of a stationary. (0.8345-0.125)/0.8345 = 0.85 

480hz divided by 85% = 565hz, exactly what I measured at peak firings!

The smaller the electrodes are, the closer you get to the theoretical value.  
The measured frequency will always be higher than the theoretical.

On the other hand, if I made my stationary electrodes equal to 0.8345, a 
rotating electrode would always be in front of a stationary, and the equation 
would go to 0, when in reality, it would revert to being a static spark gap 
and fire at 120hz.

I've been trying to add these equations to WinTesla, but the various 
geometries of the gap design makes it difficult to come up with one true 
equation.

Hmmmm......something else to think about........-dot-comments?

R. Scott Coppersmith