Pulse Power--with a fan!

I just rewired an old MRC HO powerpack into a transistorized throttle. The new low speed control of my locos is very smooth, but I can't avoid
that little jackrabbit jump since the output is very pure DC.
I tried a little experiment. Don't ask me how I thought of it but this actually worked. I wired a small PC power supply fan in series with the pack output. I couldn't believe my eyes. The little fan started to spin first, then my loco began to crawl at that one tie a minute pace I'm so used to in my other packs. The loco didn't get up to a very high speed at full throttle because the fan was using 8 volts leaving only 5 volts for the loco tops. So this method could never bee used for full range.
Can any of the electronic wizards on the group explain how a little PC fan was able to pulse a loco? The fan amp rating was .15 amps.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
robby wrote:

The problem with the loco on pure DC is stiction. The bearings present more friction in the transition from stop to go than they do once moving, so by the time there is enough amperage flowing to move the loco there is enough to cause it to run at a highish speed. At zero rpm the motor has a low resistance (say 12 ohms) but once it is spinning the effective resistance rises. (to perhaps 36 ohms on 12 volts at max speed)
It sounds as thought the fan doesn't have good commutation if it's output is pulse power. Now you need a wirewound rheostat across the fan to adjust the available voltage to the fan and so give your loco the full 0-12 volts. (the original controller rheostat would do nicely)
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I once put a scope across the terminals of a DC motor. The motor turned the straight line of DC to a haze of noise and spikes on the screen.
I'd expect a PC power supply fan to be quieter, but I'd guess it's adding noise to the DC, and producing enough voltage spikes to give your loco motor a bit of a kick to keep it going at low RPM.
--
Bill Kaiser
snipped-for-privacy@mtholyoke.edu
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Those fans (the burshless ones) are electronically controlled and the power is chopped by the electronics. In addition, the regular DC motors that are in some have a bit of stuttering as the brushes for current just like the motor in our loco and that provides some pulsing of the power. The thing to do with your throttle is to look at some of the circuits on the web and most of the ones that use pass transistors in analog mode have a little circuit attached to the control lead to provide pulses of some kind to the control line. Anything from square waves from an oscillator to a little feeding in of the AC input waveform si done depending upon the way that the designer did their thinking on the circuit. Then there are the electronic controllers that use the power amp in a switch fashion and those put out pulses of varying width for the output and they do an excellent job of low voltage crawling of the loco. Unfortunately, the Linn Westcott series of TAT throttle schematics aren't availbe on the web (at least, not the last time that I looked) otherwise you could see the whole evolution of the basic electronic throttle from what you have built to a throttel with braking, constant speed control and so forth. Most of the DC output throttles use some variation of those circuits as their base. Probably the simplest for you would be to put together a 555 circuit as an oscillator (run it around 200Hz) and feed the square wave into the control line with a resistor so that it makes a few volt AC signal on the DC line.
-- Why isn't there an Ozone Hole at the NORTH Pole?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.