Saturday, March 31, 2012

Pegasus 1/48 Martian War Machine--WIP Part 3

Previously on Martian War Machine WIP...

Not a lot of pretty pictures to show this time. I completely re-leafed the upper hull, since I was unhappy with the mottling that occurred due to my lack of control of the patina process. I did discover that there is something really simple that helps change copper from bright, almost white, to a duller reddish tone. Finger oil. Just rubbing your finger over the copper ages it a bit. The effect is subtle, but shouldn't be totally unexpected: after all, what happens to pennies? Aside from exposure to earth's atmosphere, all that happens to them is that people touch them. Most pennies don't get dipped in household cleaners or exposed to any other chemicals that aren't normally found on your bare hand. If experimentation doesn't reveal anything better for giving the copper a somewhat aged appearance, I will have to settle for this.

The heat ray projector swivels back and forth on the "real" thing and on the model as well. I'm taking advantage of this feature to motorize the MWM.

First, I chose the motor and reduction gears. The motor is a cheapo, a little can motor I picked up at Edmund Scientific years ago for 25 cents. It runs at something like 500 RPM 5000 RPM when 1.5 V is applied with no torque. I superglued some aluminum tubing onto the shaft to match the bore through the worm that will be used. This will turn a 50 tooth gear with a coaxial 10 tooth pinion, which in turn drives another of the same, which in turn drives a 50 tooth gear which is mounted to the hollow shaft of the base of the heat ray projector. I was originally going to put in a linkage from the shaft at the base of the heat ray to the third gear, like the push rod on a steam locomotive, to mechanically cause the heat ray to travel back and forth. This idea limited the travel of the heat ray to be less than 45 degrees from center.

A better idea came to me while I was nearly asleep (see circuit diagram at left). A momentary-off pushbutton can provide the feedback to prevent overtravel of the heat ray projector (which ought to be avoided because it is undesirable to twist the wires leading to the heat ray LED and because it would look silly for the heat ray to spin all the way around). The pushbutton will interrupt the current to the motor to prevent overtravel. A center-off DPDT switch will supply current to drive the motor in the desired direction, which is interrupted by the momentary pushbuttons. A simple diagram of the gearing is included. It also shows the stacked gear ratios needed to slow the motion of the heat ray projector down to 1/125 1/1250 of the RPM of the motor.

The plastic gears have, as near as I was able to measure, a bore diameter of .111", so in order to obtain appropriate diameter shafts, I ordered lengths of number 33 and 34 drill rod, at .112 and .110 diameter respectively, from McMaster. Good thing I ordered both to bracket the needed diameter: both were undersize (apparently McMaster got my order mixed up and sent 35 in place of 33), the larger drill rod measuring .110". I cut it to length and also drilled holes in a small piece of 1/4" ABS, spaced at about .592" to support the shafts holding the gears at the right distance from one another to prevent binding or slippage. Drilling precisely spaced holes is not something I am set up to do at home, and I got one good result out of three tries. I had wanted the gear shafts to be supported at both ends, but decided to settle. One of the leads on the motor needed to be resoldered, as well.

Next time: the motor and gear box all installed and working, and a video demonstration of the motion of the heat ray projector.

No comments:

Post a Comment