In lugnet.general, Beat Felber wrote:
> Thank you very much for that fine piece of software Bob!
>
> After reading your post I immediately run in my LEGO room to grab an RCX and
> the IR tower to test your software. Well, I had to find out that I had not
> enough batteries at home and thus had to get some overpriced ones at the next
> gas station :-)
>
> Download of the firmware as well as the program worked fine. The program runs
> without problems and the Bulldozer reacts immediately. Great!
>
> I have a few issues though where I'd like to ask you for some hints:
>
> - The rotation sensor sets zero at the position it was during start of the
> program. By stepwise moving the axle in one direction, the PF motor runs
> first at max. speed and then gets slower and slower. So PWM works fine. But
> how can I change the program in a way that from zero position the speed
> increases when turning the axle of the rotation sensor? Currently it
> decreases stepwise from max. speed which makes it somewhat difficult to
> control the Bulldozer (or whatever model using the PF motors).
>
> - Furthermore I found out that turning the rotation sensor in one direction,
> the motor runs in for example forward direction. I thought by turning the
> connector plate on the sensor interface of the RCX would change direction
> of revolution of the motor when the sensor is turned in the same direction
> as before. Of course, I tried all four possible orientations of the
> connector, but the motor still turns always in the same direction when
> turning the axle of the rotation sensor forward? Is this the correct
> behaviour?
>
> - Finally, does it happen to you too, that the rotation sensors loose
> calibration very fast? Just after playing around for a few cycles (forward,
> backwards, forward...) The zero position is not at the same place any more.
>
> If you could give me some hints on how to overcome these issues, this would
> be much appreciated. You will have to know that I am everything else than a
> Mindstorms professional :-)
>
> Thanks again for your work
>
> Beat
>
> PS. My intention is to build a proper remote control with two mechanical
> joysticks, each coupled to two rotation sensors to be able to control all
> four functions of the Bulldozer at the same time.
Well this was just a quickly hacked out example program to show how to use the
library function that i wrote. i realized at the time that the PWM started out
at fastest and went to slowest but i had run out of steam and never really got
back into it to fix it. i'll probably spend some more time this week and make
the changes you suggested.
i will have to play around to see if i notice the calibration issues you
mentioned. one suggestion is to build a ratcheting mechanism like in the 8288
using the bionicle teeth and the rubber part that has 2 axle holes.
bob