Programming technique..

[archiver]

Posted by bill whitley at 05/02/2001 9:13 PM EST


Student on team #70, Auto City Bandits, from Powers Catholic High School and Kettering University.



After reading that the Victors give no output for an input of 127 +/- 10%, leads me to question my programming technique. I had programmed in my own dead zone (117

[archiver]

Posted by Gui Cavalcanti at 05/02/2001 11:15 PM EST


Student on team #422, Mech Tech, from Governor's School (GSGIS) and Verizon.


In Reply to: Programming technique..
Posted by bill whitley on 05/02/2001 9:13 PM EST:



For the code space it takes up (and from the programmers' point of view) no, it's definitely not worth it. To the drivers point of view, however, small bumps to the calibration knobs usually means fighting the controls if there's no dead zone. If you have a one-joystick machine, this is usually pretty deadly. If you have tank steering (like us) then the drivers can make up for it pretty easily.

Essentially (this is a bad ending) it's all up to you

-Gui Cavalcanti
Team 422, Mech Techs

[archiver]

Posted by Gregory Ross at 05/03/2001 12:46 AM EST


Engineer on team #330, Beach Bots, from Hope Chapel Academy and NASA/JPL.


In Reply to: Re: Programming technique..
Posted by Gui Cavalcanti on 05/02/2001 11:15 PM EST:



: For the code space it takes up (and from the programmers' point of view) no, it's definitely not worth it. To the drivers point of view, however, small bumps to the calibration knobs usually means fighting the controls if there's no dead zone.

(Actually, Gui, Bill's point is that since the Victors already have a dead zone maybe it doesn't make sense to implement a dead band of your own.)

I was very surprised to hear that the Victors have a dead zone built in. Still, I think it is probably a good thing to code your own. Of course, it wouldn't do any good to make your's smaller than the Victor's, but if there's a chance you might want a larger dead zone (to suit your driver's preference), it's not that expensive -- either in coding effort or in program space. (Remember you have lots of room with the BS2SX.)

[archiver]

Posted by Lloyd Burns at 05/03/2001 9:42 AM EST


Engineer on team #188, Woburn Robotics, from Woburn Collegiate and TorDistSchoolBrd, ScotiaBank. OntPwrGen, Enbridge.


In Reply to: Re: Dead Zone Technique
Posted by Gregory Ross on 05/03/2001 12:46 AM EST:




: I was very surprised to hear that the Victors have a dead zone built in. Still, I think it is probably a good thing to code your own. Of course, it wouldn't do any good to make your's smaller than the Victor's, but if there's a chance you might want a larger dead zone (to suit your driver's preference), it's not that expensive -- either in coding effort or in program space. (Remember you have lots of room with the BS2SX.)

. In fact, while the Victors have no output for input values of 127 +/- 10, we would do well to remember that, esprcially under load, many motors would growl and refuse to turn their shafts with outputs from the Victors corresponding to that range.

. In the white papers, here or on the InnoFirst site, there is a paper about realigning your output to a new straight line, to take account of the output and the motor dead-zones.

. For controlling servo mechanisms, I always slow the motor as it approaches the demand position (to prevent overshoot), this year looking up Victor input values based on "distance to go", and the minimum Victor input value to be found is always just outside the dead-zone.