On Wed, 2 Feb 2011 21:08:09 -0800 (PST) "Peter C. Wallace" <[email protected]> wrote:
> It was pretty surprising. I was using Matt Shavers velocity mode test > hal file with halrun, so the effect was not masked by a FE as it > would be under EMC. If we had any integral gain and put enough load > on the motor that it could not match the requested velocity, and > therefore saturated the PID output, as soon as we removed the load, > instead of a velocty overshoot and recovery, the PID loop locked up > with a saturated output resulting in a runaway. We had some motor runaway during a test this morning. I wasn't physically present when it happened, but the fellow doing the testing said that as soon as the "8i20.enable" box was checked in the Pyvcp GUI, the PID output would latch at 1.0. Interestingly, I believe pid.0.enable was set to false (which should reset the integrator), but upon re-enabling, the output once again latched at 1.0. I don't know what the pid output was when pid.0.enable was false. A restart of halrun allowed the testing to continue with no changes in the config files. I looked at pid.c, but I don't know enough to mess around in there :) Test files are at: http://www.mattshaver.com/8i20/ Thanks, Matt ------------------------------------------------------------------------------ Special Offer-- Download ArcSight Logger for FREE (a $49 USD value)! Finally, a world-class log management solution at an even better price-free! Download using promo code Free_Logger_4_Dev2Dev. Offer expires February 28th, so secure your free ArcSight Logger TODAY! http://p.sf.net/sfu/arcsight-sfd2d _______________________________________________ Emc-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-developers
