Alan Condit wrote: > John Kasunich <[EMAIL PROTECTED]> writes: > >> There must be something different between the generated files and the >> working files. You will have to figure out what it is. Once you know >> what the difference is, then Jeff can figure out if it is a stepconf >> error, or if you have something in your system that is beyond what >> stepconf is designed to do. >> >> Stepconf is NOT an all purpose configuration generator - it is intended >> to cover a very popular class of simple configurations. >> >> One idea just occurred to me... while running the working config, open >> a shell and do a halcmd save (redirect to a file). Then load the bad >> config and do the same. The halcmd save should output most stuff in the >> same order each time, so diff can be used to find the differences. >> >> Good luck, >> >> John Kasunich >> > John, > > here is the diff from the halcmd save > > Alan > > <=candcnc_pinout.hal >> =my-mill.hal > 2,3d1 > < loadrt hal_parport cfg="0x0378" > < loadrt scope_rt > 5c3,5 > < loadrt motmod base_period_nsec=50000 servo_period_nsec=1000000 > traj_period_nsec=10000000 key=111 > --- >> loadrt hal_parport cfg=0x378 >> loadrt probe_parport >> loadrt motmod base_period_nsec=100000 servo_period_nsec=1000000 > traj_period_nsec=1000000 key=111 num_joints=3 > 9,30c9,35
100000 seems low for a base period, but I don't know the context, maybe your machine has very bad latency, or something. > < setp motion-command-handler.tmax 22400 > < setp motion-controller.tmax 103194 > < setp motion.servo.overruns 00000090 > --- >> setp motion-command-handler.tmax 21247 >> setp motion-controller.tmax 92276 >> setp motion.servo.overruns 00000005 I think overruns should be zero (that isn't something you set, it is an output from EMC to tell you how things are running, like lost packets from ifconfig. > 40c45 > < setp parport.0.pin-02-out-reset FALSE > --- >> setp parport.0.pin-02-out-reset TRUE > 44c49 > < setp parport.0.pin-04-out-reset FALSE > --- >> setp parport.0.pin-04-out-reset TRUE > 48c53 > < setp parport.0.pin-06-out-reset FALSE > --- >> setp parport.0.pin-06-out-reset TRUE > 52c57 > < setp parport.0.pin-08-out-reset FALSE > --- >> setp parport.0.pin-08-out-reset TRUE > 61,64c66,69 > < setp parport.0.read.tmax 30190 > < setp parport.0.reset-time 00000000 > < setp parport.0.reset.tmax 0 > < setp parport.0.write.tmax 24844 > --- >> setp parport.0.read.tmax 21300 >> setp parport.0.reset-time 00006165 >> setp parport.0.reset.tmax 16858 >> setp parport.0.write.tmax 24642 One configuration is using doublestep, one isn't. That is a HUGE difference. I'm unfamiliar with doublestep, so I'm not about to start troubleshooting it. > 67,78c72,82 > < setp scope.sample.tmax 0 > < setp stepgen.0.dirhold 0000C128 > < setp stepgen.0.dirsetup 0000C128 > < setp stepgen.0.maxaccel 21 > < setp stepgen.0.maxvel 2.5 > < setp stepgen.0.position-scale -2000 > < setp stepgen.0.steplen 0000C128 > < setp stepgen.0.stepspace 0000C128 > --- >> setp stepgen.0.dirhold 00018595 >> setp stepgen.0.dirsetup 00018595 >> setp stepgen.0.maxaccel 31.5 >> setp stepgen.0.maxvel 0 >> setp stepgen.0.position-scale 2000 >> setp stepgen.0.steplen 00018595 >> setp stepgen.0.stepspace 00000000 Step timing parameters are different - so the step pulses will be different. Regards, John Kasunich ------------------------------------------------------------------------- SF.Net email is sponsored by: Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
