Jon Elson wrote: > Mario. wrote: > >>Heisenbug. Surely ;-) I'm not sure if I've nailed this one or not, but there's some new data. I discovered a day before going to the NAMES show that the spindle threading didn't work on my mini-mill, and after some fooling around tried a previous, experimental version of the FPGA configuration that I'd done the driver development with, and it worked. I took that to the show, and experienced problems with the spindle sync, but they were REALLY flaky, and hard to reproduce. Well, after going nearly crazy trying to answer what, if anything, had any influence on whether the spindle sync would work or not, it seemed putting kernel message writes into the code "fixed" the problem in some mysterious way. John Kasunich said it had to be timing, and it appears he was dead on! That experimental FPGA config did show some timing problems on PC <=> FPGA communications. I fixed the typo in the VHDL for the encoder index detection on the current version FPGA, and this new version SEEMS to work. I hate to say anything with certainty, as the bug was so hard to reproduce.
Sorry to put you guys through all this, but the review of the code was a very worthwhile exercise. What WAS there was darn obscure. I'll have to test out John's revisions of the code. I didn't want to add additional variables to the mix until I had a handle on what was going on. Whether this has anything to do with Stuart Stevenson's homing problems (or my occasional ones) isn't real clear yet. That's different hardware that has no timing issues that I'm aware of. But, the results with the minimill and UPC shows that EMC will try to run even when communication with the hardware is suffering some garbling. That's kind of scary. Jon ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ Emc-developers mailing list Emc-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-developers
