Gentlemen, I mailed my answer to Przemek unaware that the reply address was only his private. So I mail it again to the list hoping for some help. Peter
Hello Przemek, thanks for the hint with the scope. That's what I suspected to be due all along. I did just this, bought an old oscilloscope (10 Euros, Tek 422, 40 years old, but very willing) to look at the pulses. I cannibalized a former RS232 null modem with sub D 25 plugs on both ends and equipped it with 10 LED's and half an IC socket between those plugs to make the data lines accessible. I connected this with the PC, stepper card and scope properly. Then I booted the PC with my old DOS disk and ran the program I was used to. I saw clean, regular pulses with a constant length of 80 usec and a period of 200 usec at minimum (5 kHz is the limit of the card) while the motor was beautifully turning, singing up and down with the ac- and deceleration rate entered in the program params. I left everything running and unchanged, booted the Ubuntu disk instead and started stepconf. I got pulses of a constant length of 15 usec all right, but with a varying (!) distance starting with the value adjusted on page 1 of the stepconf program. But: the distance between the pulses varied regularly twice a second by 70 to 100 usec regardless of the pulse frequency, hard for my heritage scope to get a steady picture. The motor did not sing as before with DOS, but rattled somehow with a very indecent sound, but turned. At higher speed the 100 usec wobble has greater influence than a low speed. So, the higher the speed, the more uncomfortable it sounds and the worse performance gets. The motor has only little momentum and apparently loses steps. I can tell right from the sound that something is wrong. No wonder, since every step have different lengths. I repeated this with many different pulse and pause values and latency settings. I commented the "..reset.." lines in the .hal file to turn off doublestep. I tried inverting the pulses. All this didn't bring any improvement on the large scale. Questions: - where does the 100 usec wobble twice a second come from, an how can I get rid of this? - why can't I influence the pulse lenght by varying parameters in stepconf? Thanks for any contribution... Peter Blodow At 03:57 17.03.2009, you wrote: >On Mon, Mar 16, 2009 at 3:08 PM, Peter blodow <[email protected]> wrote > > The driver card can't be the reason because with my old DOS program things > > work still ok. > >I would look at the waveforms for the steppers as generated by the DOS >program. >and try to find out how are they different from the waveforms generated by >EMC. >You'd need an oscilloscope for that (at least for the DOS program, >halscope can >display EMC waveforms). ------------------------------------------------------------------------------ Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are powering Web 2.0 with engaging, cross-platform capabilities. Quickly and easily build your RIAs with Flex Builder, the Eclipse(TM)based development software that enables intelligent coding and step-through debugging. Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
