Dave Houghton wrote: > Heinz Reimer wrote> > >> >> > I've posted previously on this topic, including the link to the guy who > thought that step time and step space for the Sherline option in the > wizard were inverted. Since then, I've done some additional testing, and > no longer believe this is correct, as it produces missed steps in my > setup. (Sherline PN 8760 driver box, one of the early ones, purchased in > 2003) > > In an attempt to understand this better, I decided to try the "reverse > engineering" approach, and measured the actual step pulses generated by > the Sherline supported EMC-BDI distribution, using the settings from the > factory Sherline ini file (max velocity .36; max acceleration 5; scale > 16000). http://imagebin.org/31335 The upper left image is jogging at max > velocity, the upper right is a direction change, also at max velocity. > The measured values are: Step time and step space, both about 80 > microseconds, direction setup, 40 microseconds, direction hold 2.7 > milliseconds. These are an order of magnitude longer than the other > common drivers! Also of note, the direction change occurs when the step > pulse is high, rather than low, as with EMC2. I don't know if this has > any relevance, and hope someone more knowledgeable will comment. > > I then looked at the steps generated by the sample stepper XYZA > configuration, using identical ini file settings. This is what I use and > have found to be reliable over several years of use. (middle images) > Step time at max velocity is just under 50 microseconds, step space is > about 90 microseconds. > > Next I returned to the Stepconfiguration Wizard and tested settings > similar to the ones I had measured. Starting with step time and step > space of 80 microseconds, I worked my way down and found that between 44 > and 45 microseconds for either parameter was where failure occurred. I > then did the same for direction setup and hold, (with step time and > space at 50 microseconds; (bottom 2 images)) and was unable to > demonstrate errors all the way down to 200 nanoseconds for both. As an > aside, the measured step time here is around 30 microseconds, despite > the minimum setting of 50. > ...[snip] > > Hello Heinz > > I also transposed the Step Time and Step Space values, now my Step > Time=6000; Step Space=1000, and I got considerable missed steps at max jog > velocity of 450mm/min. Seemed OK at 100mm/min. > My loaded values for Vel=7.5mm/sec and Acc=15mm/sec^2. Now 7.5mm approx = > 0.3" so we are similar. However my acceleration of 15mm/sec^2 approx = 0.6" > is vastly different to yours. > Also my Scale=800 now I must say here I have no idea what this Scale means > or how it is used. My main concern is the vast difference in acceleration > values. Could you comment please. > Can you please give me some values to put in Step Space, and Step Time or > how does 8000 sound for both? > One last thing Jeff Epler commented on "really long" step pulses - roughly > what values are we talking about here? > Many thanks and > Best regards > Dave > > > > ------------------------------------------------------------------------- > This SF.Net email is sponsored by the Moblin Your Move Developer's challenge > Build the coolest Linux based applications with Moblin SDK & win great prizes > Grand prize is a trip for two to an Open Source event anywhere in the world > http://moblin-contest.org/redirect.php?banner_id=100&url=/ > _______________________________________________ > Emc-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-users > > Hi Dave
Scale refers to the number of pulses required to move one machine unit. I'm assuming you have a metric machine and a Sherline driver. So with 4 microsteps/step, 200 steps/revolution, and a leadscrew pitch of 1mm, you need 800 pulses to move 1mm. I have an inch machine, hence I need 16000 pulses to move one inch. Sherline's "factory" acceleration settings don't make a whole lot of sense to me. In the metric version the default values are 10 for Y and 15 for X and Z. For the inch version they are 1 for X, and 5 for Y and Z. I use 2 inches/sec^2, as higher values tend to cause following errors. The time units in Stepconfig Wizard are nanoseconds. The shortest pulses I measured using the Sherline supported EMC-BDI at their default max velocity were 80,000 ns. That's considered "really long". As mentioned above, if you use values of at least 45,000 ns for both step time and step space, it seems to work, but as Jeff mentioned, the Wizard stops listening with values that large and gives you less. If you do go this route, I would suggest you use values a bit above that minimum value and then test carefully for lost steps. The sample stepper configuration uses much longer pulses to begin with, so seems more suited to the Sherline driver. To use this option the general approach is as follows: Open EMC2 and in the configuration selector chose the appropriate sample configuration. For you, assuming you're using a 3 axis metric mill, this is probably stepper_mm. The configuration will be copied to your home directory. From your home directory, navigate to emc2/configs/stepper, and open stepper_mm.ini with your text editor. Using the factory Sherline ini file http://www.sherline.com/emc/ini.html as a template, edit the ini file to match. The lines you'll need to edit are: In the Trajectory Planner, default and max velocity and acceleration In the Axes section, for each axis, max_velocity, max_acceleration, stepgen_maxaccel, scale, ferror, and min_ferror. Assuming you don't have any additional functions such as limit or home switches, spindle control, or estop, that should be it. Save your newly edited ini file and reopen EMC2. Choose the stepper_mm configuration which is now listed under your home directory and test. Regards Heinz ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
