Hi Everyone, Is this mail getting through?? I have an X2 mill with a spindle controller which converts pulses to voltage output. I'm planning to use pwmgen in PDM mode to generate the pulses. Is there a way to ramp up the pulse density (frequency) to allow a soft spindle start?
GCODE is something like this, but this seems a clumsy way to do it S500 G04 P4 100 S750 G04 P4 100 S1000 G04 P4 100 . . . Regards Andy -----Original Message----- From: Eric H. Johnson [mailto:[email protected]] Sent: 24 September 2009 21:34 To: 'Enhanced Machine Controller (EMC)' Subject: Re: [Emc-users] some interesting results Andy, >> Well, I finally got round to trying this, and it seems that it was well worth it. My base period jitter has gone from 20,000 to 2,509 which seems a quite unfeasible improvement. << Cool. Actually that seems about right. Running on a 1.6Mhz Atom smp I get just over 2000 as I recall. >> The next thing to do is to make use of this improvement: Stepconf always seems to use 100,000 base thread and 1,000,000 servo thread regardless of the input parameters used for drive timings and latency. Is this normal?<< Yes, that is pretty typical. >> Where are the step, space and direction-hold timings defined? I couldn't see anything explicit in the INI files, but perhaps I was not looking carefully enough. (Or are these always exactly one thread period). << In the HAL file. What driver are you using, the parport? If so, then you will want to use for example: stepgen.n.steplen stepgen.n.stepspace etc. If using another driver, they will have their equivalents to these parameters. It can also be passed through from the ini file, as in this example from the Mesa hostmot2 drivers: # timing parameters setp hm2_[HOSTMOT2](BOARD).0.stepgen.00.dirsetup [AXIS_0]DIRSETUP setp hm2_[HOSTMOT2](BOARD).0.stepgen.00.dirhold [AXIS_0]DIRHOLD setp hm2_[HOSTMOT2](BOARD).0.stepgen.00.steplen [AXIS_0]STEPLEN setp hm2_[HOSTMOT2](BOARD).0.stepgen.00.stepspace [AXIS_0]STEPSPACE setp hm2_[HOSTMOT2](BOARD).0.stepgen.00.position-scale [AXIS_0]SCALE setp hm2_[HOSTMOT2](BOARD).0.stepgen.00.maxvel 0 setp hm2_[HOSTMOT2](BOARD).0.stepgen.00.maxaccel 0 setp hm2_[HOSTMOT2](BOARD).0.stepgen.00.step_type 0 The values specified as "[AXIS_0]<parameter>" would then be specified in the AXIS_0 section of the ini file. >> My main interest is in running a faster thread to read the spindle encoder. Currently I have the pulses connected through a HAL software filter, but I am running into problems with the granularity (for want of a better word) of the thread that the filter runs in. If I sample over 2 periods then I start losing encoder pulses to the filter at the moment. (You might note from another thread that I am investigating hardware filtering too)<< I am afraid I do not follow what you are asking, and confess I was not following the other thread. Are you dropping counts on the encoder or do you not have a meaningful time period over which you are sampling encoder counts to get a stable velocity? Are you using the parallel port to read encoder counts? >> Any suggestions for values for BASE_THREAD and SERVO _THREAD with a 2509 latency?<< If you are using the parallel port, then the base thread needs to be faster than the fastest frequency the spindle encoder can generate, assuming there is no other device which needs to be read even faster. Regards, Eric ------------------------------------------------------------------------ ------ Come build with us! The BlackBerry® Developer Conference in SF, CA is the only developer event you need to attend this year. Jumpstart your developing skills, take BlackBerry mobile applications to market and stay ahead of the curve. Join us from November 9-12, 2009. Register now! http://p.sf.net/sfu/devconf _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users ------------------------------------------------------------------------------ Come build with us! The BlackBerry® Developer Conference in SF, CA is the only developer event you need to attend this year. Jumpstart your developing skills, take BlackBerry mobile applications to market and stay ahead of the curve. Join us from November 9-12, 2009. Register now! http://p.sf.net/sfu/devconf _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
