Heh, I have been getting practice with the soldering iron removing parts BUT that is on a board with no intention of it ever working again. I am just harvesting plugs to use. I would be very afraid to try this on a board with any life left. :)
On Thu, Mar 28, 2013 at 11:19 AM, Chris Radek <ch...@timeguy.com> wrote: > On Thu, Mar 28, 2013 at 10:51:12AM -0500, Stuart Stevenson wrote: > > > I am trying to dumb down the drive so the LinuxCNC tuning is directly > > effecting the motor motion instead of manipulating the tuned servo > system. > > My VMC uses velocity mode amps with tachs, and I had one tach that > would sometimes give a "bumpy" output that would cause a banging > once per screw revolution. > > It is a high accel machine with high res encoders. > > I had extra DACs so I fixed it by unhooking the real tach and > instead sending the hostmot2 high quality encoder velocity feedback > out a DAC and hooking that to the tach input on the drive. I had to > adjust the gains on the drive to compensate for the much lower > "tach" input (now +-10). I don't remember whether I replaced > components, or just turned knobs. > > For those interested in the results (these were apparently taken at > a time when the real tach was working right): > > http://timeguy.com/cradek-files/emc/real-vs-generated-tach.jpg > http://timeguy.com/cradek-files/emc/real-vs-generated-tach-upclose.jpg > > But this is all background information for what I propose. The > first thing the amp does, as has been pointed out, is take the > difference of the command and feedback signals. I suspect you could > do exactly the same in HAL and use just one DAC to feed an amp like > this. One of the inputs would be just zeroed (shorted). The signal > to the amp would now be velocity error, not velocity command. > > The amps on my machine also have at least one safety feature that > uses the tach input: it has a max speed setting and just compares > the tach input to a setpoint and faults if it exceeds. I think this > is a basic runaway detection. I think this still works in my setup > now but I didn't test it. I say this only to remind people to watch > for similar gotchas if you try to "trick" the amp. Shorting the > tach input would disable this safety feature, but perhaps shorting > the command input and hooking my error signal to the tach wouldn't!? > > This might help you scratch your itch of simplifying the parts that > are outside linuxcnc/hal in the hardware realm, but it doesn't > eliminate dependence on tuning in the amp. I have had such good > success with velocity mode amps (sometimes after tuning them with a > soldering iron) that I don't share your goal of making them dumb, > but if you succeed at that (likely the only way is also by using a > soldering iron), I would love to hear your results. > > Chris > > > ------------------------------------------------------------------------------ > Own the Future-Intel® Level Up Game Demo Contest 2013 > Rise to greatness in Intel's independent game demo contest. > Compete for recognition, cash, and the chance to get your game > on Steam. $5K grand prize plus 10 genre and skill prizes. > Submit your demo by 6/6/13. http://p.sf.net/sfu/intel_levelupd2d > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users > -- dos centavos ------------------------------------------------------------------------------ Own the Future-Intel® Level Up Game Demo Contest 2013 Rise to greatness in Intel's independent game demo contest. Compete for recognition, cash, and the chance to get your game on Steam. $5K grand prize plus 10 genre and skill prizes. Submit your demo by 6/6/13. http://p.sf.net/sfu/intel_levelupd2d _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
