On Sat, 2011-02-12 at 13:36 -0800, Doug wrote: > What do the letters stand for in the hal. It seems anything I increase make > it run rough. P is works great a small value of 75 to 125. any increase > other that 0 on I and D makes it rough again. What values can I put in > here. You said before you think these old velocity drive only take a small > p value if I remember correct but could be mistaking. I read the link > before and follow the instruction the best I can but I think it a bit over > my comprehension. > > Arch Fitters > Doug McCurtain (C.Ped) > President > P = gain or stiffness I = integral ... shoves the axis toward equalibrium D = differential .. pushes harder as the following error gets larger FF1 = velocity feedforward FF2 = accel feedforward
Speaking of geography I'm close (relatively) only 200 mi. But you have a very experienced emc'er in Portland. Dan Falck [email protected] [email protected] Tuning is both a science and an art. It takes time to do correctly. HTH Dave > -----Original Message----- > From: Jon Elson [mailto:[email protected]] > Sent: Saturday, February 12, 2011 12:36 PM > To: Enhanced Machine Controller (EMC) > Subject: Re: [Emc-users] Servo Balance > > Doug wrote: > > Hello, > > > > > > > > I am new to this site so forgive me if I do not understand everything. I > > have an old Knee vertical cnc Lagun mill that I install emc 2.4.6 with > > John's Pico Systems digital to analog. I have old west amp servo drives > > which I believe are called velocity drives due to they speed up with dc > > voltage increase. The system, after Johns great help is working great > > except for the it does not seem to be balanced. With Pico system shut off > I > > have the driver cards balanced to zero and motors hold with no movement. > > When enabled with emc there is a .008 motion to the right and then back > > again. > Because of component variations, the "zero" output of the DACs is not > exactly zero volts. > It generally is within +/- 10 mV of zero, but you can check it. You > will never be able to get the > zero to be perfectly nulled out and stay that way. If it warms up or > cools down in your shop, the > drift will come back. > > I have used hal scope to tune the servo loop to .005 error. Can > > run the system 200 inches per minute with out error. But the back and > forth > > movement while that axis is not moving cause a wave pattern. It does the > > same amount on x and Y. I am not sure what the hal values stand for to > stop > > this. Please advise and please be specific since I am new to this. > > > OK, are you saying that this movement continues, or is just when you > first hit F2 and go to machine on? > > If the movement continues, then the servo PID parameters in EMC are not > set right. First, reduce > P and try different values of D to see if you can get rid of the > oscillation. The best way to do this > is to use the jog keys to bump the axis back and forth, and look for > signs of instability at the > transition points. > > See http://wiki.linuxcnc.org/cgi-bin/emcinfo.pl?PWM_Servo_Amplifiers > for some info on servo tuning. > > Jon > > ---------------------------------------------------------------------------- > -- > The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE: > Pinpoint memory and threading errors before they happen. > Find and fix more than 250 security defects in the development cycle. > Locate bottlenecks in serial and parallel code that limit performance. > http://p.sf.net/sfu/intel-dev2devfeb > _______________________________________________ > Emc-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-users > > > __________ Information from ESET NOD32 Antivirus, version of virus signature > database 5868 (20110212) __________ > > The message was checked by ESET NOD32 Antivirus. > > http://www.eset.com > > > > ------------------------------------------------------------------------------ > The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE: > Pinpoint memory and threading errors before they happen. > Find and fix more than 250 security defects in the development cycle. > Locate bottlenecks in serial and parallel code that limit performance. > http://p.sf.net/sfu/intel-dev2devfeb > _______________________________________________ > Emc-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-users ------------------------------------------------------------------------------ The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE: Pinpoint memory and threading errors before they happen. Find and fix more than 250 security defects in the development cycle. Locate bottlenecks in serial and parallel code that limit performance. http://p.sf.net/sfu/intel-dev2devfeb _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
