On Saturday 20 March 2010, Paul Keeton wrote: >So is there a converter that anyone is aware of or a method in emc that >would allow you to take quadrature encoder signals and convert them into an >analog tach signal? I have other Fanuc systems in the shop that use tach >feed back on the motor and Linear Scales on the axis with no issues. Drive >MUST see a tach signal or it will not work. The signal originally (On a >Fanuc) is brought into the CNC master PCB and then routed to the drive from >there. If an encoder is brought back to the CNC then the signal is > converted on the Master PCB and then routed to the drive. Any way it goes > the drive must see an analog tach signal. Fanuc uses a custom chip called > an "LSI" to do this. Are you saying that a separate encoder or scale is > impossible without an encoder on the motor or just more difficult? > Difficult can be worked with. I suppose I could also modify the motor to > use both.....Now that might be a simpler solution to get the best of both. > Drive would get what it needs and emc should like it also. > With all due respect, trying to make an analog voltage level out of an encoder, be it a quadrature or a pulse per increment device, the time lag involved in converting the output pulse train into a smooth analog signal, and making it work, is a good recipe to waste a week and still not get it done to a usable state. That isn't saying its impossible, I can think of one or two ways that might be made to work based on sawtooth wave generators based on miller runup ideas, and doing a sample-hold on the resultant. But that is going to be a pretty complex design that I as a C.E.T., would gladly leave to those with more patience than I. Using the sample and hold, one could fabricate a step function analog signal that was valid till the next edge came out of the encoder. But even that on warming the wet ram would mean it would be low if the motor was being accelerated, and high if it was being decelerated, enough that stability during those phases would be very problematic.
This really does call for a very tightly coupled dc generator, with a finely spaced commutator and lots of slots in the armature. I think it would be the only way to get a degree by degree of rotation speed signal for this. Its also a 1930's way to do it. >----- Original Message ----- >From: "Stephen Wille Padnos" <spad...@sover.net> >To: "Enhanced Machine Controller (EMC)" <emc-users@lists.sourceforge.net> >Sent: Saturday, March 20, 2010 12:10 PM >Subject: Re: [Emc-users] G52 and Fanuc conversion to EMC > >> Paul Keeton wrote: >>>> If there's a tach on the motor, then you should be able to route that >>>> directly to the drive. My understanding is that some controls would >>>> synthesize the tach feedback from encoder feedback. If you have one of >>>> those, then it gets more complex (and I don't know the answer) >>> >>> I would think that the simplest solution would be to change the motors >>> to tach feedback and then install either linear scales on the iron or >>> install >>> encoders on the ballscrews. I would think this would be the simplest way >>> to >>> handle it? >> >> It may be, it may not be. >> >> If there are already encoders on the motors, you should keep those >> installed if at all possible. PID only works well when the feedback >> device is rigidly coupled to the actuator - ie, when the encoder is >> mounted on the motor. If you use only a linear scale for feedback, and >> there is any backlash, then it will be difficult or impossible to tune >> the PID. You can use a scale in addition to an encoder, as was done on >> the G&L at MPM (I think there's a case study on the linuxcnc.org >> website, but I'm not sure where). >> >> Encoders on the screws should be OK if there's no backlash between the >> motor and the screw. It's a good rule of thumb to consider that every >> step the encoder is moved away from the motor makes PID tuning harder. >> >> - Steve >> >> >> ------------------------------------------------------------------------- >>----- Download Intel® Parallel Studio Eval >> Try the new software tools for yourself. Speed compiling, find bugs >> proactively, and fine-tune applications for parallel performance. >> See why Intel Parallel Studio got high marks during beta. >> http://p.sf.net/sfu/intel-sw-dev >> _______________________________________________ >> Emc-users mailing list >> Emc-users@lists.sourceforge.net >> https://lists.sourceforge.net/lists/listinfo/emc-users > >--------------------------------------------------------------------------- >--- Download Intel® Parallel Studio Eval >Try the new software tools for yourself. Speed compiling, find bugs >proactively, and fine-tune applications for parallel performance. >See why Intel Parallel Studio got high marks during beta. >http://p.sf.net/sfu/intel-sw-dev >_______________________________________________ >Emc-users mailing list >Emc-users@lists.sourceforge.net >https://lists.sourceforge.net/lists/listinfo/emc-users > -- Cheers, Gene "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) To err is human, to moo bovine. ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
