On Saturday 13 June 2020 12:30:46 Lester Caine wrote: > On 13/06/2020 15:49, Gene Heskett wrote: > > On Saturday 13 June 2020 09:29:47 Lester Caine wrote: > >> It seems I have a set of GYS701DC2 and mating servo drivers sitting > >> in a big metal box with a laser sitting in the top of it. I think > >> it's due for stripping down so I can salvage all the usable bits, > >> but I've hit a bit of a brick wall I can't find any data on the > >> motors such as what encoder they are fitted with. Anybody got an > >> old manual as google is proving lacking. > >> > >> While trawling around I cam across the JMC iHSV57-30-18-36 which > >> looks like a possible contender if I was starting from scratch. > >> However once again it does not list the resolution of the encoder, > >> playing instead on it's 'stepper motor' like setting. I've > >> established that the built in driver can handle any encoder from 1 > >> to 2500 lines and it even lists 1000, 1250 and 2500 as the > >> 'standard' encoder resolutions, but not how to recognise just which > >> one is fitted :( In my book, the 2500 line encoder gives a native > >> 10000 step resolution without any 'micro-stepping' and when > >> combined with a 3mm pitch ball-screw ( on the Taig mill ) looks > >> like a magic combination. > >> > >> Anybody using the iHSV57-30-18-36 and worked out what the native > >> resolution is? ... any feedback as to if it's even worth > >> concidering? > > > > Thats a problem I will face for the second time in a bit as I've > > bought a pair of the 3NM stepperservo's myself. > > > > These are equipt with 1000 line encoders, I think that are actually > > 250 line encoders. So for the lathe, I will have to plug in the > > sprocket ratios and z screw stuff to get a usable z scale. > > > > But I faced a similar problem with my g0704 when I put an encoder on > > the top of the spindle motor because I had no clue what the gear > > ratio in the head of the g0704 had it either gear. I intended to > > get, and have but its failed at the moment, an index pulse from the > > spindle rotation. > > > > So I cobbled up in the hal file, a counter which gave me the number > > of pulses from the 1000 line on the motor, for 100 turns of the > > spindle, measuring for both high and low gears, so I now have a > > couple switches on the rim of the gearshift knob to tally which gear > > it is fully seated into which in turn control a couple muxes so the > > tach and such stay calibrated. By using a mux4 for the speed control > > I can also insert a very low speed commend when its not quite seated > > in either gear. So I can reach up and shift gears when its running > > and its turning very slowly, the next gear flat faces mesh > > perfectly, and the selected speed resumes when one of the tally's > > goes good. > > > > That counter is still in the hal file, but commented out. and I just > > put a copy in my web page for you, or anyone else, you can get it > > all from my site in the sig, click on that and when you can see me & > > the missus, add lathe-stf/ to the browsers address bar and ENTER. > > You should then see another subdir named GO704-5i25-7i76/, click on > > that and help yourself, that is a copy on the config dir for my > > GO704. The main GO704-5i25-7i76.hal contains this counter. Copy it, > > fix the missing addf's etc. I used halshow's command line to reset > > it for each run, and a halmeter to read what it said. When reset, it > > counts indexes from the third to the 103rd, for any axis you feed it > > from that has an encoder, so divide what the halmeter says when it > > stops counting to get the actual number of pulses for one turn, > > divide by 100. > > > > I hope this is helpfull, Lester > > I get exactly where you are coming from, but I think I've got a > slightly different take on things. I have found reference to the > iHSV57-30-18-36 having a default of 4000 and that being '1:1' so this > suggests the motor has a 1000 line encoder ( 4000PPR ) but I would > prefer even just the 1250 line encoder to restore some of the > resolution lost moving from the 20TPI thread on the original leadscrew > over the 3mm pitch of the ball screw ... I suspect now that the 2500 > line encoder may actually be too high a frequency for the built in > electronics which is a shame and perhaps that is where we do need to > use LinuxCNC to close the loop using faster hardware than that built > into the integrated servos? > > I don't see the point of the 'serv ostepper' for anything other than > the raw steps of the motor itself? Although I suppose micro-stepping > may be more accurate if it's following the steps of the encoder? > I'm going to find out.
With regard to the frequency, I've been known to trace out and remove the glacial speed opto's in several so-called breakout boards, by-passing that speed limit if its a mesa or lookalike encoder on the receiving end and I'm using one such board in the GO704 right now, connected to P2 of a 5i25. Works well when I can get an index from a separate src into it, which is whats failed ATM. But first I have to go visit the woof, whose in the shop with end stage COPD. So the mill is the 2nd or 3rd on my list of toodoo's today. > > Stay well and safe. > > > > Cheers, Gene Heskett Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) If we desire respect for the law, we must first make the law respectable. - Louis D. Brandeis Genes Web page <http://geneslinuxbox.net:6309/gene> _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
