The CUI capacitive encoders are relatively inexpensive, and there are line driver "lump" cables available if needed, but there are two minor cautions that may or may not affect you. One is sensitivity to magnetic fields, and the other is quadrature count "hunting".
You might want to be cautious about mounting the AMT102/103 capacitive encoders on the back of a motor (or anyplace else there is a strong magnetic field). The encoder uses a "wavy" variable capacitor that repeats every 45 deg, and a hall effect device to determine which of the 8 repeats is the origin (for the index pulse). So strong magnetic fields may confuse operation of the index pulse. This is mentioned (at least for stepper motors) in the fine print at the bottom of p.1 of the data sheets: http://www.cui.com/pdffiles/AMT102-V.pdf http://www.cui.com/pdffiles/AMT103-V.pdf Why do they specifically mention steppers? Maybe steppers leak more, or maybe the fields are closer to the back? Maybe this hasn't been a problem with servos because they already have a feedback device? I don't know but thought I'd mention it. At any rate it might be prudent to mount them away from electric motors, regardless of type. If you must mount them on the back of a motor, it should still work fine if you are willing to give up the index pulse (which usually means giving up proper homing). The quadrature count hunting means that you cannot use just the A output (or just the B output) to measure RPM. At certain positions the encoder is allowed to hunt back and forth between two count values, even though no real motion is occurring. So there will appear to be extra counts there, if not observed using both A & B in quadrature style. But the quadrature count is said to remain correct, even though it will sometimes seem "nervous". The "nervous" count may mean that at certain positions, the position display might be jumping back and forth between two values, maybe in a blur, even though that axis/joint is not moving. If it's a stepper system, and you're just using the encoder to confirm position and monitor the following error, then maybe the "hunting" (on the display) won't really bother you. But if it's a servo system, the hunting might add a little extra error or noise or vibration that you might find to be objectionable. Here's an interesting video comparing the same servo loop (in position-hold mode) first using the AMT102 encoder (it seems to be a little more "growly") and then with a traditional optical encoder (it seems to be a little quieter). AMT102 vs R35i.avi (henrols) http://www.youtube.com/watch?v=jIrQLfcP_04 I'm sure that in many applications the AMT102/3 encoders will work just fine, I just wanted to point out a couple of minor problems in case they might affect your situation. John plans to try them on his Bridgeport II as stepper monitors when he gets the time to install them. I'll look forward to seeing that and reporting on the results. And when I get some free time I'll post about these two "features" of the AMT102/3 on the wiki. Thanks, Kim Flying Electron wrote: > Thanks for the info on mounting the encoders. > > I decided to go with the mesa 5i23, those ebay amps, keling motors, and > the digikey encoders. > > I'll let everyone know how it works out when everything gets here. > > Thanks everyone for all the advice, it would have been impossible to > choose a combination without everyone's input. > > sam sokolik wrote: > >>http://www.electronicsam.com/images/KandT/pcbmill/bldcenc2.jpg >> >>The keling servos I got (also the KL23BLS-115) have 2 very shallow >>tapped holes that line up to the mounting holes of the encoder. >>I found some laptop assembly screws that just happened to be the right >>length. Mariss seems to love these encoders over the cheap us-digital >>ones for his step/dir servo drives.... >> >>YMMV >> >>sam >> >> >>On 1/7/2010 03:36 PM, Flying Electron wrote: >> >> >>>Wow! Thank you so much for the link. The amps on ebay looks pretty >>>much exactly what I am looking for. I looked at the datasheet for the >>>issue with the drive inputs being commoned at +5V like you pointed out, >>>but I'm pretty sure I can hack up something to deal with that, like run >>>the signals through a microcontroller to do whatever translation is >>>necessary. I especially like that they are all mounted in a nice box >>>for me already :-) >>> >>>Getting closer. Mesa card + Servo Amps are picked out. Keling motors >>>look good. >>> >>>Do you know if that amt102-v encoder need anything special to mount on >>>the back of a NEMA 23 brushless motor like an adapter plate or anything >>>like that? Or do encoders and motors have standard mounting patterns so >>>that all encoders fit all motors nicely? >>> >>>Yesterday I had no idea what I needed to buy or even what questions to >>>ask, and today everything looks doable. >>> >>> >>> >>>sam sokolik wrote: >>> >>> >>> >>>>there are always deals if you are patient. Like... >>>>*KL23BLS-115*<KL23BLS_115.pdf>*: $52/pcs brushless servos >>>>digikey amt102-v encoder 29.95.<-up to 2048 line - adjustable. >>>>ebay 3 amc amplifiers that take pwm+dir >>>>http://cgi.ebay.com/3-AXIS-A-M-C-PWM-Brushless-SERVO-AMPLIFIER-ASSEMBLY-NEW_W0QQitemZ260516353589QQcmdZViewItemQQptZLH_DefaultDomain_0?hash=item3ca7fc1635 >>>>179.95 >>>> >>>>Not including shipping for 3 axis... (not including mesa hardware) >>>>$425.80 >>>>closed loop - priceless... ;) >>>> >>>>You would want to do your homework though - like the drive inputs seem >>>>to be common-ed to +5v. (probably not an issue) >>>>(if I did the math right) >>>>sam >>>> >>>> >>>>* >>>>On 1/7/2010 01:29 PM, Flying Electron wrote: >>>> >>>> >>>> >>>> >>>>>I'm deciding on what servo drive to use with the mesa FPGA card and it >>>>>doesn't look too bad since there are not too many low cost options. I'm >>>>>constrained to a NEMA 23 size motor since a NEMA 23 stepper is on the >>>>>machine right now and I want it to be a direct swap. So far I've found >>>>>from searching on google and from people's suggestions here on the >>>>>mailing list: >>>>> >>>>>Drives >>>>>------- >>>>>[brushless] Pico Brushless PWM Servo Amp $150 >>>>>[brushed] Pico PWM Servo Amplifier $125 >>>>> >>>>>Motors >>>>>-------- >>>>>[brushless] Pico Size 23 Brushless Servo (61ozin continous, 180ozin >>>>>peak, 1000 CPR) $120 >>>>>[brushless] Keling KL23BLS-115 Brushless Servo (61ozin continous, 180 >>>>>ozin peak, 1000 CPR) $134 >>>>>[brushed] Keling KL23-130-60 Brushed Servo (50ozin continous, 350ozin >>>>>peak, 1000 CPR) $162 >>>>> >>>>>Not Suitable >>>>>------------- >>>>>dmm-tech.com -- RS232 based communication not PWM >>>>>gecko -- EMC2 cannot control PID loop (step/dir input, not >>>>>pwm) >>>>>viper -- EMC2 cannot control PID loop (step/dir input, >>>>>not pwm) >>>>>rutex -- EMC2 cannot control PID loop (step/dir input, >>>>>not pwm) >>>>> >>>>>Most of the servo drives I have found have the PID loop built into >>>>>them. Is this the normal case? >>>>> >>>>>Is it possible to run a brushed DC servo with just a simple H-Bridge >>>>>controlled directly by the PWM output and a direction signal? I was >>>>>looking at the Freescale MC33886 which I think could make a very simple >>>>>and cheap H-Bridge. >>>>>http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=MC33886VW-ND >>>>><http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=MC33886VW-ND> >>>>> >>>>>Anybody know of any companies or products I may have missed in my search? >>>>> >>>>> >>>>> >>>>> >>>>>------------------------------------------------------------------------------ >>>>>This SF.Net email is sponsored by the Verizon Developer Community >>>>>Take advantage of Verizon's best-in-class app development support >>>>>A streamlined, 14 day to market process makes app distribution fast and >>>>>easy >>>>>Join now and get one step closer to millions of Verizon customers >>>>>http://p.sf.net/sfu/verizon-dev2dev >>>>>_______________________________________________ >>>>>Emc-users mailing list >>>>>Emc-users@lists.sourceforge.net >>>>>https://lists.sourceforge.net/lists/listinfo/emc-users >>>>> >>>>> >>>>> >>>>> >>>>> >>>> >>>>------------------------------------------------------------------------------ >>>>This SF.Net email is sponsored by the Verizon Developer Community >>>>Take advantage of Verizon's best-in-class app development support >>>>A streamlined, 14 day to market process makes app distribution fast and easy >>>>Join now and get one step closer to millions of Verizon customers >>>>http://p.sf.net/sfu/verizon-dev2dev >>>>_______________________________________________ >>>>Emc-users mailing list >>>>Emc-users@lists.sourceforge.net >>>>https://lists.sourceforge.net/lists/listinfo/emc-users >>>> >>>> >>>> >>> >>>------------------------------------------------------------------------------ >>>This SF.Net email is sponsored by the Verizon Developer Community >>>Take advantage of Verizon's best-in-class app development support >>>A streamlined, 14 day to market process makes app distribution fast and easy >>>Join now and get one step closer to millions of Verizon customers >>>http://p.sf.net/sfu/verizon-dev2dev >>>_______________________________________________ >>>Emc-users mailing list >>>Emc-users@lists.sourceforge.net >>>https://lists.sourceforge.net/lists/listinfo/emc-users >>> >>> >>> >> >>------------------------------------------------------------------------------ >>This SF.Net email is sponsored by the Verizon Developer Community >>Take advantage of Verizon's best-in-class app development support >>A streamlined, 14 day to market process makes app distribution fast and easy >>Join now and get one step closer to millions of Verizon customers >>http://p.sf.net/sfu/verizon-dev2dev >>_______________________________________________ >>Emc-users mailing list >>Emc-users@lists.sourceforge.net >>https://lists.sourceforge.net/lists/listinfo/emc-users >> > > > ------------------------------------------------------------------------------ > This SF.Net email is sponsored by the Verizon Developer Community > Take advantage of Verizon's best-in-class app development support > A streamlined, 14 day to market process makes app distribution fast and easy > Join now and get one step closer to millions of Verizon customers > http://p.sf.net/sfu/verizon-dev2dev > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users > ------------------------------------------------------------------------------ This SF.Net email is sponsored by the Verizon Developer Community Take advantage of Verizon's best-in-class app development support A streamlined, 14 day to market process makes app distribution fast and easy Join now and get one step closer to millions of Verizon customers http://p.sf.net/sfu/verizon-dev2dev _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
