The ones I have are at least 5 years old and do have optoisolated inputs. I'm not aware of the specifics of the -NR version, but the datasheet for the ones I have says "The step clock (SCLK) and direction (DIR) inputs are buffered using Schmidt triggered buffers for increased noise immunity ...".
I am providing a 5V opto supply to the IM483s with a homebrew switching PS, so wondering if noise on that may be an issue. Also, I do see one issue with the direct wiring I have now: if I'm only putting out 3.3V from the step/dir signals with 1.8Vp-p of noise, and it has the 5V pullup, then the opto input LEDs are seeing signals of 4.2V and 2.4V which may be enough to be detected as on/off pulses. Yes, the noise is the problem, but translating my parport signals to clean 0V-5V signals should be a major help here. All seems to be wired and working fine, except for the occasional noise. Only reason I'm not using shielded cable is because I don't have any. I'm going to find some USB cables I can hack up and try those. Cheers, -Neil. On 3/8/2015 4:45 PM, Bruce Layne wrote: > The C10S looks as if it provides 3.3V (parallel port) to 5V (driving > the stepper drivers) voltage translation, but I'd assume any > optoisolator interface would do that, including the homebrew version > you were using? > > There seems to be an IM483-NR version of your stepper driver that's > listed as "noise reduction at inputs". So the IM483 doesn't have any > sort of noise filtering at its inputs? > > http://motion.schneider-electric.com/products/im483.html > > I'd focus on killing the noise. 1.8V of noise is a killer on a 5V > signal. I'd run shielded cable for everything - motors, limit > switches... everything. Ground the shields directly to the subpanel > where the stepper drivers and optoisolator card are mounted. Leave > the other ends of the cable shielding unconnected to prevent current > loops (current flow in the shield). Don't run the power cabling next > to control signal cabling. Use a good quality IEEE 1284 bidirectional > parallel port cable. > > Are you sure the optoisolator board and/or stepper drivers are wired > correctly? The inputs on the cheap eBay stepper drivers that I use > are optoisolated. To turn on the optoisolator at the inputs would > require a few milliamps of current, which is a lot for electromagnetic > interference noise to source. If you have stepper driver inputs that > have a floating common that is internally biasing itself based on > leakage currents from the inputs, then the inputs would be much more > likely to trigger on noise. > > > > > > On 03/08/2015 03:34 PM, Neil wrote: >> Hi all, >> >> Haven't found a clear answer to this yet in the archives, so hoping >> someone can help. >> >> I've just gotten back to a homebrew CNC machine project from a few years >> ago and need to figure out why I'm getting incorrect/missing steps. >> Using an Intel D525MW and IM483 stepper drivers. I was using a homebrew >> opto-isolator board, and got the incorrect/lost steps, so I tried >> without (direct from par-port to drives) and getting a lot of chatter. >> With this, moving the X-axis gets me enough noise on the Y-axis lines >> that the Y-axis will move. >> >> An oscilloscope shows me that the parallel-port output seems to be 3.3V, >> and I've got noise of up to 1.8V on the step line! Other than shielded >> wires, I need a decent buffered breakout board. Space is very tight, so >> leaning towards the C10S from cnc4pc.com here... >> http://www.cnc4pc.com/Store/osc/product_info.php?cPath=33&products_id=696 >> >> >> Do you think this would resolve my issue? >> >> Thanks. >> >> >> ------------------------------------------------------------------------------ >> >> >> Dive into the World of Parallel Programming The Go Parallel Website, >> sponsored >> by Intel and developed in partnership with Slashdot Media, is your >> hub for all >> things parallel software development, from weekly thought leadership >> blogs to >> news, videos, case studies, tutorials and more. Take a look and join the >> conversation now. http://goparallel.sourceforge.net/ >> _______________________________________________ >> Emc-users mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/emc-users >> >> > > > ------------------------------------------------------------------------------ Dive into the World of Parallel Programming The Go Parallel Website, sponsored by Intel and developed in partnership with Slashdot Media, is your hub for all things parallel software development, from weekly thought leadership blogs to news, videos, case studies, tutorials and more. Take a look and join the conversation now. http://goparallel.sourceforge.net/ _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
