Usually I'm asking for help here but in this case I'm trying to contribute something useful.
Recently I ran into an issue with a 1979/80 mori seiki SL- Yasnac 2000B control, all yaskawa drives/motors.
This info will hold valid for any mori/2000B of this era.
The machine would auto zero return in z but in x it would hit the creep feed proximity switch, start creeping, and then proceed to creep feed right to the over travel limit switch.
Now, this is a common problem, encountered on machines across the board, not just this machine/control platform.
The problem can usually be tracked right down with 100% confidence in the diagnosis using the machines diagnostic input/ output state checking function(refer to your manuals to access)
Usually it comes to be that a single piece of swarf is stuck in front of a prox switch or gunk and swarf have bound up a mechanical limit/home switch. Sometimes it can be traced all the way back to a board having pooped the bed.
In this case what the problem was had nothing to do with either. The first thing I did was pull both axes encoders in order to make the problem change axes by installing the encoders on each others original axis. This will tell you without a doubt that th encoser is or isn't the problem. In this particular case the encoders were swapped, the machine powered up as usual and each axis homed. Heres the kicker, both axis homed properly first time around. What happened you ask? Whaaaaale.....upon removing each encoder unit I inspected both for any physical damage, dirt/oil on the lens, burnt smells, etc. While I found that the encoder I was using as a known "good" unit had a completely wrung out shaft bearing, I knew it worked. When I inspected the suspected "bad" one I noted a blotch of gook stuck to the lens spanning over 30the scan bars and covering the "home pulse" signal on the track( easy to see, its the only clear area in an otherwise shaded 360 tracl of the lens) and set about snatching up the last qtip in my cabinet and a capful of alcohol. Needless to say, the splotch of evaporated gunk, even though it was semi transparent was enough to hide the home pulse marker as well as a few pulse counter marks. This is why when the encoders were switched the machine was miraculously cured. Basically what I'm saying is if you run into this issue with your machine and switch encoders and find that you have a bad encoder, check that baby out with a damn magnifying glass if you must. You could save yourself over $1,000. Need I say more.
I should also add that changing parameter no.62 (zero return final speed) to make it easier on the control to count the pulses can sometimes fix a sketchy home signal.
Standard param for x is 10, you want to knock this down to 4. Any slower and will take ten minutes to get to the home pulse from the deceleration trigger.
Imagine counting fan blades, or tracking them with your eye, obviously if I slow the fan down the blur of blades becomes crystal clear. A valid test that won't get you dirty.
Recently I ran into an issue with a 1979/80 mori seiki SL- Yasnac 2000B control, all yaskawa drives/motors.
This info will hold valid for any mori/2000B of this era.
The machine would auto zero return in z but in x it would hit the creep feed proximity switch, start creeping, and then proceed to creep feed right to the over travel limit switch.
Now, this is a common problem, encountered on machines across the board, not just this machine/control platform.
The problem can usually be tracked right down with 100% confidence in the diagnosis using the machines diagnostic input/ output state checking function(refer to your manuals to access)
Usually it comes to be that a single piece of swarf is stuck in front of a prox switch or gunk and swarf have bound up a mechanical limit/home switch. Sometimes it can be traced all the way back to a board having pooped the bed.
In this case what the problem was had nothing to do with either. The first thing I did was pull both axes encoders in order to make the problem change axes by installing the encoders on each others original axis. This will tell you without a doubt that th encoser is or isn't the problem. In this particular case the encoders were swapped, the machine powered up as usual and each axis homed. Heres the kicker, both axis homed properly first time around. What happened you ask? Whaaaaale.....upon removing each encoder unit I inspected both for any physical damage, dirt/oil on the lens, burnt smells, etc. While I found that the encoder I was using as a known "good" unit had a completely wrung out shaft bearing, I knew it worked. When I inspected the suspected "bad" one I noted a blotch of gook stuck to the lens spanning over 30the scan bars and covering the "home pulse" signal on the track( easy to see, its the only clear area in an otherwise shaded 360 tracl of the lens) and set about snatching up the last qtip in my cabinet and a capful of alcohol. Needless to say, the splotch of evaporated gunk, even though it was semi transparent was enough to hide the home pulse marker as well as a few pulse counter marks. This is why when the encoders were switched the machine was miraculously cured. Basically what I'm saying is if you run into this issue with your machine and switch encoders and find that you have a bad encoder, check that baby out with a damn magnifying glass if you must. You could save yourself over $1,000. Need I say more.
I should also add that changing parameter no.62 (zero return final speed) to make it easier on the control to count the pulses can sometimes fix a sketchy home signal.
Standard param for x is 10, you want to knock this down to 4. Any slower and will take ten minutes to get to the home pulse from the deceleration trigger.
Imagine counting fan blades, or tracking them with your eye, obviously if I slow the fan down the blur of blades becomes crystal clear. A valid test that won't get you dirty.
Last edited:
