Hey Steve, This might not be what you're asking for but from a process point of view, (and maybe you've already thought about this) this part looks like it would really lend itself well to some soft jaws and a fusion probing routine. Put 6 or 7 parts in there at a time with sufficient locating features to provide reasonably repeatable part positioning (at least enough to keep it within the overtravel range of the probing cycle in X and Y directions). You could mill a reference feature into the fixed jaw to define G59 X and Y, then set G59 Z off of whatever surface the finished side of the part would be resting on. Set up a probing cycle in fusion with your initial origin set to that point, and then go in and grab G54 P1-6 (or whatever) and store those as your various work offsets. That would cut you down from a manually run probing cycle for every part to a single initial set up of G59. The rest would be completely automated and could be cooked into the rest of your OP2 program. One button push for half a dozen or so finished parts. Chuck
When I make a video it always seems so crystal clear to me what I’m trying to convey. 😂However given the emails, txt, and the few comments here I didn’t come close to the mark. It’s really about adding functionality to the probing screens in very unique situations where you have a tiny feature to probe from and whether or not to add that to the screen sets. All great comments about work holding and setups however.
Soft jaws? The way I do complicated parts is I make aluminium jaws, put a thin parallel between the jaws, just to have some clearance, machine part shape or in this case considering the size - multiple parts, use same G54 you used for the jaws, for the part, you don't even need to use multiple coordinate systems
If you’re already jogging it to center of the part by eye, just estimate the distance to the tangent of the ball and use that number for that part instead of eyeballing it perfectly on the sides. Since you’re centering off a circle with a ball and the ball is going to the same z height for the probing then if it’s off slightly it won’t be any more than your current method. You could do that and you could set up a small stop on the one side and you can just probe the first one and every subsequent part will locate to the same home.
The feature you mention doesn't seem like a priority to me. Can't the issue be resolved with soft jaws? If soft jaws hold the part (or better - parts) with geometry that references the part, and you know the fixed soft jaw location, then you don't need to probe that dimension of each part
I’d have to agree. Disclosure: I don’t have a spindle probe, and just “work with what I’ve got”. There might be a better use-case for the feature you’re proposing. For the “hundreds of parts” scenario it’s worth either making soft jaws or parallel(s) + hard jaw + end stop along X-axis for consistent part placement. If you’re just decking the hat off then take a pass, measure the resulting Z and dial it in. If you need to also make features in the hat material a sacrificial part or two (out of “hundreds”) while dialling in X & Y is no issue. Soft jaws and/or hard stops make for very easy and repeatable part loading, and (for me at least) is probably more consistent than always pressing correct buttons over 100+ cycles.
does the screen show gcode when it's probing? it's probably calling a g65 to a sub program stored in the machine that you can alter the default z height values higher than standard. can't tell if you're asking for a solution to avoid the break off nub during xy centering or something else. the optional retract value might work looking at the conversational inputs. could probably also set the g54 p1 value to the +.25 for the probe cycle.
After you find the position you need to start probing from with the first part, would moving there via a line of gcode in MDI or a separate program work?
Maybe I wasn’t clear in the video what I was trying to convey. What you’re suggesting would work however it would be the most difficult way to do it. So I’m thinking there’s a disconnect.
Hey Steve,
This might not be what you're asking for but from a process point of view, (and maybe you've already thought about this) this part looks like it would really lend itself well to some soft jaws and a fusion probing routine. Put 6 or 7 parts in there at a time with sufficient locating features to provide reasonably repeatable part positioning (at least enough to keep it within the overtravel range of the probing cycle in X and Y directions). You could mill a reference feature into the fixed jaw to define G59 X and Y, then set G59 Z off of whatever surface the finished side of the part would be resting on. Set up a probing cycle in fusion with your initial origin set to that point, and then go in and grab G54 P1-6 (or whatever) and store those as your various work offsets.
That would cut you down from a manually run probing cycle for every part to a single initial set up of G59. The rest would be completely automated and could be cooked into the rest of your OP2 program. One button push for half a dozen or so finished parts.
Chuck
When I make a video it always seems so crystal clear to me what I’m trying to convey. 😂However given the emails, txt, and the few comments here I didn’t come close to the mark. It’s really about adding functionality to the probing screens in very unique situations where you have a tiny feature to probe from and whether or not to add that to the screen sets. All great comments about work holding and setups however.
Soft jaws? The way I do complicated parts is I make aluminium jaws, put a thin parallel between the jaws, just to have some clearance, machine part shape or in this case considering the size - multiple parts, use same G54 you used for the jaws, for the part, you don't even need to use multiple coordinate systems
If you’re already jogging it to center of the part by eye, just estimate the distance to the tangent of the ball and use that number for that part instead of eyeballing it perfectly on the sides. Since you’re centering off a circle with a ball and the ball is going to the same z height for the probing then if it’s off slightly it won’t be any more than your current method. You could do that and you could set up a small stop on the one side and you can just probe the first one and every subsequent part will locate to the same home.
Pop up window with a pre-fill from the previous entry, so one only has to hit Enter if no changes are necessary.
I'd love to help but first I need to buy my Syil CNC and learn to machine parts! 😮
The feature you mention doesn't seem like a priority to me. Can't the issue be resolved with soft jaws? If soft jaws hold the part (or better - parts) with geometry that references the part, and you know the fixed soft jaw location, then you don't need to probe that dimension of each part
I’d have to agree. Disclosure: I don’t have a spindle probe, and just “work with what I’ve got”. There might be a better use-case for the feature you’re proposing.
For the “hundreds of parts” scenario it’s worth either making soft jaws or parallel(s) + hard jaw + end stop along X-axis for consistent part placement. If you’re just decking the hat off then take a pass, measure the resulting Z and dial it in. If you need to also make features in the hat material a sacrificial part or two (out of “hundreds”) while dialling in X & Y is no issue.
Soft jaws and/or hard stops make for very easy and repeatable part loading, and (for me at least) is probably more consistent than always pressing correct buttons over 100+ cycles.
put a reamer hole in your jaw, if you just facing just probe your z ans for the x,y just centering with your eyes
does the screen show gcode when it's probing? it's probably calling a g65 to a sub program stored in the machine that you can alter the default z height values higher than standard. can't tell if you're asking for a solution to avoid the break off nub during xy centering or something else. the optional retract value might work looking at the conversational inputs. could probably also set the g54 p1 value to the +.25 for the probe cycle.
After you find the position you need to start probing from with the first part, would moving there via a line of gcode in MDI or a separate program work?
Maybe I wasn’t clear in the video what I was trying to convey. What you’re suggesting would work however it would be the most difficult way to do it. So I’m thinking there’s a disconnect.