Block Edges is Acting Weird

[[Ry...]]

Hey Folks,

So this weird thing happened to me today. I had an operator mention that he was having trouble with a program. (Turns out it had nothing to do with the program, but was an operator error in how he took points on the plane during manual alignment).

However, it initially got me digging into my program to determine if I made an error that was causing him the troubles. And while going over every aspect of the program, I checked the blocked edges and discovered this (See attached picture).
What should be the normally circled edges of all the -Z faces, are now on the -Y edges.
This is the most basic of programs. A basic rectangle with a top plane for rotation in space, and a 2D line for planar rotation. And then the basic 2D lines for X and Y origin and plane for Z origin. Nothing out of the ordinary. I normally dont worry about the blocked edges in my basic programs, unless its more complex with star probes and tight clearance around the part. This hasn't caused me any troubles to date so far.

Now, years ago I would have issues when I would try and change my base alignment, calypso would hiccup and flip the part over and then say that -Z plane was facing up and the circled blocked edges would be on the top. I could never figure out how to fix this and re-establish the base alignment to the correct orientation, so I would usually just start over as it wouldn't even allow me to probe the top plane as it thought it would go through the part.
Anyhow, I haven't had that issue in years until today. The funny thing is that the program works flawlessly as is right now.
But I'm curious as to why the common blocked edges look like this.

Thanks for your help,

Calypso Alignment Block Edges Error.pdf

[[Gu...]]

Hello,

If you flip your part defining Base System rather than using CAD transformation, blocked edges became crazy :p
Just run your Base System (offline or online) and then recheck your blocked edges, this should solve your problem.


Hope it helps.

Guillaume

[[Je...]]

.

Please sign in to view this quote.

.



Ryan,

The reason your base alignment and blocked edges were inverted has to do with the fact that 2D lines have vector. As you mentioned, the operator may have been probing in an incorrect order during the base alignment. Is it possible that he was also probing the line that constrains rotation in the incorrect order (vector), either left to right when it was supposed to be opposite, etc.? Calypso specifies the line's normal vector during the base alignment by showing in the CAD window where the first point (origin) of the line begins. If it is probed in the opposite order, this changes the space plane that Calypso assigns to the part, which is why the blocked edges appear changed.

I could be off my rocker, but this is the cause of similar situations that I have run into. In your particular case, using a 3d line from a circle in one of the holes to a circle in the opposing hole would constrain rotation nicely. Or, you could substitute planes for the current 2d lines. Just ideas. You could also use a start alignment to simplify the manual probing, but this requires that the part be oriented fairly accurately in the fixture.


Jeff Frodermann
Meier Tool & Engineering
Anoka, Minnesota


.

[[Ow...]]

Jeff Frodermann is not off his rocker here, I've seen it happen many times. As he mentioned, the best way to fix it to where it doesn't happen again is use planes on the faces instead of 2-D lines because you can't take the points on the plane in the wrong order (only 3 is needed) and NOT mess up the space axis, which flips the alignment along with the blocked edges.

If you're adamant about using 2D lines, you can create two individual points on the face, name them by the position they're in (at left front face, at right, ect) and then recall them into a 2D or 3D line. This way the operator reads the name and position of the point being taken if he's not looking at cad that should show it in the correct place to be taken.

To correct the alignment once fixes are implemented and to reset the blocked edges, go to resources, utilities and all the way at the bottom of the menu is "set base alignment to zero", click it and it will say "base alignment deleted". Then go to the blocked edges and they will be back in the correct place with the circle at the bottom (you never want to travel underneath a part :-)_). THEN..... run a manual base alignment making sure the points are in the correct order if you chose to stay with 2D lines.

[[Ry...]]

Gentlemen,

Thank you for your help.

I managed to delete the base alignment, and redo it in the correct order of probing. Blocked edges are back to normal.

Now out of curiosity, I went back and pretended I was the operator and probed everything in the incorrect order. The result, was nothing happened to the base alignment or the blocked edges. Once I finished the manual alignment, CNC took over and probed everything in the correct order, no matter what order I probed the 2D lines in.
This is what I am used to for the years I've been programming. I use 2D lines (instead of planes) for their 2 point minimum simplicity and easy and quickness of probing for my operators to setup the alignment and get back to their machines as quickly as possible. Same reasoning I dont use the holes for alignment, as that would require 6pts minimum of the combined circles. I know this is splitting hairs for the time required to probe, but if I can keep the operator away from the CMM as much as possible means more time on the machines.
I've never had a 2D line affect the base alignment no matter which way it was manually probed. CNC probing always probed in the correct orientation and the base alignment during the run has always been correct.
As well, I normally use a Start alignment for most everything, however, this part is so thin that I'm only taping it to the table and relying on the operator to align it to the machine, hence its not guaranteed square and unable to use a basic 3 pt start alignment.

I'll continue to program with 2D lines but really watch for this in the future and try and mimic what caused to figure out exactly.

Thanks,

[[Ke...]]

Please sign in to view this quote.

Just a reminder that a very simple Start Alignment can reduce operator error, and save time.