Skewed Datum set up without Probe

[KevinV_MEI]

I have a Heidenhain 426, which has the capability to use a probe to define datums, alignment, etc. I have the chance of winning a project that is slightly longer than the X axis of my machine (38.5") at 41". It is a narrow rail that I could run on an angle. I see in the manual how to use a probe to align to fixture and it's axis, but how would someone align a fixture using the controller and say two pins? I've seen it used on other machines and know this one can do it just need to know where to poke around.

This may not be the right forum for this topic but I think is is more Heidenhain specific than process specific.

Kevin V.

[awemawson]

Guessing here, but probably the probe/skew alogorithm is popping values into dedicated Q parameters. If you read your manual, just possibly you can find the locations and pop them in yourself.

Mind you a probe is a mighty fine tool to have - I'd be tempted to see if eBay could provide!

[gus]

write your program as if it were straight, then rotate the coordinates.

do the same for machining your fixture.

[KevinV_MEI]

Originally Posted by gus write your program as if it were straight, then rotate the coordinates. do the same for machining your fixture.

Thanks Gus, that's what I want to do. Where/How do I do that on the controller?

[gus]

Cycle Define

Coordinate Transformations

Rotate

I would suggest reading the book

I believe there are programming manuals on the heidenhain website

[KevinV_MEI]

Thanks Gus. I think what you're referring to is a G73 rotation. That would be useful if I knew the rotation angle. What I want to do bolt a fixture or workpiece down without regard to its parallelism to the table, pick up a point, say a hole center, then pick up another, and then have the controller compensate for the angle between the two in reference to the X axis.

[christinandavid]

You can use functions in your program to calculate the angle between your hole centres (look into trigonometric Q DEF FN:s and also ANG and LEN). The result you store as a Q value and this can then be called up in your Rotation CYCL.

Without a probe you will need to input the positions into your program every run, so if you can set one position as zero every time you've halved the chance of a balls-up.

I have implemented this method successfully on Heidenhain conversational on a job that was too large and heavy to tie up precisely with the fixture.

DP

[gus]

First, if you are programming in G code, you have lost half the value of this control

second, gawd man, just do a little math. 10 seconds of trig will give you the numbers you need, then write the freakin program. I think I have spent more time typing than it would have to calculate the hypotenuse needed, plus a tool diameter plus the hypotenuse of the short side to do it all

[KevinV_MEI]

Thanks guy. I came to the conclusion over the weekend I'd have to pick up one point, pick up the other and just trig out the triangle to come up with the angle. No problem. And I also agree Gus, the power of this controller is lost in g code, it's just the two programmers I'm using haven't had time to get their conversational posts tweaked and keep up with due dates, etc. We're going to keep moving in that direction, it just may take time.