Jump to content

Welcome to eMastercam

Register now to participate in the forums, access the download area, buy Mastercam training materials, post processors and more. This message will be removed once you have signed in.

Use your display name or email address to sign in:

PROBING MATH EQUATION QUESTION


Recommended Posts

So this one is for brains bigger than mine. I have a casting and the datum target points are the center of a diameter. This diameter lies perpendicular to my Z.  Ultimately it is not hard for me to probe to find the center of diameter in Y and then probe the Z to adjust my Work offset. See program below

O7699
G00 G17 G20 G40 G80 G90
G91 G28 Z0.
T299 M6 (CHANGE TO RESHAW PROBE)
M69 (UNLOCK)
M11 (UNLOCK)
G00 G17 G90 G57 B0 C0.
M68 (LOCK)
M10 (LOCK)
G43 H299 Z2.0 F100.
G65 P9832 (TURN ON PROBE)
G65 P9810 Y0. F250. (PROTECTED MOVE IN Y)
G65 P9810 X-.6875 F250. (PROTECTED MOVE IN X)
G65 P9810 Z0. F100. (PROTECTED MOVE IN Z)
G65 P9810 Z-4.674 F50. (PROTECTED MOVE IN Z)
G65 P9812 Y11.6 S4. (PROBE Y AND SET G57)
G65 P9810 Z-7. F50. (PROTECTED MOVE IN Z)
G65 P9811 Z-7.427 S4. (PROBE Z )
G65 P9810 Z0. F50. (PROTECTED MOVE IN Z)
G65 P9833 (SWITCH PROBE OFF)
G91 G28 Z0. (REF RETURN)
G91 G28 X0.
G91 G28 Y0
M30
 

The problem is the castings are all over the place in the diameter, So I need a way to calculate the true diameter from the 3 points so the Z can be adjusted properly. Hopefully that's clear enough to understand what I'm asking. I know I can store values and do calculations by adding some macros, but I'm not sure what the equation should look like. Any help is appreciated, sample file attached so u can see what I'm doing. Thanks in advance

CASTING PROBING EXAMPLE.mcam

also a screen shot for those that dont wanna download the sample file

casting probing.JPG

Link to comment
Share on other sites

Pete, I would try measuring with a circle or a cylinder routine. Collecting just points will require you to do some least squares match to figure out from a nominal diameter what you are getting from the collected points on the cylinder and then make decision about how much from the correctly fitted shape that deviation is to then see if the difference is enough to accept or reject. Where as using predefined routines that logic is already part of the macro and you don't have to do a lot of math with the collected data.

Link to comment
Share on other sites
25 minutes ago, crazy^millman said:

Pete, I would try measuring with a circle or a cylinder routine. Collecting just points will require you to do some least squares match to figure out from a nominal diameter what you are getting from the collected points on the cylinder and then make decision about how much from the correctly fitted shape that deviation is to then see if the difference is enough to accept or reject. Where as using predefined routines that logic is already part of the macro and you don't have to do a lot of math with the collected data.

Great suggestion, bad situation for it. The model I made is far from a true representation or the part. I has a .750 lip around that feature from the side. Otherwise I would have just rotated it 90 and probed it with a 3 point vector circle. to do it on this part would require changing the probe tip to a big disk, And the machine has 5 other pallets running various parts that require a standard 6mm probe ball. That and at this point I'm infatuated with how to do the math.. lol

Link to comment
Share on other sites

I would give them to the Inspection and ask them to measure them on the CMM and tell them to give me a reference back to a feature on the Fixture. Use a Flat Plane and Tooling ball and have them tell you where the Zero needed to be for each part.

Here is a little Light Reading on Fitting of Circles using LSM.

Link 1

Link 2

Link to a book

MIT Link for Linear Algebra

  • Like 1
Link to comment
Share on other sites
1 hour ago, crazy^millman said:

I would give them to the Inspection and ask them to measure them on the CMM and tell them to give me a reference back to a feature on the Fixture. Use a Flat Plane and Tooling ball and have them tell you where the Zero needed to be for each part.

Here is a little Light Reading on Fitting of Circles using LSM.

Link 1

Link 2

Link to a book

MIT Link for Linear Algebra

I do like that idea , Might have to run that up the chain of command. I can tell you that we have grown accustom to waiting around weeks at a time for an inspection, So I don't see how that could complicate things any worse.. lol that was a joke, but definitely a viable solution. Knowing the radius on each part would make my original probing routine completely full proof, just adjust the Z number in this line of the code

G65 P9811 Z-7.427 S4. (PROBE Z )

Link to comment
Share on other sites

Also I think I'm getting closer to a math solution to define the radius. Basically I can get the linier width (record it) and I can probe the z from there that length = the saggita, with those 2 numbers I can calculate the radius

 

excel formula looks like this =(B2/2)+(B1^2/(8*B2))

B2 Represents the box B2, which has the height (sag)

B1 represents the boxB1, that has the linier width

 

with that radius I might be able to write all the macros I need to set the z on the machine?

  • Like 1
Link to comment
Share on other sites
3 hours ago, motor-vater said:

here is a more accurate representation of what I am dealing with, the opposite end of the part also has a over hang, so really getting the 3rd point in the z is the only option I can see

 

casting 2.JPG

Hi Pete,

Why not just perform Vector Calibration, and then "hit 3 points on the pieces of the Arc"?

You just need to call O9823, with A-3. B-29. C-174. 

The angles of the 3 Points, are measured +-180. Degrees, from the X+ Axis, as "zero" for the angles.

Drop the probe 1" above the hole with Protected Positioning, then just call:

G65 P9823 A_ B_ C_ D_ Z_

A = 1st Angle for Vector measurement (taken from Diameter Center)

B = 2nd Angle

C = 3rd Angle

Just need to be sure all 3 of those angles are actually "achievable" with wherever that angled wall cuts off the measurement.

Measure the Circle, don't measure the angled wall, as that thickness could be off, but the circle may not be.

https://www.haascnc.com/content/dam/haascnc/en/service/manual/probe/english---renishaw-inspection-plus---programming-manual---2008.pdf

Page 8-10.

  • Like 1
Link to comment
Share on other sites
17 hours ago, Colin Gilchrist said:

Hi Pete,

Why not just perform Vector Calibration, and then "hit 3 points on the pieces of the Arc"?

You just need to call O9823, with A-3. B-29. C-174. 

The angles of the 3 Points, are measured +-180. Degrees, from the X+ Axis, as "zero" for the angles.

Drop the probe 1" above the hole with Protected Positioning, then just call:

G65 P9823 A_ B_ C_ D_ Z_

A = 1st Angle for Vector measurement (taken from Diameter Center)

B = 2nd Angle

C = 3rd Angle

Just need to be sure all 3 of those angles are actually "achievable" with wherever that angled wall cuts off the measurement.

Measure the Circle, don't measure the angled wall, as that thickness could be off, but the circle may not be.

https://www.haascnc.com/content/dam/haascnc/en/service/manual/probe/english---renishaw-inspection-plus---programming-manual---2008.pdf

Page 8-10.

Honestly I did not know I could do a vector measurement in Z? Ill have to look into that a bit more. If that actually works its a done deal. But in order to do it in X I need to rotate to B90. and then use a disk style probe to clear the lip. The circle is cast and there for very choppy. I would rather use the same Datum target points the print calls out because Inspection will be using those as well. That datum is "as cast" to target points... so it never gets cut.... Real PITA!

Then even if I could pull it off I have never really tried probing anything rotated, would no doubt require a G68.2 and some macros to record values and update X to Z and things like that. But I could be wrong. Definitely been hitting the boards hard seeking information about probing under rotation. Not coming up with a lot of information. I'm probably over thinking the whole thing and will end up setting multiple work offsets , but I hate doing it. Makes the operators job 3 times as hard for something that really is a few cuts... I like to help make my guys job ez so they can be successful, Hard enough to find good guys anymore, hate to scare the ones I have off. LOL

  • Like 3
Link to comment
Share on other sites
36 minutes ago, motor-vater said:

Honestly I did not know I could do a vector measurement in Z? Ill have to look into that a bit more. If that actually works its a done deal. But in order to do it in X I need to rotate to B90. and then use a disk style probe to clear the lip. The circle is cast and there for very choppy. I would rather use the same Datum target points the print calls out because Inspection will be using those as well. That datum is "as cast" to target points... so it never gets cut.... Real PITA!

Then even if I could pull it off I have never really tried probing anything rotated, would no doubt require a G68.2 and some macros to record values and update X to Z and things like that. But I could be wrong. Definitely been hitting the boards hard seeking information about probing under rotation. Not coming up with a lot of information. I'm probably over thinking the whole thing and will end up setting multiple work offsets , but I hate doing it. Makes the operators job 3 times as hard for something that really is a few cuts... I like to help make my guys job ez so they can be successful, Hard enough to find good guys anymore, hate to scare the ones I have off. LOL

I would highly recommend just placing a call to Renishaw, and getting them involved to help you figure out the process. Their Customer Support is outstanding.

Renishaw Inc
 
1001 Wesemann DriveWest Dundee IL 60118USA
  • Like 4
Link to comment
Share on other sites
4 hours ago, Colin Gilchrist said:

I would highly recommend just placing a call to Renishaw, and getting them involved to help you figure out the process. Their Customer Support is outstanding.

Renishaw Inc
 
1001 Wesemann DriveWest Dundee IL 60118USA

When you are right you are right, I remember making a commitment to myself years ago, If I couldn't figure something out in 30 min I would ask for help. Here I go breaking my own rules.. lol Thanks Colin.

4 hours ago, mwearne said:

https://www.cncci.com/post/three-point-arc-center-macro

He is doing XY while you are wanting YZ, but the math should be relevant. 

This is outstanding and exactly the kind of stuff I love finding on the interweb, Thank you for sharing it. Weather or not I use it now it is definitely getting saved into my probing reference materials. Thank you

  • Like 2
Link to comment
Share on other sites

So I talked to the Guys at Renshaw. Looks like to do what I want would require a strain gauge probe and newer software than what this control has. This is Just a OMP60 from 2011. But he made a recommendation that I am going to give a go. Since I am already probing the Y with a P9812 (web probe cycle he recommended writing the error which is recorded to #143 to a #500 then using that error to update my z value after running my P9811 ( z probe cycle). And that might get me close enough. I am going to play with this more Monday.

I am letting you guys know all this because I have always gotten frustrated when searching the achieves for info, and the post went dead when someone figured something out. I am all about sharing, so once I get code that works I will post it for the world. Lol

  • Thanks 1
  • Like 6
Link to comment
Share on other sites

Here is what I came up with! More details to follow

 

%
O7699
(PROBE PARKER 7967460)
G00 G17 G20 G40 G80 G90
G91 G28 Z0.
T299 M6 (CHANGE TO RESHAW PROBE)
M69 (UNLOCK)
M11 (UNLOCK)
G00 G17 G90 G57 B0 C0.
M68 (LOCK)
M10 (LOCK)
G43 H299 Z.0 F100.
G65 P9832 (TURN ON PROBE)
G65 P9810 Y0. F250. (PROTECTED MOVE IN Y)
G65 P9810 X-.6875 F250. (PROTECTED MOVE IN X)
G65 P9810 Z-4.0 F50. (PROTECTED MOVE IN Z)
G65 P9812 Y12.5858 S4. (PROBE Y AND SET G57)
#540=#143/2 (DIVIDES LINIER ERROR FROM Y MEASERMENT AND SAVES TO #540)
G65 P9810 Z-7. F50. (PROTECTED MOVE IN Z)
G65 P9811 Z-7.427 S4. (PROBE Z )
#541=#5283-#540 (SAVES ADJUSTED RADIUS MEASUREMENT TO #541)
#5283=#541 (ADJUST Z TO ACCOUNT FOR RADIUS ERROR)
G65 P9810 Z0. F50. (PROTECTED MOVE IN Z)
G65 P9833 (SWITCH PROBE OFF)
G91 G28 Z0. (REF RETURN)
G91 G28 X0.
G91 G28 Y0
M30
%

  • Like 1
Link to comment
Share on other sites

A few things I had to factor in

The linier distance in Y must be a perfect representation for this to calculate the correct error. I did that by calculating the Y value of where the probe stylus actually contacts the diameter see illustrations.

1635233278_PROGRAMMERILLISTRATION1.thumb.JPG.c7c80c2e53f901d737514811ed385d5f.JPG

241701791_PROGRAMMERILLISTRATION2.thumb.JPG.cd3948719484635c286226fe21e2280c.JPG

As you can hopefully see the radius of the 6mm ball actually contacts the part a little higher than the actual z-4 value called by the program, so I found the contact point and was able to come up with and actual linear value based on that contact line.

Summery: All though this is functional, it is not a solution for this situation, problem is with out starting from a theoretical perfect z zero position, the linier value of the contact line will change, which upsets the error value. If it is higher then expected the linier would be longer, if lower it would be shorter, that could potentially lead to a work offset that is not where expected in the z. 

In the end I am just going to have to manually check the radius of each part on the CMM and update the z value of my original program from part to part, it will always find the center of Y no mater where in the depth I probe it and if I have a known radius I have my z value. Thank for all the help guys, hopefully this helps someone else down the road.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.

Join us!

eMastercam - your online source for all things Mastercam.

Together, we are the strongest Mastercam community on the web with over 56,000 members, and our online store offers a wide selection of training materials for all applications and skill levels.

Follow us

×
×
  • Create New...