spindle probing a cam profile

[Andris Skulte]

I've been asked if we can use our Renishaw MP7 probe to automatically reverse engineer cam profiles. Since this is new to me, I figured I'd start with a forum search...

 

Basically, what I'd like to do is bolt the cam onto a rotary table, jog the probe into position by hand, and then have a program index 1 degree, touch off the probe, note the location, back off 1", index 1 degree, and repeat until the full 360 degrees is complete. Right now we've got an operator manually probing the cam profile, and writing down the locations on scrap paper.

 

James mentioned a "custom macro b" here:

http://www.emastercam.com/cgi-bin/ultimate...t=012444#000000

What is that? We've got a Fanuc 11M control on the 'ole girl w/ the MP7.

 

I also saw this post:

http://www.emastercam.com/cgi-bin/ultimate...t=012563#000001

 

What should be my next step in learning how to use the probe? I wish I could go to a course, but there's no funds available... Thanks for the help, guys (and gals!).

[?Mark]

Metrolosys could propably do what you are asking so give them a call.

Highly recomended by James Mayette which is good enough for me

 

Kind regards, Mark

[chris m]

Andris

 

Try Sheila Schermerhorn, the 'Regional' rep for Renishaw: (413) 528-8954

 

C

[McRae]

If you have a probe, then you most likely have the capability to run custom macro b.

 

For that matter, with a program with 360 lines, probe along an axis and the program could either be configured to output the point to the RS232 or, you could write it down.

 

For the time it would take you to learn the custom macro B is the time it would take the guy to finish the "Manual" recording of the cam.

 

Are there a pile of Macros with your machine? I have the Renishaw book and that is how I learned the routines.

 

As Chris has stated - the Renishaw office is a good place to start.

[Jack Mitchell]

Andris,

 

I would lean towards sweeping the “Y” axis to ensure true center is established.

Next I would use a dial test indicator and rotate the rotary axis and record the “X” axis variance across all 360 points as you have described.

 

To test the result is to program in Mastercam without a spindle speed. You can figure this I am sure. The largest benefit is that the mounted tool (Test indicator) will follow the contour and you can see variations throughout the cam path. Another method would be to test cut the existing contour about .1” deep using a ½” solid carbide finisher and to feel the witness mark variations.

 

Once the data is obtained then it is also possible to calculate the throw using mathematics.

 

 

Regards, Jack

[Andris Skulte]

Excellent! Thanks for the good replies. RobK emailed a copy of the Renishaw Inspection Plus manual, and I downloaded the users guide from the Renishaw website, so I'll start there, and give Sheila a call once I actually know what I'm talking about

[John Summers]

If the follower movement is linear, you might be better off if you just find the angle of the beginning and end of the rise and fall sections, then using the cam program in CAM.MPK in MISC... on the FTP site. The reason is that cam profiles are very fussy, and it it is unlikely that digitized data would produce as smooth a profile as a cam generator program.

I haven't yet made a program for followers on a pivot, but I would consider doing so.

[Andris Skulte]

Over the last few days, I’ve been learning about Custom Macro B, with the intent of programming a Renishaw MP7 probe for reverse engineering cam profiles. We currently use a rotary 4th axis, index 1 degree, and come down on the Z, so there would be 360 data points. The biggest area of confusion right now is how to get the probed data out to a text file (RS232 to DNC?), to import to excel. From what I can tell, the only outputs from macros are using POPEN, then DPRINT or BPRINT, then PCLOS. Since the machine and the DNC is quite slow, I figured the safest way would be for the macro to make it’s own program (say O0001), write each position as a new line (DPRINT[N1*Z*POSITION*AT*1*DEGREE*IS***#501]), and have the operator punch out the program to the DNC once it's finished.

 

We're just using this data to get rough industrial cam profiles from customers that have ancient, worn out cams and no drawings, and with the data, we can figure out a better motion for the new cam. Since we have more time than $ (right now), I'm using this as an opportunity to learn macros, probing, etc... The machine is about 10 years old, and has the old original Renishaw macros loaded in the 8000 series, FWIW.

 

So… What I’m wondering is how to get the probed data out of CNC world, into people world.

 

Thanks to RobK, JamesM, and Sheila from Renishaw who sent over some documentation about macros to get me started, and all you smart Mcam guys who've been at this a lot longer than myself...

 

Andris

[McRae]

We use the DNC port to send out data all the time on our Fanuc, Tosnuc and Fusion 640 controls.

 

We just initialize the port, and keep sending it out, then at the end, close the port. The DNC program will capture any data that you send to it and you will need to adjust the timeout setting on the DNC unit to allow the machine time to do this. Then save the file in the DNC system and that will do waht you are asking.

[thoskins]

You can use capture rext in Hyperterminal. I have done this before to collect data from an Renishaw probe.

[Andris Skulte]

Andrew & Thoskins,

Thanks. I've only used the DNC to punch and download programs to the CNC's, so this is new to me. I guess it's similar to punching out programs, and should be in the "Received Files" area of the DNC program...