Star Probe Help

[Wes Cisco]

I have built a star probe using 3mmx10mm tips. I have a dual sphere qualification tool. I am using all 5 tips in the A0B0 position only. I select the tips and send it off to calibrate DCC. Tip # 3 must calibrate on the bottom sphere, whereas the other 4 use the upper sphere. I did not create the artifact and the help menus are no use, but the software seems to know I have two spheres and where they are located as it goes to the bottom sphere for #3. BUT when I call up Tip #3 to use it in the program the demon thinks it is in a different place from where it really is by the difference in the actual location of the two spheres. Any help would be greatly appreciated.

[george frick]

i know there must be a better way- but in the past i have had to manually calibrate star probes i have built to have them measure accurately. you can try this and check it to a ring gage or other artifact to verify the results.

[J Temmen]

Delete the cal spheres and recreate them so you are sure of the vectors and locations, then redo allprobe cals.

[John Kingston]

Wes, I'm not at work right now, but use a star probe all the time. I've always gotten all 5 tips to calibrate on the upper sphere.
First, you need to have your star square to the machine.
What I do is place a 123 block on my plate.
I have a Rayco plate so it is easy to get that square.
When I put the star in I eyeball it square to the block.
Your dual calsphere needs to be pretty square to the machine also.
Next, I calibrate each tip individually, starting with #1.
It might take a few tries, but it'll work.

Let me know if that works.
Good Luck!
John

[JamesMannes]

Dual sphere and star probes

What you need to do is what I call a "dual sphere" qualification.
Qualify all your tips on the top sphere("A" sphere; 0,0,1 vector) that you can. Answer "yes" to the question "Has the qualification sphere moved?". Take one sample hit on the top of the sphere and qualify. Then create a new qual sphere(ADD TOOL; call it "B" sphere). At this location, make the vector direction 0,-1,0(if you have the spheres set up such that the top sphere is at the top and farther away(+Y) from the lower sphere), or 1,0,0 if the lower sphere is to the right of the top sphere(+X). Begin by qualifying your 0,0 tip(answering "YES" to the question "Has the qualification sphere moved. Qualify 0,0. Then go on to qualify tip #3 of your star probe configuration. What you are doing is building a relationship between sphere "A" and sphere "B" by qualifying 0,0 at both locations.
See attachment(hope I"ve helped).
Call me if you want 262-377-3270 x212 6:30 a.m. - 2:30 p.m.

[McGyver]

Star probe qualification

All 5 probes MUST be qualified on one sphere to ensure accuracy.

I had the same problem (on A90B0 & A90B180 where I had to qualify tip 5 on sphere 2) the results was that all roundness' of the circles measured were 0.5mm up to 0.9mm out, I tried everything I could think of but could'nt get it right. I even entered an offset into tip 5's tooldata, it worked but it is not the solution.

In the end I turned the calibration spheres to allow access to qualify all 5 probes on one sphere which solved my problem.

[Wes Cisco]

Thanks for the responses

Thanks to all who replied.
John K: How do I calibrate tip #3 on the upper sphere without fouling tip #1 on the shank of the calibration tool? I have the star square to the machine axis, (pretty darn close at least, I eyeballed to my alignment nest); likewise the calibration tool is
square to the y axis with the upper sphere y+ from the lower.
Jammes, Thank you for the doc. I will try this when I have a chance. Due to the need to get the job done, I gave up on the star last night and wrote the program using a single tip indexed. I can not get as many hits on some surfaces as I would like, but I can get at least 1 or 2 hits every where I need to verify the print dimensions. The objective this time was to verify surface plate inspection rather than regular recurring inspection. But I will probably be comming back to this job soon to write a full fledged program and I will need the star probe then.

[John Kingston]

I use a 1mmX10mm and a 2mmX15mm. I have an exentsion in between the star and my TP20 module. Both work, after a few calibration runs. Tip3 is usually the most troublesome. If you keep running its cal, it'll adjust for it. I forgot to bring in my PDA, so I can't get you a picture of my set up. I'll try to do that Monday.

[Rick Coleman]

star probe

The tips do not have to be calibrated on the same sphere, the spheres do have to relate to each other though. I have used the attachment here on several occasions and it works out pretty slick. Got it from a techie at B&$.

[McGyver]

Thanks Rick,

I have qualified my star probe (A90B0 & A90B180 and tip 5 being a problem) on one sphere only (I did'nt know about and how to use parameter sets etc.) the pdf document you submitted was very helpfull and cured a headache I had for many days.

thanx
McGyver.

[Wes Cisco]

I know it's Saturday

Hello all,
The job is back and I must get the star probe to work now. I have printed both documents posted here and have tried to follow them very carefully several times. Tip #1 always fouls on the qual. sphere shank when I try to perform the first calibration on the second sphere. I have deleted and made new calibration tools several times. I am positive that both the artifact and star tips are aligned to the machine axis within .005". I have tried (0,-.7071068,-.7071068/ 0,-1, 0/ 0,-1,-1/ and probably a few other vectors for shank of the second sphere. The upper sphere's shank vector is 0 .7071068,.7071068 and it works fine. Logically, since it is lower and in y- from the 1st sphere I think it should be 0,-.7071068,-.7071068.), but nothing works.
I have been in touch with my predecessor Ken Amason, (who many of you know), and he says I should be able to get all 5 tips to calibrate on the top sphere. My set up is as follow, I am using a dual sphere both .75" dia. The two spheres sit at 45 degrees to the z axis. The top sphere is y positive. If I try to calibrate tip #3 on the top sphere it fouls on the qual. sphere shank. You say you get all 5 tips on one sphere. JOHN K.: Please post how your sphere is set up, and what parameters you use for calibration. (start/stop angle, number of hits/levels, prehit retract etc.) I am back to the 1mmX10mm tips for the star probe, they are what I really need. (I had switched to the 3mm tips b/c I thought one or more of the 1mm might be bent. I now know they are not, as I have switched them all into tip#1 postition and they calibrate fine.)
This is driving me crazy. I will not have any hair left by Monday night if I do not get this figured out. I am putting it to the side for now and working on a couple other small projects until Monday. Thanks in advance to all for all of your help.

[John Kingston]

Wes, I sent you an email with the pix from my PDA. They were to big to post, and reducing the quality of the poor quality pix it takes would be worthless.

[brianparker]

Wes, I'm not at work right now, but use a star probe all the time. I've always gotten all 5 tips to calibrate on the upper sphere.
First, you need to have your star square to the machine.
What I do is place a 123 block on my plate.
I have a Rayco plate so it is easy to get that square.
When I put the star in I eyeball it square to the block.

Your dual calsphere needs to be pretty square to the machine also.
Next, I calibrate each tip individually, starting with #1.
It might take a few tries, but it'll work.

Let me know if that works.
Good Luck!
John

Seems to me there would be a better way to get the star probe square, like through calibration efforts. Am I the only one that has this crazy idea, with no plausible way to execute it?

[John Kingston]

Seems to me there would be a better way to get the star probe square, like through calibration efforts. Am I the only one that has this crazy idea, with no plausible way to execute it?

Are you talking physically square or square to your alignment?
Not sure what you are asking now. You just need to be relatively square physically. Your calibration (I use DCC + DCC) will take care of the vector squaring.

Here's another thread about the method you quoted.
http://www.pcdmisforum.com/showthrea...o-your-machine

[ruidoso]

Are you talking physically square or square to your alignment?
Not sure what you are asking now. You just need to be relatively square physically. Your calibration (I use DCC + DCC) will take care of the vector squaring.

Here's another thread about the method you quoted.
http://www.pcdmisforum.com/showthrea...o-your-machine

i just hope is not another 6yrs old thread...