hexagon logo
Register Sign in
  • Replies 14 replies
  • Subscribers 15 subscribers
  • Views 186 views
  • Users 0 members are here
Options
Related

Automatic Probe Qualification

InspectorJester
Good morning!
Running PC-DMIS 2013 MR1 SP5 on Windows 7 x86
----------
I'm embarking on the typical quest of autocalibration routines.
This quest involves an Optiv Performance 443 Camera probe, as well as a plethora of touch probes. I'm moreso concerned with the Camera and 1 probe.

I was able to create a standard autocalibration routine for the single probe using Parameter sets.
However, being an Optiv, I have this big camera I would like to calibrate.

This is a little more complicated unforunately!
However, I am confident. Perhaps overly so?

Regarding this post: https://www.pcdmisforum.com/forum/pc-dmis-enterprise-metrology-software/pc-dmis-vision/21364-auto-qualification-issues-with-optiv-1-vision-system-with-pcdmis-2001

I am having similar problems as the individual in the aforementioned post. When I calibrate the touch probe, I am using a parameter set with the defaults and MAN+DCC. I am able to start the program and run it, automatically calibrate/qualify the touch probe. However, when I calibrate the camera, the crosshair Hit Target "EM1" to 3 does not show up. I am curious as to why?

I am unable to autocalibrate the camera because of this. To compensate I manually calibrate the camera and have an autocal program for the probes. But I'd like a 1 button solution to both if possible!

Anyone have success with autocalibrating a camera?
  • I don't know about Optiv, but I read on the forum that on this kind of cmm, you first calibrate the optical probe, and then the ttp.
    In this case, you just have to run the autocalibration in dcc mode only, answering no, the tool hadn't moved.
    Do you proceded like this ?
  • Yep!
    I know the process, the camera is the "master" probe while the next tip is the link between the ring and the sphere. I'm just trying to automate it so I can run one program and maybe take a point, then walk away and come back X amount of time later and be fully calibrated (or whatever)

    If I set it as DCC and the tool hadn't moved I can take points on the camera. But what if I need to tell it it had? How can I update the ring position automatically, or is this not feasible?

    If that is the case I would have to manually calibrate the camera and then I could DCC the tips. But that's what I do anyway!
  • I think I'm trying to overcomplicate the process.

    All I need to do is automatically define the location of the ring, and update it via the measurements of the camera. I would think that I should do this every time I calibrate, for posterity if nothing else. I thought that's the whole point of calibration, relating the camera location to the ring location, then a probe to the ring location so the probe and the camera know where they are in relation to one another.

    If the camera doesn't know where the ring is, how could it accurately calibrate? And if I can only manually find the location of the ring using the camera outside of an autocalibration routine, then having the camera in an autocalibration at all would be pointless, because I would have to initally define its location manually.

    Right?
    Someone help me derail this mess
  • I'm not sure to understand exactly (don't forget where I come from ! LOL !)
    Do you move the ring or not ?
    With the cmmv, I can create a parameter set dcc with the ring, and the auto cal works find.

    Another way is buying a matte (dull ?) sphere, then calibrate the optic and the ttp on it, it should work (measure a point at the top, the a circle at the center level, create a constructed sphere and call "single tip calibration with the name of the sphere)
  • in reply to InspectorJester
    As far as I understand the process, I am to use the camera saying YES the ring has moved. Then measure it with a probe saying NO it hasnt. Then measure the shere with the same probe, saying YES the sphere has moved, then calibrate all other probes on the sphere
    This is why I would like to say yes the ring has moved. Ideally I would like to take 1 point and then calibrate all the probes
  • Thanks!
    Basically I would like to take 1 point and calibrate every tip. Ideally I would take 1 point on the ring with a probe, define a general location of the ring for the camera to find, then DCC everything else.

    The ring moves very little, but enough for it to sometimes work and sometimes not with DCC. Thus I would like to just manually define its location. The camera has a hard time finding the ring, but the probes don't have trouble finding the sphere. If I manually locate the camera everything will be seamless, but I can't manually locate it using AutoCalibrate probe
  • in reply to JEFMAN
    This is great!
    I'd like to use custom-defined objects to calibrate everything. I feel like it would just be easier for me. Then I could make a whole custom mount with a small calibration station permanently mounted to the CMM.
    Your post will help me immensely
  • Normally, when I manually calibrate (using Man+DCC!) the camera, it starts the routine where I am tasked with taking 3 points on the ring (hit target "EM1" through 3). I move the camera to the ring, drag the crosshair over and take my point. (3x) From there, it automatically focuses, takes points, etc. and calibrates.

    HOWEVER, when I try to do this with the AutoCalibrate feature, I don't get to drag the crosshair. I get the prompt to take hit target "EM1" on the ring. I can move over to that point, but there's no crosshair for me to drag!
  • in reply to InspectorJester
    Basically I can't "manually calibrate or locate" automatically
© 2024 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement