hexagon logo
Register Sign in
  • Replies 6 replies
  • Subscribers 15 subscribers
  • Views 103 views
  • Users 0 members are here
Options
Related

Finding spline issue

WM321
Hello.

Using Quindos 7 with gear pack. I am trying to fix a recurring issue with an existing program.

The part is a cylinder with a flange at one end, the flange is comprised of four pads, each with a hole through. The cylinder has an internal 39 tooth spline shaped through it. The pitch diameter of the spline is datum A. The surface on each pad of the flange is datum B. The program then uses a rotation between one of the holes and datum A. All these features, plus the pitch of the spline, are probed with the single 1mm stylus with no issue.

The program then pulls in a 4 way stylus with 4 x Ø1mm tips, each at 90° which each have a very short stem, approx. 4mm. This is used to check the profile and helix. This tip will now crash at the gear command, even if I have find activated. When attempting to find the gear, it will generally take 1 point, then take an unusual path, skipping past 4 or 5 teeth and crashing into the spline in a diagonal path. I have never had any real issues with the find gear before this. It is using the same alignment and same SETCMPAR as the pitch gear command. The machine has been qualified.

Is there anything I could be missing here?

Also, I was attempting to scan the spline to find a valley, then rotate to it to try running the gear command with the new alignment to figure if there is something suspect with the probe configuration / parameters etc. I used ME2DE to scan a segment of the spline, however I'm not sure how to extract points from the scan. I would like to extract the point with the highest X value, this would equal the valley, then use COLPTS to create a line between -A- and the extracted point. If everything is where it is meant to be, the gear command should run in this alignment.

How do I extract the point with the highest value in X from the scan?

Thanks.
  • I know nothing about gear, but your last question I can handle 

    ASSIGN/V1==MAXINDEX(SCN1.HIT[1..SCN1.NUMHITS].X)
LARGEST_X  =GENERIC/POINT,DEPENDENT,CARTESIAN,$
            NOM/XYZ,<SCN1.HIT[V1].TX,SCN1.HIT[V1].TY,SCN1.HIT[V1].TZ>,$
            MEAS/XYZ,<SCN1.HIT[V1].X,SCN1.HIT[V1].Y,SCN1.HIT[V1].Z>,$
            NOM/IJK,<0,0,1>,$
            MEAS/IJK,<0,0,1>


    Also, a stab at the probe thing - perhaps the software is seeing the active tip (Say T3A0B0) as a straight down tip with no obstruction?
  • in reply to RandomJerk
    I wish it was this easy. This is pretty much what I am looking to do, only I need the appropriate Quindos commands to do it.
  • in reply to RandomJerk
    My bad, you said Quindos, I missed that.
  • You can use FXTREM (find extreme points). Select the first option under "type of evaluation" and this will give you 6 result elements - minimum in X, maximum in X, minimum in Y, maximum in Y, minimum in Z & maximum in Z. (command would look as shown below) 

    FXTREM          (NAM=(XMIN,XMAX,YMIN,YMAX,ZMIN,ZMAX), OBJ=ME2DE(1), TYP=CSY, MOD=NOE)
  • Thanks for the tip Neil, this should come in handy.

    However the fix I had to apply to this was some what of a work around. I had to orientate the spline tooth closest to the centre-line of the flanges to the fixture, then set the gear command to manual in container 5, run the program and manually probe the spline. Once run through I set the gear command to automatic in container 5, and as long as the operator has set the part / spline as per the work instruction, it runs sweet every time.
  • the 'issue' is the usage of only a 4-star and Quindos automatically selecting the appropriate styli to use. Not sure what version of Quindos/GEAR you are using, but there is an input in GEAR's 'Configuration' dialog where you can enter values for the 'min. angle between probe and surface', this will alter when Quindos decides to change probes due to the relative angle of the probe to the tooth surface.
© 2024 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement