hexagon logo

Long probe tip

We are cutting a cone inside an 8" tall cylinder. We do have a callout for the smaller ID (approx .38") , the larger is ref (approx .55") controlled be .0005" flatness, angle is 2° ±.005°, & .0005" circularity. Customer is using a custom tapered gage (I know, don't get me started Astonished ... 'not traceable, how much force is used etc. we are stuck & may tell them we cannot work that way BUT after some parts were rejected we wanted to see what we were getting so split a scrapped part lengthwise & measured that way - there is definitely a belly. Company wanted to run some tests so 4 x 100 mm carbon fiber probe with a 100mm ceramic extension - ONLY using A0B0, part is held vertically in a V block(TP 20 on Hexagon Global 575)(. I lowered movespeed to slow enough it doesn't fault out from the long probe momentum & touchspeed is at .4mm . Getting a 'repeatable' .0003-.0004" calibration with this (basically using the majority of our tolerance without toughing the part yet). Taking circles at certain heights where we know what diameter to check angle & circularity should be & for straightness taking 70 hits (over 8") per vertical line - 4 lines at 0,90,180 & 270 degrees (top front is marked by operators in machine) to check straightness & angle. Is there a better way without cutting open a part or sample ??? Not a big customer so they aren't planning to invest in different probe head etc a few hundred okay, but not worth a few thousandConfused
Parents
  • Matt,
    Actually came out better than I expected, made a program to check ring gage ID with varying number of hits (nothing more than .00012" different than gage actual) Astonished


    Wow, that's quite good.

    I have used longer probe builds than Renishaw suggests, but I have never tried pushing it to that extreme.​

    How did the roundness check when mearing that ring gage with that 200mm tip build up? Did you see much lobing error?

    I'm just asking out of curiosity, it should only matter to you if you need to report any sort of form measurements (Circularity, runout, profile, that sort of thing). Your application may not require it.
Reply
  • Matt,
    Actually came out better than I expected, made a program to check ring gage ID with varying number of hits (nothing more than .00012" different than gage actual) Astonished


    Wow, that's quite good.

    I have used longer probe builds than Renishaw suggests, but I have never tried pushing it to that extreme.​

    How did the roundness check when mearing that ring gage with that 200mm tip build up? Did you see much lobing error?

    I'm just asking out of curiosity, it should only matter to you if you need to report any sort of form measurements (Circularity, runout, profile, that sort of thing). Your application may not require it.
Children
No Data