irregular wear on diameter of a seal groove

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • irregular wear on diameter of a seal groove

    Picture attached is the wear on a seal groove, I'm writing programs that will check used parts and give me the smallest diameter of the wear.
    Depending on the number of hours in service, severity of service etc, the location from the groove wall will vary a fair bit from the groove wall so taking a circle of hits at any axial location is not an option.

    Touch probing only here, what would be your approach to checking this? I'm already settled on a measurement plan but thought I'd see if any better ideas pop up, I'm not too far along to change gears if for the better.

    I have a 1.5x20mm tip that can hit it no problem and the wear radius should not really be smaller than tip radius... but how would you construct it?



    wear.JPG

  • #2
    Optical comparator would make short work of it.

    Comment


    • Douglas
      Douglas commented
      Editing a comment
      That is what I'm using to check how my CMM result correlates to it. Thr part is alreadyvon the CMM for other sizes so I'm adding it

  • #3
    Just by looking at the geometry...if I were writing the insp plan for this part, knowing that the wear will be irregular (and therefore a PITA to program), I wouldn't put this dimension on the CMM.

    Would suggest making an inspection mylar with dotted profile lines and throwing it up on an optical comparitor. Check it in 0.2 seconds/part instead of messing around with hit algorithms and filtering and all of that jazz. I understand the part is already on the CMM and it would be cool to add this to your inspection plan, but sometimes trying to get a certain dimension on the CMM is more trouble than its worth!
    Last edited by DAN_M; 07-25-2020, 09:33 AM.

    Comment


    • DAN_M
      DAN_M commented
      Editing a comment
      I am positive, in all definable matrices, that my industry is just as (if not more) stringent and annoying than yours. We deliver top notch customer service, as well, via the utilization of programs that I don't have to constantly fiddle around with. That being said, when the CMM isn't the correct tool to accurately and efficiently measure a part, we realize this and don't dedicate expensive programming resources to the issue. I am sure you CAN figure this out, I am just politely and respectfully trying to get you to realize that our of respect for your company's bottom line, maybe go the route that is cheaper, takes 90% less time, and is so easy that you could train your janitor or CEO (whoever is smarter) to employ this measurement method in about 5 minutes.

      All of the above being said (and hopefully well received as was intended)...
      Let's say all of your tolerances on that area are ± 0.005". You just need to eliminate the solidity of the CAD, get a solid line for the nominal profile, and create offset lines from nominal. (one being 0.0025" above nominal and one being 0.0025" below nominal). Blow it up to the scale you'll be measuring on (usually 20X or 50X). Print this out on that clear/see-through mylar paper using a laser printer. Then, slap the part on on the optical comparitor with the mylar on the screen and see if the shadow of the part fits within the lines. Using this method, you could line up a whole bunch of parts and bang them out in minutes as opposed to painstakingly editing points for EVERY SINGLE PART. [email protected] that would make me go bonkers lol.

      If you're telling your boss that the CMM is the best way to go because you like your overtime, I promise I won't tell!! I respect your hustle! lol
      Last edited by DAN_M; 07-28-2020, 08:21 AM.

    • Douglas
      Douglas commented
      Editing a comment
      I am basically in the process of re-writing a set of programs that I inherited... to be polite I will say 'simple' programs(some do not even have a primary alignment... no not at all not even one they gotta go), and adding more to the CMM inspection to streamline and replace the old with proper programs that include manual alignments and proper part alignments... currently about halfway through implementation. In many cases I am able to write a program that replaces several old ones... this is something we have definitely decided to roll out in the slow pandemic times... better programs could mean one less inspector, and much better data to our customer, plus I have extra time for improvements to that process right now because covid19

      First we nailed down programs for new parts coming in the door pretty good... now I am moving on to what is used in a rebuild to check used parts... used parts are actually quite a challenge it turns out, one part I did an alignment in 3 stages just because I could not trust what was worn and not, had to pre-align to get at some internal features with less wear to level the final alignment or who knows where the hits could land..

      this is what I ended up with as a feature for that cylinder(V5 is assigned to be the Z value of a hit on the groove wall)... might stagger the z-heights by a thou on 2 of the tips yet but the result is very closely mimicking the comparator... close to the uncertainty of the comparator

      Code:
      OD_POLYPAK_TOP=FEAT/CYLINDER,CARTESIAN,OUT,MAX_INSC
                  THEO/<0,0,V5-.069>,<0,0,-1>,4.875,0.048
                  ACTL/<-0.0003,-0.0002,-0.7994>,<0.0062429,-0.0002092,-0.9999805>,4.876,0.0784
                  MEAS/CYLINDER,68
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.045>,<1,-0.000001,0>,<2.4387,-0.0005,-0.7758>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.048>,<1,-0.000001,0>,<2.4389,-0.0004,-0.7771>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.051>,<1,-0.000001,0>,<2.4387,-0.0003,-0.7813>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.054>,<1,-0.000001,0>,<2.4385,-0.0003,-0.7844>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.057>,<1,-0.000001,0>,<2.4387,-0.0002,-0.7875>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.060>,<1,-0.000001,0>,<2.4387,-0.0002,-0.7904>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.063>,<1,-0.000001,0>,<2.4386,-0.0002,-0.7933>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.066>,<1,-0.000001,0>,<2.4386,-0.0002,-0.7964>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.069>,<1,-0.000001,0>,<2.4386,-0.0002,-0.7995>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.072>,<1,-0.000001,0>,<2.4388,-0.0002,-0.8025>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.075>,<1,-0.000001,0>,<2.4387,-0.0002,-0.8053>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.078>,<1,-0.000001,0>,<2.4385,-0.0002,-0.8084>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.081>,<1,-0.000001,0>,<2.4384,-0.0002,-0.8114>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.084>,<1,-0.000001,0>,<2.4382,-0.0002,-0.8146>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.087>,<1,-0.000001,0>,<2.438,-0.0002,-0.8174>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.090>,<1,-0.000001,0>,<2.4381,-0.0002,-0.8203>,USE THEO=YES
                  HIT/BASIC,NORMAL,<2.4375,0,V5-.093>,<1,-0.000001,0>,<2.4381,-0.0002,-0.8234>,USE THEO=YES
                  MOVE/POINT,NORMAL,<2.5000,3.5000,(V5+V6)/2>
                  TIP/T5A0B0, SHANKIJK=0, 1, -0.0008, ANGLE=0.0937
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.045>,<0.000001,1,0>,<0.0004,2.4385,-0.7757>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.048>,<0.000001,1,0>,<0.0002,2.4385,-0.7764>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.051>,<0.000001,1,0>,<0.0002,2.4386,-0.7811>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.054>,<0.000001,1,0>,<0.0002,2.4386,-0.7843>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.057>,<0.000001,1,0>,<0.0001,2.4384,-0.7872>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.060>,<0.000001,1,0>,<0.0001,2.4383,-0.7902>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.063>,<0.000001,1,0>,<0.0001,2.4383,-0.7932>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.066>,<0.000001,1,0>,<0.0001,2.4382,-0.7963>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.069>,<0.000001,1,0>,<0.0001,2.4384,-0.7993>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.072>,<0.000001,1,0>,<0.0001,2.4384,-0.8022>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.075>,<0.000001,1,0>,<0.0001,2.4383,-0.8053>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.078>,<0.000001,1,0>,<0.0001,2.4381,-0.8084>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.081>,<0.000001,1,0>,<0.0001,2.4382,-0.8113>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.084>,<0.000001,1,0>,<0.0001,2.4381,-0.8142>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.087>,<0.000001,1,0>,<0.0001,2.4381,-0.8173>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.090>,<0.000001,1,0>,<0.0001,2.438,-0.8203>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,2.4375,V5-.093>,<0.000001,1,0>,<0.0001,2.4378,-0.8234>,USE THEO=YES
                  MOVE/POINT,NORMAL,<-3.5000,2.5000,(V5+V6)/2>
                  TIP/T2A0B0, SHANKIJK=-1, 0, 0.0016, ANGLE=0.047
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.045>,<-1,0.000001,0>,<-2.439,0.0006,-0.7754>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.048>,<-1,0.000001,0>,<-2.4391,0.0002,-0.7776>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.051>,<-1,0.000001,0>,<-2.4389,0.0001,-0.781>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.054>,<-1,0.000001,0>,<-2.4391,0.0001,-0.784>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.057>,<-1,0.000001,0>,<-2.439,0.0001,-0.787>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.060>,<-1,0.000001,0>,<-2.4387,0,-0.7901>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.063>,<-1,0.000001,0>,<-2.4389,0,-0.793>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.066>,<-1,0.000001,0>,<-2.4388,0,-0.796>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.069>,<-1,0.000001,0>,<-2.4388,0,-0.799>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.072>,<-1,0.000001,0>,<-2.4387,0,-0.8021>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.075>,<-1,0.000001,0>,<-2.4387,0,-0.8051>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.078>,<-1,0.000001,0>,<-2.4387,0,-0.808>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.081>,<-1,0.000001,0>,<-2.4385,0,-0.8111>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.084>,<-1,0.000001,0>,<-2.4386,0,-0.8142>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.087>,<-1,0.000001,0>,<-2.4384,0,-0.8172>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.090>,<-1,0.000001,0>,<-2.4383,0,-0.8201>,USE THEO=YES
                  HIT/BASIC,NORMAL,<-2.4375,0,V5-.093>,<-1,0.000001,0>,<-2.4382,0,-0.823>,USE THEO=YES
                  MOVE/POINT,NORMAL,<-2.5000,-3.5000,(V5+V6)/2>
                  TIP/T3A0B0, SHANKIJK=0, -1, 0.0008, ANGLE=-0.0937
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.045>,<-0.000001,-1,0>,<-0.0004,-2.4388,-0.7757>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.048>,<-0.000001,-1,0>,<-0.0002,-2.4387,-0.777>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.051>,<-0.000001,-1,0>,<-0.0001,-2.4388,-0.7812>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.054>,<-0.000001,-1,0>,<-0.0001,-2.4388,-0.7842>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.057>,<-0.000001,-1,0>,<0,-2.4387,-0.7873>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.060>,<-0.000001,-1,0>,<0,-2.4386,-0.7902>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.063>,<-0.000001,-1,0>,<0,-2.4385,-0.7934>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.066>,<-0.000001,-1,0>,<0,-2.4386,-0.7963>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.069>,<-0.000001,-1,0>,<0,-2.4386,-0.7993>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.072>,<-0.000001,-1,0>,<0,-2.4386,-0.8023>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.075>,<-0.000001,-1,0>,<0,-2.4385,-0.8054>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.078>,<-0.000001,-1,0>,<0,-2.4385,-0.8084>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.081>,<-0.000001,-1,0>,<0,-2.4384,-0.8113>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.084>,<-0.000001,-1,0>,<0,-2.4385,-0.8142>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.087>,<-0.000001,-1,0>,<0,-2.4384,-0.8172>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.090>,<-0.000001,-1,0>,<0,-2.4382,-0.8203>,USE THEO=YES
                  HIT/BASIC,NORMAL,<0,-2.4375,V5-.093>,<-0.000001,-1,0>,<0,-2.4382,-0.8234>,USE THEO=YES
                  ENDMEAS/

    • Douglas
      Douglas commented
      Editing a comment
      I also was not trying to include anything about other industries,i just mean our closr competitors... we are the sort of shop that CMMs are not really found in and it gives us the edge, any comparison I made was meant for just our competitors sorry if I was unclear that way. Our competitors use visual and maybe some manual inspection on a rebuild... we have CMM inspection to truly hold engineering wear allowance

  • #4
    By irregular, I assume you mean the depth or height are not consistent. I also assume the wear to be equal around the diameter at any given height. If this is not true, then the CMM isn't the best tool.

    If it is true, then just measure a series of points from along the diameter from top to bottom(or bottom to top). Then assign the X values of each to an array. Finally construct an offset point with the min value.
    Code:
    ASSIGN/V1=ARRAY(PNT1.X,PNT2.X,PNT3.X)
    PNT_LOW_POINT=FEAT/POINT,CARTESIAN,NO
    THEO/<7.3919,0,0>,<0,0,1>
    ACTL/<7.3919,0,0>,<0,0,1>
    CONSTR/POINT,OFFSET,ORIGIN,MIN(V1),0,0
    PC-DMIS 2016.0 SP8

    Jeff

    Comment


    • Douglas
      Douglas commented
      Editing a comment
      Yeah, as wear starts it is like that but as it progresses with more hours in service the maximum wear occurs further from the shoulder... I ended up with a series of points every .002 at every 90°, 4 places to construct a cylinder, minimum inscribed fit is giving me very good results... the IJK of the result is true enough to give me confidence also.
      Anything I tried to do on just one side of the part seemed sketchy as I monitored results...

Related Topics

Collapse

Working...
X