Questions about accuracy measuring a threaded hole diameter CMM versus tri-mic!

  • Filter
  • Time
  • Show
Clear All
new posts

  • Questions about accuracy measuring a threaded hole diameter CMM versus tri-mic!

    Okay, this is quickly becoming a hotly contested issue in our shop. We are making this aerospace part that has 2 threaded ID bores that are quite large. The tolerances on the part are all wonky because they are using normal dimensioning AND total positioning, and a whole slew of other headache-inducing stuff that has made this part something nice to cut my teeth on. These threaded ID bore locations and size have to be fairly accurate. We have a go/no go gauge out by the mill that the operator is using to check parts that are coming off. Still, the print is complicated and I have to check things on it (there is also a larger counterbore above the threaded portion). The threaded portion is fairly deep and wide, so lots of room to probe and lots of points that can be taken easily. The diameter that my machinist wants to keep is 2.418 from thread tip to thread tip. He insists that the tri-mic is reading the threaded portion at 2.418 almost exactly, maybe 2.4181 or 2.4182. My CMM is getting very repeatable and accurate results in my opinion. I'm using a 4mm ruby tip, and I do select the pitch option so that the circle is measured along the curve of the threads. I'm taking about 40 hits all the way around 360 degrees up the threads. NinjaBadger, I read a thread from back in 2015, and you are where I get the idea from (management does not want to buy the location gauges at this time either and I can't really blame them seeing the price). I made it measure 8 different levels of the thread as 8 different circles on pitch. I could have gone down even further, the threads are deep. The CMM inspector from the shop that we sent them to said they are in tolerance, but slightly out of circularity, which I would expect when a circle is milled and then threadmilled instead of lathed and the threads cut on a lathe. I am def picking up on the circularity issues too, PC-DMIS is telling me that the circle has about .002" out of circle-ness. The number that I am picking up on my CMM, is anywhere from 2.4200 to 2.4204 depending on which thread I want to analyze. That number (2.4200 and larger) is technically out of spec even though the thread gauge says the Go goes and the no-go doesn't.... Our customers quality guys say it's okay. I'm a bit worried, and my machinist is worried about the numbers our cmm put out, but insists that the tri-mic is right anyways. I'm basically in CYA mode right now, and since I'm new at this, I want to know that our part is right. I'm guessing that this is either down to the circularity, possible probe deflection....chatter in the threads? The data I got back was: (Diameter of threaded bore going DOWN starting at the top) 2.4200, 2.4199, 2.4200, 2.4205, 2.4208, 2.4208, 2.4208, 2.4207. I can probably sit here all day and look closer in the threads, and there probably is some trash in them, I'll be honest, maybe some dust. But the average I'm getting is around 2.420, a hair above it really. Which, is not the 2.418 that my machinist is getting. I know it's only .002 difference. I know. And maybe some of you will say that is just going to be the battle between the CMM operator and the machinist until the end of time, idk. I've heard that this is where things can get crazy in a shop. I just want to know what the correct measurement is, not who gets a golden star. My data is just very similar and repeatable. And his tri-mic does read the 2.418 fairly clearly (it has a clicker thimble). There is no way I'm forcing that tri-mic .002 bigger in an honest manner. THANKS IN ADVANCE x100 FOR ANY HELP!!!
    What are you doing using your big school words? Just use normal people words and I'll understand what you're talking aboot.- Ricky LaFluer

  • #2
    Measuring threads with a CMM is a fun time, as I think you've picked up on elsewhere in the forum. First off, what diameter are you in fact measuring? The tri-mike almost certainly has three cylindrical surfaces it's pushing against innermost surfaces of the threads - it will give you the minor diameter and nothing else. The CMM may give you the minor diameter, but only if the ball is landing on the thread peaks. If it's falling down between the thread peaks, you may still get an accurate-ish position if you have the pitch setting set correctly, but the diameter will be meaningless.

    I'm not familiar with the NinjaBadger thread you mention - perhaps the way your multiple measurements are spaced out one of them is definitely on the minor. But I personally wouldn't count on it unless you get in there with an eye loop and look at it.

    Your threads are probably large enough you could get it lined up fairly well. Most of my threads are pretty small, and they're one of the few features where I will readily side with hard gauging over the CMM results. There's just too much inconsistency for it to be fully trusted.

    Also, the thread gauge results are totally different, since the thread gauge contacts the actual 60-degree angled surfaces of the thread. The tri-mike's not doing that, and neither (almost certainly) is your CMM ball. One could be OOT small while the other could be OOT big. Basically, the thread gauge will tell you what the D-comp on the thread mill needs to be, while the tri-mike and CMM (at best) will tell you what the D-comp of the end mill that cuts the hole needs to be.


    • UKCMM
      UKCMM commented
      Editing a comment
      Not sure why you think it weird that a Minor diameter gauge checks the part good against using the CMM and guessing the hits are exactly on the thread crest. A CMM check of an internal thread will almost always report big unless you have a large flat crest to aim for if you have a full form crest then forget using the CMM it can be done but takes way to much time.

    • steelweaver120
      steelweaver120 commented
      Editing a comment
      UKCMM, I don't think it's weird, I do trust the gauge over the CMM...I just want it to be closer I suppose? The hole is quite large to where I can visually look in the hole and see where the probe is taking points. I just figured it would be nice to have another trick in my bag. I actually did not know that a CMM would probe a minor diameter as big! Thank you! ^_^ This is the kind of info they don't teach you in school. I'm loving this forum.

    • mbatten
      mbatten commented
      Editing a comment
      When you say "Thread diameter tolerance" are you 100% sure you're talking about the minor diameter? That's critical to what you're trying to do.

      Also, if you spin the tri-mic to move the contact points around, do your size results change? A severe out-of-round condition could easily explain a few thousandths disagreement. The tri-mic number will only be based on three points, while the CMM will average over however many points you take - possibly dozens.

  • #3
    Try checking the diameter with a 2pt bore gage and compare it with the 3pt. part could be egg shaped or tri-lobe shape.
    Time for the Trolls to leave.


    • steelweaver120
      steelweaver120 commented
      Editing a comment
      We have def confirmed the presence of egg-shapedness/circularity problems! That's what makes this even more maddening! The CMM and tri-mic readings are super close (within .002), but I know for a fact that those holes being measured with a tri-mic are out of round so there easily could be something off. The CMM readings just seemed big to me.

  • #4
    Use the thread gage to buy off the threads and associated dimensions.

    Use the CMM to report location of the hole.


    • steelweaver120
      steelweaver120 commented
      Editing a comment
      This is likely what we are going to do. I'm just trying to wrap my head around this/get as close as humanely possible. I would just like to have some tricks hidden in my sleeves for the tough times. It always seems like my shop has that one part that just really throws as many wrenches in my path as possible.

  • #5
    A CMM is a poor choice for measuring threaded hole diameters, and a fool's errand for trying to measure things like pitch diameter. Using thread following, you are not guaranteed as to where on the thread profile you will land, only that you should land on essentially the same part of the profile on all hits, making your circular center more accurate. A minor diameter for an internal thread can be semi-accurate if you take enough hits using a barrel/shank probe.

    What diameters PRECISELY are you trying to get?
    I'm upping my standards........

    Up yours!


    • BKulpa
      BKulpa commented
      Editing a comment
      As I've said in the past the barrel probes don't calibrate correctly. I've tested it in the past, and interestingly enough I'm going to do it again today.

  • #6
    measuring thread diameter on a cmm doesn't work very well

    measuring thread position on a cmm works well (within 0.002"TP) if:
    -threads are FOD/burr free
    -using correct pitch value for your hit path

    use gages for the thread sizes and you'll be good


    • #7
      steelweaver120 When I have to verify size or position of the minor diameter of the thread on the CMM (some of our customers specifically ask for that, go figure...) I use a large diameter probe let's say 3mm and run several(6-8 depending on diameter) line scans parallel to the centerline of the thread, then construct a MAX_INSC BF Recomp cylinder.
      Another option is to use Auto Feature Cylinder MAX_INSC and choose Adaptive Cylinder Line Scan.
      Just to emphasize, these only work for the minor diameter which can be way different from the actual thread position, esp if a shitty drill was used to create the hole.



      • #8
        Max_insc and a barrel probe if you're normal to the threaded hole may get you closer to your target if that's what you're trying to do, but like everyone else, I woudn't use a cmm. But then again, the treads I'm usually measuring I can barely see by eye. Sounds like you have massive thread to work with. Are the minor diameter edges fairly wide? Even when I have seen large threads that had big flat areas to measure, they looked a little dished.

        Anyway, I was just wondering what you ever figured out with this situation.


        Related Topics


        • Ravenmaniac
          Helicoil threads....
          by Ravenmaniac
          Good afternoon....again I'm dipping into the best knowledge pool I have for a question that is not exactly Pcdmis related but will always promote a great...
          03-28-2007, 08:03 PM
        • quality2214
          Thread measurment
          by quality2214
          Here is the deal. I am going to have to measure some minlon threaded caps. They want the Major, Minor, Pitch Diameter and also the pitch of the threads....
          09-18-2006, 01:04 PM
        • FAMOUSX1
          TP of threaded diameter
          by FAMOUSX1
          Hello All,
          I have been reading up on the correct way to measure a threaded diameter but I am still unclear on what the correct way is to do this....
          04-12-2016, 05:37 PM
        • B. Jacobs
          by B. Jacobs
          I hope this is a simple question.
          I am measuring the diamater of an OD thread. I have the auto circle selected for this measurement and also the...
          07-25-2006, 12:44 PM
        • Darrollh
          Measuring Threads on the CMM...
          by Darrollh
          Good evening,
          What is the best way to measure threads on the CMM? I would just use a thread gage and put it in the report but I was asked to write...
          01-17-2019, 12:08 AM