Thread measurment

[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. Is this possible with the CMM. If so what is the proper way? Thanks in advance.

[Hilton Roberts]

Best advice I can give you is not to try it. The CMM is not the proper tool for such measurements. It is not accurate enough.

If you are going to measure external threads, measure the major diameter with a standard measuring machine and the pitch diameter with the addition of thread wires and the appropriate pressure for the thread.

Your best bet is a set of plugs and rings set to master plugs to give you gages for the measurement of internal or external threads.

Minor diameters are ALWAYS a problem if you are trying to do physical measurements that need traceability.

Stay away from using the CMM for such measurements.

HTH

Hilton

[#2#]

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. Is this possible with the CMM. If so what is the proper way? Thanks in advance.

Are these internal or external threads? I'm guessing internal since you say they are caps. The CMM is not going to be very good at giving you any of the dimensions. Your best bet is to get a Go/NoGo thread plug. You can measure the minor diameter with a bore gage.

[cmmguy]

MTG Gage

Best tools for checking thread charateristics or elements

There are others but none better.

[craiger_ny]

I think it is Johnson Gage that makes some very nice thread measuring gages. They used to make a lot of thread measuring stuff for the government if I am not mistaken. The gov't used to require their suppliers to use Johnson thread measuring gages.

Craig

[cmmguy]

That was some bad info that Johnson Gage was foisting onto the industry in the early 90's. They got one of the congressmen buddies to write LAW in place that said that class 2 threads had to be checked by the Johnson Gage. But fortunately it is was thrown out. The spec is the B1 standard.

The tri-roll thread gages are equivalent to the Johnson gage vee and cone method. The problems with the vee and cone method is that the vee is fixed, which will not let you isolate thread drunkeness from the measurement - no, that was not a hijack), and the fact that you need threaded masters to set the vee and cone.

The MTG gage is the purest method to measure the thread. It uses two floating best-size balls for locating the threads on one side and a fixed ball on the other. Thread drunkeness is subtracted from the measurement automatically. The gage can be set with gage blocks or master plain cyl ring gages - not threaded ring gages.

The Johnson Gage and the Southern Gage Triroll are good gages, but the MTG gage is best.

[craiger_ny]

I did not know about the congressional thingy. I only said that because a few guys I know buy Johnson from the days when they had to when they did biz with the gov't. Someone, somewhere in the gov't was requiring it. I do not know if they have improved things since you checked out their gages but they do check for that as well. I believe it is a 3 gage type of set-up. Different gages in the system check different thread attributes as no thread can be measured completely (pitch, minor, major, and drunkeness) with a single gage. I have not purchased a Johnson gage so I can not testify for them.

[quality2214]

Thanks for the help. I thought as much, but wanted to verify through the experts here.

[Jan d.]

How big is the thread you are trying to measure? ID or OD?

Generally, a CMM is not the best way to measure any threads. Only if the thread gets big enough so that you can hit the root with a "sizeable" ball diameter. I hate 1mm tips. 1.5mm is much better. I always try to avoid going under 3mm though (but obviously I have to do that many times, just like all of us). That typically only leaves REALLY large threads on the OD.


Jan.

[Hilton Roberts]

We tried the MTG gage and others back in the late 70's or so and the problems we ran into involved traceability back to some officially recognized standard. The small ball sizes for some of the pitches really required one to accept at face value the diameters supplied by the manufacturer. We were unwilling to do that and were never able to accurately verify the actual diameters of the balls versus the nominal values.

I do not think we ever used them in Saint Louis because of that fact.

It is a lot like accepting the stamped value of a thread wire by looking at the label on the container.

In Saint Louis, we always physically measured each and every thread wire at both ends and in the middle. Our values were always different than the nominal values supplied by the manufacturer.

I would think that if the policy at your company is to accept calibration reports supplied with the things you buy, you probably will not have problems. If you question everything and run tools through your metrology lab, I think you will see a big disconnect in what is supplied and the values you determine through measurement.

There again it will boil down to what your metrology lab is able to provide you.

If you have a really good lab ( read lots of expensive stuff ) you will get measurements a lot different than what you would get in a gage room.

The more expensive and capable a lab is, the higher the order of measurements that can be made. It also follows that higher order calibrations must be made of the equipment making those measurements. It is a never ending task to make sure the calibration of your equipment is perforemd at the proper level.

For example feeler stock can be verified with a micrometer. A micrometer can be verified through gage blocks. Gage blocks can be verified through transfer standard blocks. Transfer standards are verified through reference standards. Reference standards are traceable through NIST.

That is why it is critical that you use higher order standards to validate lower order standards and it was that hierarchy that prevented us from using the ball type gages for measuring threads.

HTH

[cmmguy]

While I dont disagree about the necessecity to certify the tools you use, I always felt that the certification on some items could be accepted at face value, especially if the grade was high enough and you would audit them once in a while. Where the MTG gage really shines is on ID threads. There is no better way to check ID Pitch Diameter. Vee and Cone methods are merely transfer methods from other ID threads. Too much error in the vee and cone method when the PD deviates from the set size since the contact point is different from the master. The Vee and Cone method also does not account for lead error (drunkeness) because of the fixed nature of the vee.

Adjustable AGD stlye Thread Ring gages are just plain aweful. If you ever single element checked one you would never use one again. Fixed Rings are the only way to go, but then again how do you confirm their PD?? You cant use wires...

The problem with wires is that you are actually deforming then with the prescribed gaging pressure to "make" them conform to the lead. By using the gaging pressure, you are flattening the sides of the wires each time you use them. I can't think of another gaging practice where you exert force onto the measuring piece in an effort to deform them.

If you can find an H28, it prescibes the ideal method as using balls and that how the developement of the MTG gage came about.
it is possible to remove them and certify them.

[Farmer]

The best way to "measure" threads is on a projector.

[dwade]

I always preferred to use ITW Zero Systems Thread Comparators out or Erin, Tennessee.

[Brian Gudauskas]

Felt I need to clarify a couple of things.

It's not that the CMM is not accurate enough. It's plenty accurate. The problem is that you can't get major diameter or pitch for most thread sizes because of the size of ball versus thread pitch/size. We have made routines before that can scan for minor diameter very accurately, but they are time consuming and complex. For threads that are large enough to get a ball down the face of the thread, some (also very time consuming and complex) code can be written to capture all data. Again, it's really not very practical.

The best thing to do, is for fairly open tolerance position, you can use Auto circle with pitch, ignore the diameter (or understand that it does not mean much), and realize that you are not getting the typical accuracy you would expect.

PS: Cylindrical probes are NOT the answer here either, how do you know the probe is parallel to the hole?

[Matthew D. Hoedeman]

PS: Cylindrical probes are NOT the answer here either, how do you know the probe is parallel to the hole?

Do you know how many times I have told people to STOP using those cylinder probes for that very reason? A ball is better!