Recent discussions with a customer over problems with threaded parts reminded me just how complicated some measurements can become. Problems arise because the many different ways they are inspected don’t always come up with the same results. Many of the methods used are not actually checking the same features in the same way and that makes a comparison of results difficult. 

The functional size of a thread will vary due to changes in the form, linear pitch, and overall geometry. If the methods used don’t account for all of these elements as a unit, those that are missed mean one system will accept parts while another may reject them. Despite this situation, it could turn out that the parts will assemble anyway just to make life more interesting.

If the goal is to ensure mating parts will assemble, it’s hard to beat fixed limit or go/no go gages as they represent theoretically perfect mating threads at their respective limits. The beauty of these gages is that they are so simple even I can use them. When taking direct measurements, however, things can get a bit messy.

Measuring the pitch diameter of a thread is an example of what I mean. If you use thread wires and a micrometer to do so, you’ll notice that the wires sit in the vee-groove of the thread form or if you prefer, the space between the threads. If you use a pitch micrometer, the pointed anvil does the same thing but the vee-groove anvil locates on the thread rather than in it. This means you’ll get different numbers. If the pitch mike anvils are fixed in place, which is the case for some instruments used to cover a limited number of thread pitches per instrument, the results will not be close to readings from a mike and wires. If the pitch mike contacts are not properly cleared, they won’t contact on the pitch line but over the surface of the thread flank. And that means different readings again. The same effect will occur if the thread angle is incorrect.

Products produced by a rolling process can result in threads that measure okay but are out of round or lobed, a condition that neither of these methods will detect. But if this roundness condition takes the functional size out of tolerance, the product won’t pass fixed limit gages.

If the thread has a linear pitch problem—not an uncommon byproduct of thread rolling—direct measurement of pitch diameter won’t detect it, but fixed limit gages will reject it if the functional size goes out of limit.

Thread comparators, typically with three rolls on them, are often used to obtain direct pitch diameter readings but if the wrong rolls are used, you’ll be no further ahead in sorting out problems. Single rib rolls designed to contact on the pitch line will not react to linear pitch errors while multi-rib or functional rolls will not give a true simple pitch diameter reading because of their form.

Optical comparators are essential for any facility that has to inspect threads. Thread form is the obvious reason for this but linear pitch on threaded product can also be checked using these versatile devices. Some threaded products have poor surface finish on the form and that will have a direct effect on their overall accuracy. Since they do not touch the thread to make the measurements, the readings obtained will usually vary from those devices that do. This applies to automatic machines used for thread inspection that use lasers as the sensing system.

 You can be forgiven for wondering if there is any hope that different methods for gaging threads will ever deliver comparable results. When questions arise, it is worthwhile to check exactly what a certain method is actually measuring—the whole thread or only certain parts of it. Alternatively, determine what is not being checked by that method when the results are not satisfactory. What’s missing in one check could be the reason for rejection by another.

 One system of thread inspection that has withstood the test of time, is comparatively low cost and easy to use, is the aforementioned fixed limit gages. It is assumed, of course, that said gages are properly made and calibrated but then that applies to whatever you use. Gage calibration can be a frustrating exercise in its own right but I’ll leave that subject for another day.