Other Dimensions: Elementary, My Dear Watson
June 30, 2009
The problems encountered with screw threads are legendary. The most common one occurs when the gages do what they are supposed to do: reject bad parts or accept good parts. When the gages accept parts, they are rarely called into question unless mating parts won’t assemble. However, if the gages reject parts, the flags go up and the gages become suspect because nobody makes bad parts, do they?
A process of elimination is used to resolve the problem-the same method used by detectives the world over. When all of the usual suspects have been sorted out, the one that can’t be cleared is usually the culprit. While we don’t have DNA to work with or the many wonders on the crime-scene TV shows, a little basic police work will suffice. At least we get a break in our case: unlike TV detectives, we’ve usually got more than one hour to solve our mystery.
The PartsIf the gages are rejecting the parts, measure the pitch diameter with a micrometer and wires to see if it is the guilty party. Check the major and minor diameters to see if one of them is the culprit.
When the gages accept mating parts but the parts won’t assemble, another culprit may be at work. If the component threads are longer than the gage member length, there’s a very real possibility that there is a linear pitch error in one or both. This can be detected if the go gage lengths are the same as the threads in the components.
Sometimes what seems to be impossible actually occurs. By this I mean the no-go gage will enter the part but the go gage will not. This can be caused by form errors that will not be detected by the no-go because it is truncated close to the pitch line.
A broken or chipped threading tool can produce an oversized minor diameter on external threads and an undersize major diameter on internal threads. Both can prevent the go gage from accepting the part. A measurement of pitch diameter would not detect this condition. A check on an optical comparator would.
One difficult scenario occurs when the parts fit together one way but not another. You do all the usual measurements but everything appears to be okay. Then you use an optical comparator to check the thread form and it measures 60 degrees. If you’re like most folks, you check the included angle as an isolated feature rather than in relation to the axis of the part. So you miss the fact that the 60-degree form is correct but it is tipped at an angle to the axis of the thread.
Some thread forms, such as those with a radius on their roots or crest, may appear to be okay when gaged, but the gages involved have the usual truncated form and don’t come close to verifying that the root and/or crest are within limits. This problem is particularly acute with European pipe threads because North American gagemakers frequently offer gages with truncated forms to check them. The problem is not discovered until assembly time-or worse, when they fail after assembly.
The GagesWhile the problems noted above are not as prevalent with gages due to the way they are made, many of them do occur and may require the services of a forensic specialist to find them. If the products measure up, then it’s time to check out the gages by sending them to an accredited laboratory that specializes in thread gage calibration.
Gages may be inaccurately made or accurately made to the wrong specification. If the gages are from an offshore supplier, the application of the gagemakers’ tolerances may be at odds with North American practice. Alternatively, they may be applied correctly, but the measurement technique used may not be in compliance with North American practice.
Ninety nine percent of the time, the above pool of suspects will yield the troublemaker to a good detective. If they don’t, you may have to ask Sherlock Holmes for help-if you know how to get in touch with him.