measuring thread depth for blind hole

[dog241]

Hi,


Q1) Our QC discovered faulty parts with a threaded hole of M12 x 1.25 that should be 18mm deep, but the mating part bottomed out. When this issue was brought to the attention of our supplier, they insisted that the parts should indeed be threaded to a depth of 18mm with a 1mm relief. They also questioned the method of measurement, as we measured the thread depth based on the insertion depth of the mating part, resulting in a measurement of only 16.5mm.

Q2) Additionally, for the same part, we occasionally encounter instances where the parts are out of concentricity by 0.22mm, while the drawing specifies a tolerance of 0.02mm. The supplier claims that the measurement is taken before threading M12x1.25, which measures between an outer diameter of 15.7mm and the blind hole. They argue that after threading, they cannot obtain accurate readings due to variations in thread clearance on both parts.

What is your opinion?

 

[btrueblood]

Q1 - did your inspector verify there was no debris in the hole or the threads before running the mating part in?

Q2 - Concentricity on a tapped hole to less than .001" is pretty gutsy. 0.010 is more typical, though I know shops that reliably get better than .005. Thread milling might get closer. It all then boils down to what is used to find the hole position - thread pitch diameter? Thread major or minor diameter? Sticking an off-the-shelf bolt in the hole and measuring from the protruding end creates a lot of issues (did you specify a projected tolerance?).

 

[dogbural]

Yeah, there was no debris. It appears the thread was not complete. We ended up cut it open in half and confirmed this.

No, drawing does not specify the projected tolerance.

 

[AlanBatty]

Hi,
Can't speak for all but I usually treat hole depth and thread depth separately (if both are specified). Out of habit I check hole depth first (using gauge pins) then thread depth with a plug gauge. It is very rare to use a mating part to check depth and sounds very unreliable unless you know got sure the dimensions of it. Going by the information given, this is probably why the issue occurred in the first place.
Measuring concentricity/runout in a threaded hole screams errors to me and is a good example of adding a pinch of salt to the value obtained. I don't think I have ever managed to achieve the required value (using a CMM). What I would probably do is measure a feature machined at the same time and as close to it, if indeed possible.
Alan

 

[geesaman.d]

Tapped hole location requires a Lavezzi Tru-Pos or similar.

Tapped hole depth should be assessed using a thread go-gauge.

Unless an inspection procedure was provided to support their (janky sounding) method of inspection/criteria, they're out of line.

 

[Rank Li]

If the drawing specifies a tolerance of 0.02mm,the actual is 0.22mm,then it's really hard to accept