six sigma and pipe thread metrology

Got a call from Singapore this week. Problem with a patented acoustic lubrication device we make.
The bad comnponent is 1/8 in stainless npt pipe nipple. We had one left in stock and luckly it also was defective similar to the shipped component.
here are pics--

https://picasaweb.google.com/pulsenpal/PIPENIPPLE?authuser=0&feat=directlink



So the vendor supplying the pipe nipple has one web page extolling their goodness.
"Quality Is What Sets Us Apart"
Really, it is user beware for almost every manufactured commodity. And inspecting the threads on the pipe
fitting revealed 0.003 in first full thread over allowable with crest thickness twice allowable. The nipple would only twist in by hand 1/2 turn instead of the usual 3 to 4 turns.
here is one website giving dimensions of thread gauges.


http://www.pmclonestar.com/pdfs/BasicDimensions.pdf

jh

Threads are a pain, pipe threads are a worse pain - usually create a lot of arguments between suppliers and customers. But, it sounds like they need to have a first piece, last piece inspection at every tool change to track down this problem. Even if they used a thread ring gage, they should have detected that. Did you ask them for a corrective action?

Bob

I was informed of the issue one day ago. Our office is set up for electronics assembly but limited inspection instruments.
So I took pics and dimensions in my lab for the boss to use in confronting part vendor.
Vendor of this part is one of the largest hydraulic fitting manufacturers. We purchase in bags of 50.
As you may have seen in pics, one end of fitting is satisfactory while the other is unacceptable.
I am looking forward to the vendor's response.

jh

Did you gauge the female part? Did you use verify with a thread ring gauge? I'm not so sure about using a thread mic on a tapered thread.

Is this NPT or NPTF?