Capability or Gage R&R Study for Leak Tester?

[SeanKarson]

I have a really hard trouble making an analysis of my leak tester

We put the pipe plugged into the device (leak tester)with the opposite side blocked, then 150 psi air pressure fill the interior of pipe in order to verify any possible leak.
Once this device has completed the test, It displays a leak value which It should not be greater than the leak value specified in the print. If the value is greater, the part has failed because It has a leak.

Logically, this is not a replicable test, because same part never will show same values. We can not control the value, we can only control if the part is going to leak or not by checking all joints where It could be a leak. That is the most important thing here.

The problem is that a normal capability study does not represent anything here since we do not control the distribution of the data and I do not care if I have a CPK 1.33 in this case.

Someone recommended me a Gage R&R study Nested. I have already done that in Minitab but I can not find how that can help this.

Does anybody of you guys know something about how to conduct a kind of capability or Gage R&R Study for this type of situations? I really need some help.

Thanks in advance.

 

[Ronen E]

An Honest Gauge R&R Study by Donald J. Wheeler

 

[Marc]

Any specific help/advice will be appreciated. Thanks in advance.

 

[Ronen E]

It doesn't get more specific than that. An honest GR&R study lets you know whether the problem is in the testing equipment, or with the testers (operators), or the actual product. If that can't help, I don't know what can.

 

[Marc]

I was just asking if someone who has actually worked with leak tests could help. Don't take my reply here personally. I was a lab manager in a lab that did leak tests, but that was in the 1980s and I don't remember much other than the operational aspects. Not to mention I'm not a GR&R guru nor am I a math/statistics wiz. The original poster probably isn't either. More specific help from a statistician with or without hands on experience would be appreciated.

 

[Ronen E]

I have some of both - experience with leak testers and with GR&R. I don't consider myself an expert/guru on either, and of course additional input from one would be great. However, I think what can really help here is a clarification from the OP what exactly is the issue. It is my impression something is wrong with the test set-up or the test method, because the results can't be replicated to a satisfactory level. IMHO a straightforward GR&R study should go a long way in this case, but maybe there's something important I don't know / don't understand here.

 

[optomist1]

sometimes, sharing the specifics of the test process can shed some light....Equipment Setup, Specifics air as the pressurized gas, device being tested...etc., etc.,

 

[William55401]

Think about this from a test method validation standpoint. Do you have a known bad part? Hopefully, you have several representative of the defects you are trying to prevent in your process. Does the tester reject these parts? Can you translate the leak rate of known bad into a leak rate that your tester can detect? Is your test method providing protection to your customer? There are both variable and attribute approaches to TMV. The capability studies and the GRR will help you understand inherent variation. However, IMHO, the goal is to ensure that known good (and bad) parts are accurately detected. Your leak test may have ability to analyze variation in measured pressures during Charge, Settle, and Test times. Enough rambling. Have fun.

 

[Miner]

Logically, this is not a replicable test, because same part never will show same values. We can not control the value, we can only control if the part is going to leak or not by checking all joints where It could be a leak. That is the most important thing here.

Are you expecting a measurement device to always provide identical results? This only happens when the measurement device has low resolution. A measurement device with sufficient resolution will always show variable results.

Do your results vary randomly around an average value? Do they trend in a particular direction? What specifically do you find illogical about it?

Note: I do have some experience with leak testing using Helium and a mass spec.

 

[ScottK]

You can purchase calibrated leak standards to validate your test equipment.
My former company (maker of solenoid valves) was looking into that while evaluating automated leak detection...
Here's an example: Air Flow Leak Standards | Custom Leak Standard Solutions