Need Help with Cavitations and Vibrations on Glass

[selamens]

Hello,

Currently I'm interning at a Manufacturing Company for Military Materials, MREs, Chemical Detection kits, etc.

I have a situation that I can't figure out, and I'm really hoping you can help.

Situation: The company creates these small 1 1/4 inch glass ampules. They create about 50,000 of them with materials inside, some liquid and some powder. They ship these materials to another company, and this company is complaing because there are to many rejects. just about 1%.

To solve this problem, they have been placing these ampules into a Ultrasonic Cleaner by Eumax, model number UD100SH-4L. With the frequency of 50/60 hz. They place around 800 ampules into the cleaner and allow it to vibrate for 15 minutes. Once complete they place all of them into a vacuum filled with water to suck out all the rejects.

The problem is, after testing the ampules twice, every time they keep getting 1 %, you would figure it would stop. I believe every time they test they are weakening the glass at a 1% rate, thus everytime, they get 1% rejection.


Solution: I need to find an optimal frequency or some type of solution to accurately test the rejection rate, to test the strength of the glass, and to be able to ship 50,000 ampules at a small to none rejection rate.

Can anyone help?

Please?

 

[IRstuff]

That's a lot of writing without stating what is the criteria for rejection.

TTFN

FAQ731-376

 

[selamens]

Sorry...

The criteria is simple... It can't leak or break.

 

[IRstuff]

I think "every time they test they are weakening the glass," is slightly off the mark, the vibration is simply accelerating the failure of already weak parts. Vibration and temperature are two ways to accelerate physical failures.

The big question really isn't that you don't have a means of screening the parts, since you do, it's just that the parts are wimpy to start with, and your company's process engineer needs to get off the dime and start looking for the root cause of the problem.

As for finding a "better" test, you could look for an ultrasonic cleaner that allows its frequency to be varied. Even then, you might not get a much higher failure rate without adjusting the amplitude of the vibration and/or increasing the time duration. The simplest thing to do first is to run the same test for different times, up to maybe 8 hrs and see if the failures decrease. If even the last 15 minute test yields 1%, then you're in the flat part of the reliability bathtub curve and only a process change can really do anything for you.

TTFN

FAQ731-376