Leak test for Production of water treatment product 1

[vladrath]

Greetings,

1st I'm pretty new to these forums and hope that I am posting this in the correct place. I'm also pretty new to my current job and want to make sure I develop an informed opinion on the following topic.

I work at a company which produces water treatment products. One of these products (Reverse osmosis units) requires leak testing since they get installed in peoples kitchens. They've had problems around here with leak testing and have settled on filling these units with water to (approximately) 150 psi. They then have them sit on a rack for 72 hours before being inspected for water leakage. Then an operator empties each unit out and sends them to be packaged.

This method seems horribly inefficient and all the hourly workers hate doing this job.

The statues quo around here is that this is the best possibly method but something just makes me skeptical of this. I also don't really want to bring this up with the quality guy without a solid opinion.

So my question basically is has anyone had experience with a more efficient way to leak test rigid plastic (mat'l: Polyfort ie polypropylene) products without using any dyes that would be dangerous if consumed. Optimally I would like to get these things tested in an hour or less.

Some of my thoughts on how to proceed could be.

1- Using a higher water pressure. Primary concern here is with breaking the units.
2- High air pressure. I don't necessarily like these since if a failure occurs it could be dangerous and although I can install whatever safety shields necessary I don't want to make it a huge hassle to test these things.
3- Create a vacuum and test to see if it holds for some amount of time.
4- Some sort of tracer. I know that Helium is often used as a tracer but they already thought about that and its too expensive I guess. But are there any other products I could use that are cost effective and safe for consumption.

So I'm hoping others can help me with some insight into this topic. I don't need anyone to solve my problem for me, instead I really could use some advice from someone with some experience with this type of thing and help finding a good direction.

 

[BillPSU]

Helium lead detection from the inside going out would be what I would look at. The recommendation is to pressurize to the normal operating pressure and sniff for the presence of helium leaking out. You would have to protect against bursting but the process could be automated if the volume was sufficient.

Bill

 

[redpicker]

My guess is that their method is just exactly what they want. I'd guess they have to deal not only with initial leaks, but also delayed leaks. Most of these units are put together with compression fittings or quick disconnect fittings that may not show a leak initially, but may show one after a day or so, but if it hasn't leaked in 3 days, then is probably won't.

Going to a higher pressure is probably not feasable since the housings probably won't take it.

Automating leak detection probably isn't necessary since I'd suspect if you have a leaker, you know it after 3 days.

It sounds as if the biggest problem is the workers don't like emptying the units after the test is complete. If so, see if you can automate the part that nobody likes.

rp

 

[MintJulep]

Pressurize with air, submerge in water, look for bubbles.

 

[MikeHalloran]

Pressurize it with refrigerant, survey for leaks with an electronic refrigerant leak detector, then recover the refrigerant for reuse (with a 'recovery machine').

No water to clean up, no danger of bacterial growth from water left within.

In days of yore, it was easier just to vent the refrigerant (it doesn't take a lot), but you can't do that anymore. The recovery machines are not difficult to use.



Mike Halloran
Pembroke Pines, FL, USA

 

[rodofgod]

I like Mikes idea but I would investigate the use of 95% Nitrogen, 5% Hydrogen as your tracer gas. It's quicker,a lot cheaper and will find smaller leaks and as a bonus,you can release the gas into the atmosphere.So no costly recovery equipment.

 

[Duwe6]

And at a 5% or less concentration, Hydrogen is not considered explosive. With Helium getting scarce [non-renewable resource], Hydrogen is becomming the 'go-to' tracer gas for sensitive leak detection.

 

[vladrath]

Thanks a lot everyone. Certainly some good potential for a solution here.

Bill, they've already looked into helium trace detection and didn't like the idea for whatever reason.

Redpicker, yes many of the fittings are compression fit plastic fittings. One issue was coming from o-rings rolling off their desired position but that issue has been remedied by some better fixtures.

I agree with you also on the last part, if it's not feasible (for whatever reason) to change the test method I'll be looking into increasing the speed where they empty the units.

Mint, Mike, Rod & Duwe thanks, I will look into these methods. The 95%N2 5%H2 sounds interesting.

 

[MikeHalloran]

Rolling o-rings suggests they need a tiny bit of grease.


Mike Halloran
Pembroke Pines, FL, USA

 

[rodofgod]

Yes, I'm going to investigate 95% Nitrogen, 5% Hydrogen a bit more, we use helium at the moment and it's good but costly!