Press fit 303 into 303.

[WGP1]

In our manufacturing process, we press a .1212/.1215 OD 303 SS insert into a .1207/.1210 ID SS body to a depth of .060 inches. The problem we are having is leaking between the insert and body in excess of 0.2 SCCM. We realize that similar metals should not be used in this assembly, unfortunately this being a medical application changing material is slow and difficult.

We have tried press to a lessor depth and the leak has dropped. Next, we're machining a counter-bore hole for alignment before the press fit section.

Any ideas for a permanent fix?

 

[NickE]

use a press fit adhesive. furnace brazing to eliminate gaps by capilary action.

 

[WGP1]

NickE,
The end product we manufacture goes into medical applications where any residue other than the base material would cause a failure. We have thought of heating the body to expand the hole and then drop the insert in.

 

[metengr]

WPG1;
I like the interference fit option, it should work well with polished mating surfaces. You can either freeze the insert using liquid nitrogen and lightly preheat the mating piece to assure sufficient diametrical clearance. I would guess that by press fitting you are galling between the two surfaces and creating a leak path. By using a true interference fit, once the parts cool to ambient temperature, the uniform metal to metal contact around the circumference should provide a nice seal.

 

[israelkk]

What is the maximum leak allowed. Can you switch to 304L and use laser welding or a micro TIG welding. We had a miniature bellow manufactured by micro TIG welding with great success.

 

[unclesyd]

First off I have been lead to believe that 303 SS wasn't allowed in medical devices. This might only pertain to bodily contact.
We have found that cooling the insert works best when using Cu or SS insert into SS especially if the component has a non-uniform shape.

We always consider that 303 SS wasn't weldable or brazeable. Either process, welding or brazing will entail a cleaning step.

 

[WGP1]

First, as a new member I would like to say, thank-you all for your suggestions and support! This is a great venue for true collaboration of ideas. Second, I'm amazed at the quickness of responses I received about our problem. This site is now a daily read.

Below is the result of an Engineering Symposium we had at work. Let me know what you think. One particular application for the part is the delivery of Nitric Oxide.

Different 300 Series
TBD possible material 17-4
Ream
Lubricant
Shorten length of insert depth
Cutting edge small ID of chamfer
MEMs
Machined in insert, EDM Lazier; Water Jet the hole
Plating
Welding
Heat/Freeze parts
Electro polish
Machine counter bore in body

 

[patdaly]

WGP1,
I would first explore metengrs suggestion. You have a fair amount of interference fit for the diameter and his suggestion that galling is providing the leak path also fits with my experience.

Is the operation done is such a quantity that heating/cooling would cause undue production delays?

 

[WGP1]

You are both correct, galling is ocurring. I have cut apart the bodies and found galling in every one. An engineer here calcuated frezing the insert but it's shrinkage was not enough. Heating the body is something I'll do today. Yes, production will be impacted greatly by heating or cooling the components.

How critical is the finish? Does a higher finish give a better press fit? What about maching grooves or a barb like configuation on the OD of the insert?

 

[patdaly]

WGP1,
Why such an interference fit?
Another possibility that I just thought of might be friction spin welding the two together, if the insert is long enough to grip.
Yes, the finer the finish the better, and though you may not remove the interference fit with heating/cooling, you help the interference fit while you press it in.
Is this on an automated line? What is the approximate volume produced per day?

 

[WGP1]

According to our vendors, .0003" is all the tolerance they can hold. So, we have a minimum of .0002", max of .0008".

The length of the insert is, .140" and .060" is in the body. Which by the way, is I beleave, too deep.

This part is pressed using a single station operation. Estimate runs are 200 per day but not every day, est 6K per year.

 

[patdaly]

WGP1, forget my first statement, I read the dimensions wrong. 8 tenths max isnt out of line.

 

[redpicker]

I like patdaly's suggestion, friction weld. After all, that is all a gall is. If you can't beat 'em, join 'em.

With a small part like that, you could probably design it so that a small CNC lathe could do the welding and with a cycle time measured in seconds, that production rate should be no problem.

 

[israelkk]

Usually stainless fittings are silver plated on the thread to avoid galling. However I do not believe silver will be acceptable as bio-compatible. Add to this that press-fit created stresses greater than threading. I would try to completely gold plate the insert which is quite small and the added price is minimal. The plate thickness is usually 0.002mm thick and it may serve as a lubricant and it is bio-compatible

 

[ccw]

Hi WGP1,

We have been using metengr's method with LN2 as the cooling agent for years. It is a more kinder, gentler operation than force fit.

To minimize scrap, we use select fit of plug to bore. Our typical plug and bore nominal diameter is about 33 times larger than yours. The material we use is 304 SS. Holding close machining tolerances on cold drawn SS tubing (our bore material form) can be problematic due to locked up internal stress causing movement after machining.

With a 16 microfinish, no electropolishing, we typically see 80-100 psig hydrotest and less than 1 x 10-6 cc/sec helium leak test.

 

[WGP1]

ccw,
I just tried liquid nitrogen, measured a .0001" shrinkage. I like your suggestion for Nitronic alloys, is it hard to machine, costly, hard to find? The application is for a valve and the insert is the part with the orifice.

 

[EdStainless]

1. You should be getting a lot more size change. If you heat the 'hole' to 400F and cool the insert to -300F you should have about 0.006" clearance, if they were identical to start with.

2. Press fit two steps, one at each edge. Make the diameter of the bottom one smaller than the top one. This way you will have two short press fits. This should prevent the long continous gall tracks that allow leakage.

= = = = = = = = = = = = = = = = = = = =
Corrosion, every where, all the time.
Manage it or it will manage you.

http://www.trent-tube.com/contact/Tech_Assist.cfm

 

[ccw]

Hi WGP1,

Nitronic 60 usage is popular and is readily available from stainless stock suppliers. It is more expensive than the standard 3XX grades but less than the super alloys.

See:

http://www.gocarlson.com/AlloyPDF/Nitronic60.pdf

for a good description of the alloy.