How to expand a metal seat ring (SS316) with a teflon insert to fit in a bore that is too big?

[NBR90]

The OD of the metal seat ring is about 75 mm (3") and about 25 mm thick. It is suppose to be fit into a bore with ID of approximately 75.5 mm (standard carbon steel). So obviously this is not a correct size as the bore needs to be smaller than the OD of the metal seat by about .30 mm. So the OD of this seat needs to 'expand' at least .80 mm or to 75.80 mm or the bore needs to be 'shrunk' by 0.80 mm or to 74.70 mm.

Welding or other type of thermal spraying the seat is out of the question as it may damage/distort the seat and or teflon insert. Rewelding/remachining the bore is also not considered due to time and work involved. I have two ideas here but would appreciate it if someone has some input to share...


1) Machine a new sleeve with ID of 74.90 mm (to shrink fit the seat into it). This new sleeve would have an wall thickness of 3 mm or OD of 80.90 mm. Once I use liquid nitrogen to fit the seat into the sleeve, I re-machine the sleeve to the exact size I need or 75.8 mm. The question is, would there be any problem when re-machining the new insert to about 0.8 mm thickness, would it crack being so thinned after being shrink-fitted? If it all works out, I would shrink fit this again into the bore.

2) Machine the bore bigger so you can insert a metal sleeve about .10 mm bigger than the bore and shrink fit it inside the bore. The new 'bore' would have a new ID approximately 74.70 mm. But how 'thick' of a metal sleeve can be machined effectively without it being distorted?

 

[racookpe1978]

Why so large an overlap between bore and seat?

Usually, an interference fit is no more than 0.006 - 0.010 inch.

 

[SnTMan]

One of these products work?

http://www.sealanddesign.com/category/Retaining/205.html

Regards,

Mike

 

[NBR90]

Yes the inteference is 0.30 mm. Sometimes it can be even larger up to .50 mm. Its for high pressure fluid services. LN2 can shrink it up to .50 mm, Ive done it many times. Just this time the wrong part was given to me and there is no time to reorder right part. Those inteference is per OEM spec, even tho I agree no more than .10 mm should be necessary. But 0 leakage is required and as long as LN2 can do the job, it doesnt matter to me how big of interference. In any case, even if 0.1 mm, same problem persist... how to make it fit...

As for loctite, its out of question, maybe if we were talking about .05 mm gap. Even then, if the bore is bigger, the seat may pop out under 5000 psi pressure which can lead to various disaster....

 

[tbuelna]

If I understand correctly, the repair sleeve will be trapped between the seat OD and the mating bore. A thermal fit of .001" per inch of diameter between the seat OD and the repair sleeve ID would be quite adequate to allow any rework machining of the repair sleeve OD. Even if the repair sleeve thickness is reduced to 0.8mm after rework, this should present no problem in regards to achieving a suitable thermal fit with the housing bore.

The repair sleeve is thin and has minimal radial stiffness compared to the thick section seat and housing.

 

[Artisi]

I would contact Loctite, they have a wealth of experience and a lot of information to fall back on.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[NBR90]

tbuelna, from my understanding of your post, it is ok to make a sleeve about 3 mm 'fit' the seat OD (by cold-shrinking), then machine the sleeve to less than .8 mm over the seat? The sleeve would not crack or lose its integrity?

 

[tbuelna]

You can get the interference fit you need by a combination of heating the housing and chilling the sleeve. When the parts return to RT, there will be mechanical strain created in both parts. When the repair sleeve is machined and its radial section thickness is reduced, the structural stiffness of the sleeve will also be reduced. But even machining the 75mm OD repair sleeve down to a wall thickness of 0.8mm should still result in enough interference fit to keep the sleeve in place assuming the housing and sleeve have similar CTE rates.

You did not mention what the repair sleeve material was, but I can't see there being a fracture issue with any metal I can think of when machining the repair sleeve to a .8mm thickness.

As for leakage past the OD of the repair sleeve when exposed to 5000 psi(d?) fluid pressures, that is not an easy question to answer without having more specific details about the installation. If the 76mm dia x.80mm wide end face of the repair sleeve were exposed to a 5000psi pressure, that would produce an axial force of over 1460lbs on the sleeve. If there is nothing other than friction preventing the repair sleeve from moving axially (like a shoulder on the housing) then it's likely the sleeve will be displaced.

It is also difficult to obtain a leak-free high pressure seal with just an interference fit between two large diameter metal components.

 

[NBR90]

the repair sleeve is SS316 just like the seat with teflon insert in the middle. I can heat the bore, but wouldn't that negate the sleeved seat that is chilled by liquid nitrogen? As the -200 degree sleeved seat would melt as soon as it is placed into or near the bore. Why not just freeze the sleeve seat and keep the bore room temperature?

This is actually for open/close ball valve for hydrogen service. When closing, the ball valve would stop the seat from moving anywhere even if the sleeve 'wants' to wonder.

What we have done before (safer) was to re-bore the bore 2-3 mm larger and shrink a 2-3 mm sleeve inside the bore first. The bore has about 25 mm height so does the seat (thickness).

From our experience, seat/bore interference this tight always result in zero leakage at high pressure it is designed for. We tested it countless times and always get zero leakage.

Anyways, thanks for the info about the .8 mm sleeve, we will test that next time to see how it works .