Surface finish and flatness 1

[Mario11]

Hello, I am using a carbon seal assembly with a flatness of 3 helium light bands with a surface finish of 10 and an operating speed of 3000 RPM. We have been experiencing leakage of hydraulic fluid.

This design is from a legacy product from the 1970's that has never experienced leakage in the past.

Can visible surface scratches pass a surface finish call out?

Could the surface finish be the problem?

What is the rule of thumb for surface finishes on a rotary carbon seal?

Thanks for the help!

Mario

 

[MikeHalloran]

I think visible scratches can trash a carbon seal face.

... And that includes the scratch left by the stylus of a surface finish measuring machine.




Mike Halloran
Pembroke Pines, FL, USA

 

[Mario11]

Thanks Mike,

I am there too; however the runner face seals are passing QC inspection and are meeting the 10 finish measured with a profilometer. Is it possible that surface scratches can meet a 10 finish?

I am pulling these parts and having them lapped with a requirement of no surface scratches allowed under 10x magnification and a surface finish of 4.

I hope this helps

 

[KENAT]

Wouldn't the scratches have to be within the limit of the 10 to pass?

I have some samples here and the 16 & 8 microinch finishes are very smooth on the ground finish but on the flat lapping sample for 8 you can indeed see the very small scratches caused I by the lapping process.

On the 16 finish for horizontal milling, vertical milling & turning you can definitely still see machine marks.

Not sure this helps though.

KENAT, probably the least qualified checker you'll ever meet...

 

[MikeHalloran]

I'm guessing you started noticing the leaks shortly after you started inspecting the seal faces with a profilometer. So, stop doing it.

You should not be touching the lapped faces with your fingers, let alone a profilometer.




Mike Halloran
Pembroke Pines, FL, USA

 

[fcsuper]

That's kinda what I was thinking, MikeHalloran. If a test is needed to QC the parts, AQL processes may be to be implemented, with the profilometer test treated as destructive (meaning the sample of tested parts are to be scraped after testing).



Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group

 

[Mario11]

Thanks guys for the help. I verified that the profilometer did scratch the face seal of the runner, but I am not sure this is the cause of the leakage, yet.

 

[IRstuff]

If your surface is that touchy, you probably need to look at using a noncontact surface measurement tool.

TTFN

FAQ731-376

 

[MikeHalloran]

No, they need to limit their inspection to verifying the package marking. The seal itself should not be unwrapped until installation time.

An inspection operation that would even consider using profilometers on carbon seal faces is probably doing more harm than good in other ways too, e.g., leaving the packaging open in other than a clean room, mixing up the parts among seals in a lot....






Mike Halloran
Pembroke Pines, FL, USA