Surface Textures and Finishes? 2

[Superslinky]

Hi all, I'm hoping someone can help me shed some light on this. Occasionally (like all of us) I have a machined part that is mating with another machined part. I've always seen symbols depicting a "finish" but I'm having a hard time understanding what values are used in/on what applications. For example, if I have a round plate and I plan to machine a flat recess into it to accept the mating part (maybe partially for locating), should it have a controlled finish? When do controlled surfaces come into play and if I need to add a controlled surface how do I determanie the value? I have grown custom to seeing values of 16 and 32 mostly.

Thanks all..

 

[drawoh]

Superslinky,

Buy yourself a surface roughness comparator. It really helps to have something in your hand to show you what the numbers mean.

JHG

 

[EngineerTex]

Sounds like you want a feel for what surface finishes are.

A 250 microinch surface finish can easily be achieved by a water jet cutter, some laser cutters and in certain instances plasma or gas torch cutters. Also applicable to some rough sand-casted parts.

A 125 microinch surface is about what you would expect after quickly running a part through a milling machine or getting a rough turn on a lathe or the amount of mill scale on a quenched and tempered part.

A 64 microinch surface is the result of more controlled machining, often the inner surface of a drilled part or the surface of as-rolled steel products.

A 32 microinch surface is the result of further controlled machining operations and can be achieved by good milling practices and is about the roughest you would get with a grinding operation (not a hand grinder).

A 16 microinch surface is shiny, but wouldn't be mistaken for a mirror finish and is usually achieved with a grinding operation. This is often used in groove surfaces for sealing with o-rings at high pressures. This is about the roughest that a journal bearing will be, though they are usually smoother.

An 8 microinch surface can be ground but can also be lapped. This is close to a mirror-like finish and is used for producing bearing races and the rolling elements in rolling-element bearings.

A 4 microinch surface is a mirror-like finish and will almost certainly require grinding, lapping and honing. Typical applications would be surfaces of dies used to produce plastic injection molded parts that have super-smooth surfaces as well as high-precision rolling bearings.

For your application, I think of something like the alignment hub on a hydraulic motor that centralizes the motor into a mating brake. I have never included a surface finish on these in the past any better than the 125 that is standard on our drawing title block. However, that being said, I would doubt that the actual finished machine surface is any rougher than a 64 and I'm sure it's not any smoother than a 32.

The explanations of all of these numbers is fairly general and it is obvious that any of the machining operations I referenced above could vary widely depending on machine precision (or lack thereof.)

Engineering is not the science behind building. It is the science behind not building.

 

[flash3780]

The Machinery's Handbook has some good information on what surface finishes are achievable by what operations (in the form of a general range). Generally speaking, most production parts that I've seen don't specify a finish smoother than 32 microinches except in extremely specialized applications.

Many drawings that I've seen have a default surface finish for machined surfaces of 125 microinches. In general, things like bearings will require very smooth surface finishes. Seal-shaft interfaces generally require a maximum surface roughness of 16 microinches. I have always specified maximum surface roughness of 32 microinches for both surfaces of press fits - which was derived from company best practices.

The best practice, I suppose, would be to look at what was specified on the last design, what was achieved, and how successful it was.

 

[unclesyd]

Adding a little to the post by EngineerTex don't confuse polished surfaces versus smooth surfaces unless it's by design. You can polish a pretty rough surface and it's pretty but not smooth. We have fought that battle many times as our lapping process produces a super finish and a very flat edge to edge part but it has a matt finish. When it comes to the production people saying what they want they will choose the shiny one though it maybe many times rougher .

The link to Flexbar is for finish comparator kits. The link to Testex if for a replica tape for CMA among other things.

http://www.flexbar.com/shop/pc/Surface-Roughness-Testers-and-Comparator-Plates-c25.htm

http://www.testextape.com/index.php

 

[Superslinky]

Thanks guys, I think EngineerTex and unclesyd probably had the best explaintion I was looking for. I really just needed to visualize the differences. I did look in the machinery's handbook but without the visualization it didn't help much. I think that a surface comparator kit will do me some good and seeing as though they're only around 80 buck, its probably well worth the money.

Thanks again!

 

[KENAT]

An important thing to realize is that the same number from a different process will not look the same.

For instance a ground surface looks pretty darn good even at 63 microinches.

However, 63 achieved by say horizontal milling won't look very good at all.

Also, depending on application, the other aspects of surface call out such as lay, may make a big difference both functionally and aesthetically.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[GannetAM]

I believe the attached figure will answer all your questions about surface finish and how it affect relative cost and tolerancing. It should be noted, that the surface finish exists within the tolerance range.