Jump to content

Surface Roughness Question


---
 Share

Recommended Posts

Man.... Surface Finish is a rabbit hole to go down.

The Machinist's Handbook is a great overall resource for deciphering Roughness Standards.

A lot of it boils down to the parameter (Ra, Rz, etc.), and the surface finish you are trying to achieve.

An example would be if you are trying to achieve a surface finish between 0.025 to 0.1 Rz (in microns) then you need to use a cutoff length of 0.08.

There is also information regarding what type of production process you have, and the type of stylus that you use (size of diamond as well).
Link to comment
Share on other sites

Please sign in to view this quote.

Thanks guys, we have to hold Rt so this is extremely helpful. I'm going to look in some of our machinist's handbooks to find out more because it has been one hell of a ride since our main contracted company started requiring a workmanship standard focusing more on surface finish.
Link to comment
Share on other sites

So from my understanding of the Machinist Handbook, Rt/Ra cut off lengths decrease the tighter the tolerance. Does that mean when we are trying to hold a 16Ra with an Rt of 48, the stroke length should be less than .100in based on the cut off length filters? I found this nice little chart somewhere on the ole google...


2725_03f658419fc5bbb71a9dba10e05932d0.png
Link to comment
Share on other sites

Please sign in to view this quote.

I've never had to measure waviness, but I know that material ratios, especially on the Primary side, is a pain in the butt, and you can't get a standard to check against, so you just hope for the best.
Link to comment
Share on other sites

Please sign in to view this quote.

I'd have to look into it some more, but I don't lose a lot of sleep on the stroke length (length of trace) that much. I just know that if I need to use a 0.8 cutoff that my trace has to be at a minimum that long.

I do believe the standard will say that for this cutoff to measure this much length, but that gets thrown out the window as soon as someone throws surface finish on a surface that is too short to achieve that. There are also cases where you would need to step down on the filter, but that is a risky thing because it can make the surface finish look much better than it is.
Link to comment
Share on other sites

Please sign in to view this quote.

What led to me asking these questions is because we had a bore that required a 32Ra finish and when we changed our stroke length from .050in .192in the Ra reading was dramatically different. .050 was 12Ra & .192 was 78Ra
Link to comment
Share on other sites

That can be for a number of reasons.

The general rule of thumb is to use the correct cutoff for your application, but measure as much of the surface as you can. The cutoff is just how many times it slices up the trace length. You want as many as possible to help detect abnormalities.

It's also good to look at the raw data and see why it is showing higher.
Link to comment
Share on other sites

Repeatability with a profilometer is not really that good, especially if you hold it in the same place where the diamond tip will actually un-visually scratch/smooth the surface of most materials. Do a full gage R&R with one and you'll probably pull your hair out. 🙄
Also, it's a good idea to re-calibrate it with a master fairly frequently if its moved around a lot and used by a lot of different people.
Link to comment
Share on other sites

There are several methods for determining the proper cutoff to use, based on Ra, Rz or "sm" mean peak spacing parameter. According to the chart above and what I have as well the standard cutoff for 16 Ra would be .01. Standard evaluation length is generally 5 cutoffs so in your example that would be .05".
Link to comment
Share on other sites

I've has some training from someone on the ASME 46.1 board.

To be brief : 4µin or less should be .01" cutoff x 5,
4-80µin .03" x5
80-320µin .100 x5

Also square up your roughness 'patch' and measure, your results , should be ± 1µin.

Profilometer Manufacturers may state to allow ± 2µin.

Good luck.
Link to comment
Share on other sites

 Share

×
×
  • Create New...