How to Determine acceptance Threshold for Surface Mismatch on Machined Parts

[deckypoo]

I'm in a bit of a quandary. My precision inspectors don't maintain consistent acceptance criteria when it comes to measuring steps or surface mismatch on a machined part. I'd like to impose a common acceptance criteria to reduce the engineering burden of dispositioning so many suspect parts with mismatch issues.

The shop I work with deals with model-driven CAM programs and prints with global part profile callouts in the GD&T notes per ASME Y14.5. I have half a mind to stipulate that any step or gouge that does not repeat with a tool path on the part should be evaluated exclusively under part profile. Most of our profile callouts are on the order of 0.050". This rule would presumably allow many visually noticeable steps through inspection. I'd like to challenge the verbiage in the print as much as possible. While I don't see stipulations on surface continuity on the prints we work with, I am worried the default to evaluation by profile only will lead to the presentation of parts that are not "aesthetically pleasing."

Has anyone thought much about this issue?
Can anyone refute my instinct with a requirement I'm missing in Y14.5?
Any thoughts on the principle of delivering an "aesthetically pleasing" part over delivering one that meets the print?

 

[John Predmore]

You want to provide common acceptance criteria. If the judgement of acceptability is entirely aesthetic, it might help to adopt boundary samples of what is acceptable. In my factory, we use pass/fail pictures of defects from customer returned samples . For machined surfaces, you might prefer physical samples, so operators can run their fingernails over the steps, or study reflected light from different angles. This idea is based on surface roughness scoring standards or cast surface samples. I imagine someone sells samples of machined surface discontinuities, or maybe you can fabricate your own.

 

[Bev D]

It all comes down to what is required for the function of the part in it’s final assembly and any customer requirement for the aesthetic need for the finish. Does your Customer need the thing to ‘look pleasing’? That isn’t up to you it’s up to the Customer.

I am reminded of the classic fubar of when Rolls Royce decided to use GM’s new automatic transmission. (The hydrostatic transmission). The Rolls Royce people didn’t like the looks of the rough surface of the inside of the transmission housing so they polished it to a beautiful shine. Anyone who knows how an automatic transmission works is now screaming - as were the Rolls Royce engineers and salesmen when the darn thing wouldn’t work right. It jerked. It shifted when it shouldn’t and didn’t shift when it should. Because the fluid dynamics relied on the rough as forged surface for proper function. Polishing it changed all of that. And what owner of a rolls Royce looks at the INSIDE of their transmission?!

 

[BrokenProbe]

I'm in a bit of a quandary. My precision inspectors don't maintain consistent acceptance criteria when it comes to measuring steps or surface mismatch on a machined part. I'd like to impose a common acceptance criteria to reduce the engineering burden of dispositioning so many suspect parts with mismatch issues.

The shop I work with deals with model-driven CAM programs and prints with global part profile callouts in the GD&T notes per ASME Y14.5. I have half a mind to stipulate that any step or gouge that does not repeat with a tool path on the part should be evaluated exclusively under part profile. Most of our profile callouts are on the order of 0.050". This rule would presumably allow many visually noticeable steps through inspection. I'd like to challenge the verbiage in the print as much as possible. While I don't see stipulations on surface continuity on the prints we work with, I am worried the default to evaluation by profile only will lead to the presentation of parts that are not "aesthetically pleasing."

Has anyone thought much about this issue?
Can anyone refute my instinct with a requirement I'm missing in Y14.5?
Any thoughts on the principle of delivering an "aesthetically pleasing" part over delivering one that meets the print?

My short time making cell phone covers we had Class A, B and C surfaces. Class A would be customer facing most of the time so say the front of the phone, that surface needed to be almost perfect, Class B would be the back of the phone or battery side so more defects allowed, Class C was internal and rarely customer facing. I would determine it along that criteria. As machining goes, the standards I have seen from our customers are for radius to flat mismatches. .005" for open angles meeting at radius and .010" for closed angles meeting at radius. casting to machined mismatches can go as high as .030"