Inspection of Painted Parts 2

[swertel]

My team mates are having an internal argument to define when parts should be painted.

I recall from early in my career that there is a general guideline that says non-organic finishes get inspected after finishing (anodize, plating, etc.) whereas organic coatings get inspected prior to finishing (painting, powder coating, etc.).

Is there a specific industry standard: ASME, ASTM, or ISO that defines when inspection should routinely take place?

--Scott

http://wertel.eng.pro

 

[swertel]

I should clarify the situation here.

As many companies, we order end item parts. Therefore, our vendor machines and paints the parts. We perform a receiving inspection of fully painted parts. Well, of course some dimensions are going to be a little high on the tolerance side due to the thickness of the paint. Since they don't meet the print, they get rejected; but the parts are actually fine. Now Receiving Inspection wants the vendor to ship painted and non-painted parts just so they can inspect them prior to painting. Logic is not winning this argument - I need more ammo.

--Scott

http://wertel.eng.pro

 

[KENAT]

Not quite what you're asking but:

If you work to ASME Y14.5M-1994 and prepare your drawings accordingly this shouldn't bee too much of an issue. "2.4.1 Plated or Coated Parts" basically says that you shall (yes one of those rare shalls not should) specify on the drawing if dimensions apply before or after finishing.

If your drawings don't define this then it's debateable how legitimate it is for you to reject parts the way you describe.

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

 

[MadMango]

I agree with Kenat.

My last company we use to powder coat all of our own parts and weldments. Dims should apply after finishing, but the shop floor didn't like doing the calculations. We ended up having to design parts to account for the paint build-up, and loosen tolerances in order to reduce the rejection rate.


"Art without engineering is dreaming; Engineering without art is calculating."

Have you read faq731-376 to make the be

 

[swertel]

A note on the drawing stating that dimensions apply after painting is the obvious answer, and I foresee a lot of drawing revisions in the near future. I was really hoping that the guidelines I remember was an industry standard I could reference and that it wasn't just an internal standard operating procedure.

And speaking specifically about paint, how do you control glops, drips, and bubbles? If I were to inspect at that point, then my dimension will be out even though the rest is in. On the same token, how can you keep your paint "normal" to a radius surface? I spray the top and bottom of my part. Overspray covers the edges. Fillets/rounds on the corners automatically get sharper. There is no feasible or cost effective way to get a wonderfully smooth and tangential paint on a corner fillet. That inspection will always fall out. There is no amount of math or machining adjustment to account for the variation in paint. Parts that are painted (and other organic "spray-on" coatings) must be inspected prior to finishing.

--Scott

http://wertel.eng.pro

 

[MintJulep]

I would imagine that it depends on what the part is, and what you are inspecting for.

Apparently it's not a "precision" part, or you wouldn't be covering it in an uncontrolled thickness of paint.

I would either:

Note that "dimensions apply PRIOR to painting", and specify a paint thickness.

or

Create a machining drawing and a finish drawing.

 

[KENAT]

A temporary fix until all the drawing revs get done (if ever) might be to add to your standard Purchase Orders or Terms and Conditions wording that "unless otherwise stated, for 'painted' parts dimensions apply before painting". Of course this would require cooperation of purchasing, who are probably in turn expecting Engineering to fix the problem if it's anything like places I've seen.

I think what 14.5 leads me to believe is that if it isn't explicitly stated on the drawing that there isn't actually some other Industry convention/standard that takes effect instead, it's undefined.

I'm tempted to agreethat for painted parts having dimensions apply before painting (or powder coat etc) probably makes sense for most applications. However, it does complicate inspection when the complete manufacturing of piece parts or dimensioned assy's is outsourced.

I will say that back at my last place in the UK most painting was of assemblies that didn't have any dimensional requirements at the assy level, so it wasn't much of an issue. Can any of your painting be moved to assy level?



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

 

[GregLocock]

Since you use the part in a painted state it must achieve its function in the painted state and so the dimensions needed for that function should be in the painted condition.

Sounds like your dimensioning and tolerancing are not being driven by function.



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[KENAT]

Good point Greg, star.

I just put a long post and lost it but basically I think I was wrong in what I said in my 16 Jul 08 21:46 post "dimensions apply before painting (or powder coat etc) probably makes sense for most applications" and Greg's made me see the light.

Perhaps what is actually happening is you're specifying tighter tolerances than you really need for function? I know a lot of time people seem to tolerance more based on what they think machine shops etc can hit without incurring excessive cost than on what's expressly required by function (e.g. ISO 2768). In this case the paint is taking you outside of that range.

In fact from your second post:

Since they don't meet the print, they get rejected; but the parts are actually fine.

which suggests that the tolerances on print are tighter than need be.

(A more qualified checker would have spotted this straight off;-))

As for the bubbles, globs etc, these may have been cause for rejection in their own right at the last place I worked, maybe even here from an aesthetic point of view. As Greg suggests, if these inhibit function they shouldn't be allowed, if they don't then the tolerance should be large enough to allow for them.

In practice I suppose there may be some 'grey areas' or at least ones that are more difficult to spec than is really justified.

One thing I’m thinking about, do you specify acceptable paint thickness in anyway, be it directly or indirectly within a spec/standard you call out? For instance back in the UK in defence we normally specified Def Stan approved paint systems. These specified thickness as I recall, for instance Def Stan 80-55 that we were starting to use states :” 4.5 Under normal applications conditions the dry film thickness will be (20 ± 5) _m, or as specified in the process specification for the material being painted.”

http://www.dstan.mod.uk/data/80/055/00000300.pdf

Elsewhere it talks about g/m^2 which not as direct still gives some effective thickness.

You’d really need this information, even if you work out your end item tolerances based solely on function you still need to verify they can be met by the accumulated tolerance of your machining (or other process) and paint process if not you may need to look at things again. Also the vendor will need to know the paint thickness so they can machine undersize as necessary, or you can have separate machining & finishing drawings per Mint. (Man I’m below par today).


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