Tolerances to apply before or after anodising? 1

[adventg]

Evening all,

Just a quick question on general opinion and concensus.

Generally, does one specify part tolerances as before or after surface treatments such as anodising? Clearly this is important as a tolerance in the range of a few to tens of microns, could potentially be well out of spec due to anodising?

What is the convention here, to specify tolerances having already taken account of film thicknesses, so that tolerances fall into where you really want them after treatment, or to specify tolerances as after anodising (and leave it up to the shop, which isn't in my opinion an ideal way of doing such things)

 

[pmarc]

ASME Y14.5-2009, para. 2.4.1:
"Where a part is to be plated or coated, the drawing or referenced document shall specify whether the dimensions apply before or after plating. Typical examples of notes are the following:
(a)"DIMENSIONAL LIMITS APPLY AFTER PLATING."
(b)"DIMENSIONAL LIMITS APPLY BEFORE PLATING."
(For processes other than plating, substitute the appropriate term.)"

 

[MintJulep]

It really depends on how important the plating thickness is for your specific application.

If you specify "after plating" then you are turning design control of the size of the before plated piece over to someone else. Possibly the machinist on the shop floor. With a calculator if you're lucky. By his gut feel if you're not.

If you specify "before plating" then you have nothing to inspect the finished part to.

If it's really critical then the best way may be to make two drawings. As-machined and as-plated.

Alternately, it might be possible to add a note to the effect of "subtract something from all nominal dimensions to establish as-machined dimensions".

 

[JNieman]

If your plating is to a spec with a tolerance on thickness of deposition/growth, then you may be able to simply calculate your before-plating tolerance to an acceptable/reasonable requirement.

_________________________________________
NX8.0, Solidworks 2014, AutoCAD, Enovia V5

 

[dgallup]

If we are buying the part complete with the plating, I always specify the dimensions after plating. It is up to the supplier to figure out how. If we are making the part then we make operation sequence sheets that show the part each step of the way and we have to do the hard work of figuring out how we will meet the finished part requirements. Sometimes you have to do a finish grind after plating if the tolerances are really tight.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[KENAT]

For most situations the plating thickness is significantly less than the allowable total tolerance, so 'DIMENSION APPLY AFTER PLATING' doesn't actually impose much of a burden on the poor machinist to make sure he allows for plating thickness etc.

Additionally, engineering drawing generally defines acceptable finished part taking into account functionality of that finished part - so 'DIMENSION APPLY AFTER PLATING' usually makes more sense.

However, for tightly toleranced parts where plating thickness is significant in relation to total tolerances then more thought may be required.

How this is done may depend on how tightly integrated you are with your supply chain - i.e. internal shop, external 'partner' shop, external 'random lowest bidder' shop...

For the former case, I'd probably rely on manufacturing eng/machine shop supervisor/senior machinist or whoever to come up with a routing or 'production plan' or similar 'pre treatment' drawing.

For the latter case, maybe 2 separate drawings but then you leave yourself open if failed parts come in. You're inspecting to the plated part drawing, they machined to the unplated drawing. If there is a discrepancy it's harder to tell the problem - could be bad unplated drawing, bad machining, bad plating and may not be easy to tell which.

Some platings give nice uniform coatings, some can vary quite significantly based on part geometry and other factors - for these trying to simply add or subtract nominal plating thickness may be easier said than done.

Most of the time I'm guessing the 2 things you care about are the finished part dimensions, and the plating thickness. The pre plating part dimensions are effectively secondary to function so while I'd never say never, most of the time I'd rather not get involved in it.

Posting guidelines faq731-376

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

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

 

[swertel]

I have used an internal standard (which we give to suppliers as necessary) so we don't have to place that note on every drawing. In short, the standard says

Dimensions apply after plating for inorganic coatings.
Dimensions apply before plating for organic coatings.

Anodize, cadmium, nickel, zinc, black oxide -- all apply after plating.
Paint, ink, powder coat -- all apply before plating.

--Scott
www.wertel.pro

 

[Tunalover]

I contend that tolerances for ALL parts be stated to apply AFTER PLATING (or PAINTING, or...) because it is the finished part that fits into the assembly. My experience has been that any decent fabricator can calculate how much bigger to make internal features (and smaller to make external features) to compensate for the coating that effectively makes the feature sizes grow. I have NEVER, in nearly 30yrs of design, had a fabricator complain about this practice. This could be a tribute to the quality of the suppliers I've had the pleasure to work with over those years! Note that I've generally worked with suppliers that supply as-finished parts and I don't generally hassle with two separate sources. I recommend you do the same.




Tunalover

 

[dgallup]

You obviously don't live in a world with tolerances at the micron level. A plating spec of 8+/-3 microns becomes a big deal when you need a finished part tolerance of +/-10 microns and the plating is on 2 surfaces. The 60% of your part tolerance is now being consumed by the plating.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[Tunalover]

dgallup-
You're right. I've never been concerned with tolerances that small. What kind of business are you in that requires tolerances that tight?


Tunalover

 

[JNieman]

We routinely deal with finish requirements measured in tenths (.0001) and still have issues with control. I don't like giving out feature dimensions "after plating/painting" with no indication to the thickness of the deposition, or the size of the pre-plated/painted feature. In fact, if it's painted, it's obviously not a critical item, and I'd rather just dimension the pre-paint item and understand the thickness added for paint.

@tunalover: I bet your suppliers make the painted parts to final dimensions, and the paint adds extra. No one is going to bother subtracting any allowance for paint if they're given certain dimensions. The practical tolerance on a painted part is likely such that they don't have to, anyways.

Short answer for me - if I don't control the pre-condition and final-condition, it leaves a gaping liability.

This is absolutely impractical to the point of impossibility, but for illustrative purposes, let's say you give "all dimensions apply after plating/painting" on a print. Now let's say make 1/2"dia holes oversized to 1"dia and lay a 1/4" of paint all around the inside of that hole. Not a good day. If your paint specs call out a min/max deposition, then you're covered, but if you simply say "applies after paint" and that's all - it is yet undefined.

If you spec the thickness of a coating/growth during final processes, and then a dimension after that process, then you've already fully defined and toleranced the process, even if the supplier has to do some math.

_________________________________________
NX8.0, Solidworks 2014, AutoCAD, Enovia V5

 

[Tunalover]

@JNieman
The fabricators do do the math I AM TOLD. Parts have been sent back for those that didn't for painted parts. Of course I am relying on the word of the front office representatives (who are of a sales variety). Somebody must be doing it right because the orders sent back have been few.



Tunalover

 

[tbuelna]

I have seen many examples where the tolerance of a plated surface must be controlled to within +/-.0001" or less. For example, thin dense chrome plating is commonly applied to high-precision bearing races and journals for corrosion protection. But in order to hold the tight tolerances the surface is finish ground after plating. In this case the dimension does not apply before or immediately after plating, instead it applies to the as delivered condition. If the final thickness of the coating is a concern, then you would include a note on the drawing indicating what the tolerance limits of the final coating thickness are. This will direct manufacturing planners to adjust their work instructions accordingly, and QA to perform whatever in-process and post-process inspections that are necessary to validate the final coating thickness conforms to the drawing requirements.

Plated or anodized screw threads are a common situation where some allowance during machining must be made for the coating thickness. However, precision (ie. UN class 3) screw threads are typically inspected with standardized snap gages or go/no-go gages that are sized for validating the final thread condition. FED-STD-H28 describe a standardized inspection method for UN screw threads. Attached is section 5.4 that covers the issue of inspecting coated UN threads. It basically states that when you call out a UN thread on your drawing the final thread condition must conform to the standard, which means after plating. But there is no limitation on any in=process inspection you might choose to employ.

In short, engineering drawings should define the condition of the specific part as delivered. You should think of a drawing as a set of requirements for the finished part, and not as a set of manufacturing instructions. Your drawing should describe to manufacturing minimum standards the finished part must conform to, rather than trying to tell manufacturing how to do their job.

 

[Tunalover]

tbuelna-
You are preaching to the choir.


Tunalover

 

[tbuelna]

Sorry if it sounds like that. But I saw quite a few posts that had a different perspective, so i thought it was worthwhile to offer my opinion on the matter which is a bit different than any other I read.

What I stated was that drawing dimensions don't apply either before or immediately after plating, but always apply to the final condition of the part. As I said, there are situations where the plated surface may be finish ground which would alter its dimension after plating.

However, your comment about painted surfaces presents a unique situation. Typically, close tolerance features are never painted. The thickness of a primer and topcoat can easily be .005" or more. It is also common to make a separate drawing that describes just the application of primer and paint, including masking of specific surfaces. There is a machined part or sub-assy which is called out on a finish drawing that describes the painting required. These finish drawings can get quite complicated for some components. I have made finish drawings for gearbox housings that were several sheets with detailed dimensions for masking surfaces like fastener spotfaces, sealing faces, threaded holes, studs, dowel pins, and electrical bonding surfaces. The masking of these housings would get so complicated we would have special plugs, caps and shapes of masking tape made to speed up the process.

 

[JNieman]

I believe the scope of this forum section applies to a swath of people who may be in the position to create both engineering and manufacturing (in-process) drawings. The difference of approach is neither right nor wrong in general, but only when applied to specific contexts. I can see how this could create some confusion when we have different mental-frames approaching a topic such as this.

On the 'engineering drawing' front, I agree with the statement that you should not make a drawing that infers the manufacturing method. I do believe there are exceptional circumstances that are not horribly uncommon, where an engineer would be wise to provide an "in process" requirement. Giving a bit of control over the manufacturing, when done WISELY, can shore up the supply chain and reduce do-overs. In a perfect world, you'll never source parts from someone who hasn't made parts like yours before, or isn't trying a new manufacturing method or process. The example of the chrome plated and finish-ground bearing race, as an example, would almost assuredly be ground after plating. That doesn't mean some bold manufacturer isn't going to try and eliminate that grinding step and see if he can "get away with" as-plated surfaces. When some "bold" manufacturer does this, you have a supply chain issue while you await their inevitable late shipment due to a failed 'experiment'.

You shouldn't have to do that. /Shouldn't/.

_________________________________________
NX8.0, Solidworks 2014, AutoCAD, Enovia V5