Inspecting Flat Patterns to Paper Instead of Mylar 3

[stephannelle]

Hello all,

I've been in a debate with a tooling inspector and the quality director at my company (gen aviation oem) about inspecting flat pattern parts to Mylar vs paper. For many years here all the drawings were printed on Mylar and it was policy to inspect undimensioned flat patterns to Mylar as opposed to plain paper due to its dimensional stability and robustness. We no longer have the correct style printer to print on Mylar (separate issue.) I made the argument that for a flat pattern part less than 24"x24" the difference in true drawing scale Mylar vs plain paper would be an almost immeasurable difference, and certainly immeasurable with the eye and a steel scale and even calipers. They are worried that the paper is going to grow/shrink in the non climate controlled shop by such an amount so as to cause problems during inspection. I've found thermal expansion values for Mylar but not for plain paper. Does anyone have thermal expansion rates for plain/computer paper? Has anyone dealt with this issue before? I would like to show them numbers that prove that for "x" sized drawing, a temporary paper copy is acceptable to inspect to.

Thanks,
Stephan

 

[ewh]

How do you qualify "almost immeasurable"?
I have dealt with the issue of having to re-stretch my vellum drawings every morning due to overnight growth, and would often see a difference of .060" over the length of an E size drawing.
General inspection rules should be applied, and if paper patterns are used the tolerances should be tightened accordingly to allow for shrink/stretch (after that factor has been determined).
While it may be a separate issue, an easy answer is to purchase a suitable plotter.

"Know the rules well, so you can break them effectively."
-Dalai Lama XIV

 

[stephannelle]

Ewh, Exactly, that is what I want to quantify. .060" across an E size is quite a bit; more than I would have guessed.

 

[rb1957]

you get the quality you build in ...

if you want to inspect to paper masters, have at it. sure you're bound to accept parts that are out of spec. the problem, if it exists, will become apparent downstream; and that'll mean expensive rework.

why not keep an original master in Al (or whatever) ... ie keep one conforming part as a master ?

another day in paradise, or is paradise one day closer ?

 

[berkshire]

If you used the print within a few minutes, certainly less than an hour and had a 6"x 6" check square you could check with a digital caliper. You might get away with this. You can also print test patterns on paper for caliper check, so that you can see how badly they shrink or grow as the weather changes. As you are well aware, this is not a recommended practice, however on an un- dimensioned print you might be able to produce a flat pattern for a part that is acceptable. The final proof will be if the part can be formed to dimensions that exist on other drawings.
B.E.

You are judged not by what you know, but by what you can do.

 

[ewh]

The type of parts will make a difference. Edge distance for fasteners is important, and is often pushed to the limit in the CAD file. A flat pattern off by a few thousandths could easily eat up the allowable hole location tolerance and come back to bite you.

"Know the rules well, so you can break them effectively."
-Dalai Lama XIV

 

[SAITAETGrad]

Even if you are right, is it going to be worth it in the end? Quality is process driven so it should not be a surprise when they balk at going rogue. Your approach could do some serious damage to your working relationship.

 

[WKTaylor]

stephannelle...

Suggest inspecting non-dimensioned parts ONLY on mylar** that has benn dimensionally verifyied across the background grid.

Mylar is a streched [strain-hardened] plastic sheet. For this reason, it is biaxially stable. Paper is not biaxially stable.

Strain [due to expansion of the microstructure] from humidity will almost inevitably have un-equal differences along the length of the sheet VS across the sheet.... resulting in non-symetrical dimensional deviations. What a nightmare.

NOTE.
ASME Y14.31 Undimensioned Drawings - Engineering Drawing and Related Documentation Practices has the following statement...
9.3 Reproduction Tolerances
The tolerance for reproductions shall be determined by the accuracy required for manufacturing and inspection
of the item, and by the material used for reproduction. The undimensioned drawing shall indicate directly, or
by reference, the reproduction requirements with a note placed near the title block of each sheet, such as FOR
MANUFACTURING AND INSPECTION PURPOSES, THIS DRAWING SHALL BE REPRODUCED ON STABLE BASE MATERIAL.

** Most plastic [Mylar] drawing media is qualified to L-P-519 PLASTIC SHEET: TRACING, GLAZED AND MATTE FINISH which has very strict dimensional stability requirements.

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true.
o For those who believe, no proof is required; for those who cannot believe, no proof is possible.
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion"]
o Learn the rules like a pro, so you can break them like an artist. [Picasso]

 

[tbuelna]

As others noted, I think the issue with plain paper or vellum is mostly due to changes in humidity rather than temperature. I'm old enough to remember carefully taping a J-size vellum to my drawing board so that it was smooth and tight, and then coming back the next day to find it saggy. Since the problem is mostly due to the paper absorbing moisture from a humid atmosphere, you might try taking the roll of plotter paper, unroll a length of it, leave it sitting out overnight so that it will stabilize, wind it back onto the roll, put the roll back on the plotter, and then print your flat pattern on the section of paper that has been exposed to the local atmosphere. There are also some coated types of roll paper which are suitable for use with inkjet plotters, and they are more dimensionally stable than plain paper.

Having said that, in the OP it was noted that the process in question involved a general aviation OEM. As SAITAETGrad noted, manufacture and quality assurance of aircraft components is now entirely process driven. So the inspection technique you are using to validate your flat pattern parts should conform to an existing qualified/approved process, and you should not deviate from this process without approval. Even something as seemingly innocent as changing the material your flat pattern inspection template is printed on (say from mylar to paper) could result in a big non-conformance problem for your company.

Here is a link to a public domain copy of a N-G QA requirements document for sheet metal parts. It includes detailed standards for things like dimensional tolerances of mylar templates.

https://oasis.northgrum.com/contract/miscdocs/QOS-0033.pdf

 

[berkshire]

stephannelle (Aerospace)
Do you have a process specification that would allow you to substitute paper for Mylar ? As I said before this is not a recommended practice.
If you do not, and you did what I said It would be for one piece only , and you would have to get a non conforming part disposition for that.
Do you really want to go that route?
B.E.

You are judged not by what you know, but by what you can do.

 

[gpishko]

Stephan,

It would really benefit you to get these items off vellum. I would suggest having a contractor re-draw them one-by-one as your budget allows and convert them to a dimensioned (albeit loosely dimensioned) drawing. The real issue is... do these need to be inspected at all or was the intent of using the vellum just to give a 'warm fuzzy feeling' that the parts were close to the correct size.

The ANSI Y14 family of specs would be a good reference for learning more about how this should be handled. I would guess your datum structure is not well defined in addition to the issue you are facing. If this is true then the vellum really is a false-positive anyway... but old habits die hard and what was a 'warm fuzzy' may have in time developed into a 'standard practice' at your employer.

The real issue is how long are you going to keep believing the vellum is really helping you - isn't the time and money combined with lack of confidence a risk issue for the company? Similarly if you are using a sub-standard inspection method and getting false-positives why would you resist using a true blueprint with a modern inspection technique like a laser tracker or structured light scanner - which would give you absolute confirmation of compliance?

Best of luck... I have been there before... in today's money-beats-common-sense manufacturing world you need to approach this carefully and get buy in.

 

[tbuelna]

Here's an interesting, and somewhat related, experience I had a few years back. I was hired by Boeing to help construct a 3D CAD model database of 757 airframe structure parts from existing hand drawings. The as-formed sheet metal parts were typically pencil on vellum, but the flat patterns (including mold lines, bend lines, trim lines, etc) were ink on mylar masters. To expedite the process, all of the mylar flat patterns were digitally scanned and converted into a 2D .dxf format. We imported these 2D files into 3D CATIA V4 and used a custom function to select and define features on the flat pattern such as mold lines, bend lines, trim lines, pilot hole centers, bend radii, bend angles, etc. We then ran a CATIA V4 function that automatically converted the 2D wireframe into a formed 3D solid model.

The last thing we had to do before releasing the 3D solid model file was check it for accuracy. Boeing had some fairly strict requirements for these solid model files, and it usually required a bit of manual tweaking to get them up to snuff. But I was always very impressed by how accurate the solid models produced from the scanned, hand-drawn mylar masters actually were. The guys that drew the mylar masters were very skilled.

 

[JNieman]

I skimmed the previous replies but didn't see anyone mention this - if I'm wrong I apologize for the redundancy.

Paper roll plotters also vary significantly in quality and precision. Typically printers used to create mylar prints are made to a standard of precision that's a little higher, in my experience. Your typical HP Deskjet model is not going to be printing things to a precision required for inspection purposes, IMO. Some higher end OCE or special-suited plotter may, but I know the typical 36" roll plotters HP sells at the "budget price" are not that great with pinpoint accuracy of plots.

So even if the material was stable enough... is your plotter even putting the lines in the right place?

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

 

[tbuelna]

JNieman-

Your point about plotter accuracy has merit. But there is a more fundamental issue to consider. The OP mentioned the paper plots were going to be used as an inspection device to validate sheet metal flat patterns produced by a general aviation OEM. Most aircraft OEMs are now AS9100 certified, which means the tools/gages/fixtures/templates they use for QA purposes must be qualified and calibrated. And it was not clear how AS9100 conforming qualification and calibration could be achieved/maintained with paper inspection templates produced on a plotter.

 

[Lescombes]

I can't add any real value to this thread (as others have) however I will say that the OP is not alone: I'm also at a gen aviation OEM and have had this and many similar arguments with those whom wish to inspect to paper drawings rather than a dimensionally stable material.