Flat Patterns in type design drawings. 3

[Cmott]

Flat Patterns in type design drawings.
Does anyone do this? For me I think it is a manufacturing process and should not be on a type design drawing, any thoughts on this?
I work at an Aerospace manufacturing company.

 

[KENAT]

What drawing standard do you work to?

Per ASME Y14.5 etc. you typically detail the finished part without regard to the process, except where the process directly impacts end part function.

So you don't normally show flats of sheet metal parts on the engineering drawing.

Posting guidelines faq731-376

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

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

 

[Cmott]

Yes we use Y14.5.

Thank you for backing me up on this. New owners and they want to make it easier for the shop floor and add flat patterns to drawings.
My argument is that a flat pattern is not type design and only the final product should be on the drawing.

Would you happened to know what section of Y14.5 I could find that info?

 

[KENAT]

I'm paraphrasing a bit but section 1.4 of my -94 version, especially 1.4 c & e, maybe n - sort of.

"Each necessary dimension of an END PRODUCT shall be shown. No more dimensions than those necessary for complete definitions shall be given. The use of reference dimensions on a drawing should be minimized."

"The drawing should define a part without specifying manufacturing methods...."

"Dimensions and tolerances apply only at the drawing level where they are specified..."

If you lose the battle to your management, at least make all the dims on the flat pattern reference only so that it's clear the formed part dimensions are what matters.

Posting guidelines faq731-376

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

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

 

[tbuelna]

KENAT- What you said is basically correct. But you must also remember that a sheet metal flat pattern blank is often considered an end product, similar to a raw casting or forging. Thus it would have its own engineering drawing defining any requirements for the flat pattern blank product that gets delivered and inventoried. One vendor might produce the trimmed flat pattern blank with any features such as tooling holes that can be added prior to forming, along with heat treatment of the blank. The drawing for the finished formed part would call out the p/n of the blank as the raw material.

 

[KENAT]

Sure, but it should typically be a different PN - even if by '-' number.

This is where the "Dimensions and tolerances apply only at the drawing level where they are specified..." comes in.

Posting guidelines faq731-376

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

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

 

[verymadmac]

The OEM's I have worked don't know what ASME Y14.5 is.

From what I have seen the best approach seems to be (well for small companies anyway), TC drawing stack is inspection data which controlled by the design office, while flat patterns, dxf's and the like are part of the manufacturing data that forms the work instructions (the detail of which verys greatly depending on the work force skill sets). While these may be produced by design office, they should be under the control of the manufacturing group (they often are not, at small OEM's which is how the flat patterns end up in drawings).

The other point not covered is that the flat pattern is typically these days a just another step in path to creating tool paths. It's tool paths that matter with regard to revision data control.

 

[debodine]

14 CFR 21.31 says this about type design.

§21.31 Type design.

The type design consists of—

(a) The drawings and specifications, and a listing of those drawings and specifications, necessary to define the configuration and the design features of the product shown to comply with the requirements of that part of this subchapter applicable to the product;

(b) Information on dimensions, materials, and processes necessary to define the structural strength of the product;

(c) The Airworthiness Limitations section of the Instructions for Continued Airworthiness as required by parts 23, 25, 26, 27, 29, 31, 33 and 35 of this subchapter, or as otherwise required by the FAA; and as specified in the applicable airworthiness criteria for special classes of aircraft defined in §21.17(b); and

(d) For primary category aircraft, if desired, a special inspection and preventive maintenance program designed to be accomplished by an appropriately rated and trained pilot-owner.

(e) Any other data necessary to allow, by comparison, the determination of the airworthiness, noise characteristics, fuel venting, and exhaust emissions (where applicable) of later products of the same type.

Like many issues with the FAA, the correct answer is "it depends". If the flat pattern is considered by the FAA ACO or designee authority that is approving the data to be required under any paragraph of this regulation or the sub-chapter(s) applicable to the article in question, it is part of the type design. If that authority does not consider the flat pattern to be required by any paragraph or sub-chapter of this regulation, the flat pattern is not required to be on the drawing, but it is also not forbidden. However, placing non-required information on a type design drawing can severely limit you with future changes if it turns out that the non-required information conflicts with a later design change.

You may want to have a conversnation with the FAA/designee who approves the design. Always be cautious, however. Asking any question directly of the FAA or designee always carries two major risks:

1. The FAA/designee may decide detailed investigation is needed into the issue and there may be unintended consequences to your design and/or extra effort required to respond.
2. If you are not the individual in your organization officially designated to be the liaison with the FAA/designee, you will at a minimum irritate the person who is the official liaison. Most organizations with a long history of working with the FAA tightly control who talks directly to them...and with good reason.

 

[rb1957]

my 2c ... the flat pattern is an manufacturing easement. It is an intermediate step in producing the finished s/m part. The engineering definition, the type data, is the finished part. Anything along the way is "helpful" information. Inspection inspect the finished part to verify that it conforms with the engineering definition. Once you give out flat patterns you can expect all sorts of horse poop when the finished part doesn't conform ... "but I used the flat pattern", "well then, I guess the flat pattern was wrong". Accounting for sheet metal bends is not straight forward.

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

 

[debodine]

rb1957 is right on target with the flat pattern being an intermediate step and not part of the final configuration definition of the type design. He is also colorfully correct that the outcome of non-required data on a type design drawing will be no more pleasant to work with than "horse poop".

 

[berkshire]

Flat patterns are a useful tool in the manufacturing process, however they are not the end product unless the end product is flat, the end product is the folded sheet metal part to whatever dimensions and tolerances are on the final drawing.

The bugaboo with this is that the construction of a sheet-metal part is very often process driven, and part of that process is making a flat pattern for that sheet-metal part, the problem gets worse when one person makes the flat ( pre bent part.) it gets inspected, and another person folds bends or otherwise processes the piece to its finished state. The more steps in the process , the more chances for error . A well meaning draughtsman may have specified what he/she thought were generous tolerances for that part, to find that accumulative errors in the flat patterning process, have used up all of those tolerances, and that now the person forming the part has a precision part on their hands with only a thousandth on an inch either way between acceptance and rejection.

rb1957 put it very eloquently : " Once you give out flat patterns you can expect all sorts of horse poop when the finished part doesn't conform ... "but I used the flat pattern", "well then, I guess the flat pattern was wrong". Accounting for sheet metal bends is not straight forward.
B.E.


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

 

[tbuelna]

berkshire makes an excellent point about modern aircraft manufacturing relying heavily on controlled processes for quality assurance (ie. AS9100). The certification process for an aircraft design usually includes a thorough FMEA. And for many critical components (including formed metal structures) the FMEA considers quality control used at every step of the component's manufacture, including sourcing of the raw material. Any part of a component's manufacturing process that is used for the FMEA cannot be altered after certification without approval. For example, if the profile tolerance of a sheet metal flat pattern blank is considered in the FMEA, and the profile dimensional accuracy for each flat pattern blank produced is ensured by using a controlled/validated process rather than inspecting each individual part, then even the CNC program used to cut the blank might require approval for any changes.

 

[JeniaL]

we do show flat patterns on drawings. guys who are doing routing have to verify flat pattern dimensions. how to do that if there's no dimensioned unfolded view on drawing?
i'm working for aerospace. not Boeing but IAI. take a look at all old Boeing projects. there's no way they don't show unfolded view.
if you're doing just a design so go with DPD. when it comes to production show everything on drawing what is needed for production starting from flat pattern and ending up with end product.

 

[ewh]

Do you also include inspection balloons on the drawing?
As has been noted, depending on which industry or customer standards have to be met it is often best to document the flat patterns as separate manufacturing aids, not as part of the type design (the same way QA usually creates their own bubbled drawings).
To add flat patterns to a drawing is to document a process; type design drawings should define a final, deliverable product, not the steps taken to get there.
One way I have encountered to get around this is to include such information on a separate drawing sheet which is named something other than a sheet number and does not get published (or released) with the official drawing (this was not an aerospace company).

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

 

[tbuelna]

JeniaL brings up a good point to consider regarding the current approach used for engineering definition of parts. Many companies have adopted the use of Model Based Definition (MBD) where all requirements for the part are embedded in a configuration controlled digital model file, and no drawing is produced. Other companies use a Reduced Dimension Drawing (RDD) in conjunction with a configuration controlled digital model file for engineering definition of parts. In the case being discussed of a formed sheet metal part, the digital model would include representations of both the flat pattern and the formed part. But even with the RDD approach there would not be a flat pattern shown on the greatly simplified drawing.

Unfortunately, the first thing many people (especially management) think of when they hear the terms MBD or RDD is how much money it will save by eliminating most or all costs associated with engineering drawings. But if you think that's the case, take a look at this list of requirements for Boeing suppliers working to their MBD process.

Lastly, you might take a look at this short video tutorial that shows just how incredibly quick and easy it is using CATIA V5 to create a flat pattern from a formed sheet metal 3D part model, and then add the flat pattern to a 2D drawing.

 

[berkshire]

EWH,
Back in the 90s I was working for Convair, and was part of discovering a booby trap in the method you describe.
One way I have encountered to get around this is to include such information on a separate drawing sheet which is named something other than a sheet number and does not get published (or released) with the official drawing
The company used a Yoder multi station roller to make Z sections for fuselage formers. They used a drawing called a "pre- form " which detailed the section to be rolled. After rolling these sections were heat treated then stretch formed to make the formers detailed on another drawing.

This system worked well for many years. Then the company decided to sell the roller and farm the job out to another company. We got the first of the " preforms" in and they were spot on to the " preform drawing". The problem was that when they were stretch formed, they were too small. After tearing our hair out for a couple of days we decided that there was no way the parts we formed before could have been made to that drawing. My supervisor and I after a discussion went over to the shop where the Yoder had been and discovered the foreman's desk. In it we found a marked up copy of the "Pre form " drawing about 20years old showing all of the dimensions increased by .1", which coincidently was about the amount the part shrunk when it was stretch formed.
The preform drawing was revised to the dimensions on the marked up print and the parts started coming out correct again. Apparently that pre form drawing had never been checked to the part, and as long as the finished parts were right nobody cared.
B.E.

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

 

[tbuelna]

Consider this situation which is somewhat similar. I design aircraft mechanical systems such as gearboxes. And these products often use castings and forgings. The raw casting or forging is procured under its own part number and engineering documentation. It then gets machined according to another set of engineering documents and is re-identified with a new part number. After that it has some additional parts permanently installed which makes it a non-separable assembly, and it is assigned another part number. The non-separable assembly that uses the casting is part of a flight critical system, so every part number in the design including the casting must be configuration controlled. The casting is produced using a controlled/validated process and tooling. After the end item it is used on becomes certified, the process and tooling used to produce the casting become a controlled process, and no changes to the controlled process are permitted without approvals. You cannot even change the vendor that supplies the qualified casting.

 

[3DDave]

One supervisor I had suggested all the engineering drawings could be thrown in the trash because the MIS's were sufficient. That way no ECOs were ever needed; just make changes on the factory floor.

 

[berkshire]

3DDave (Aerospace)
Dave, if you do that , how do retain traceability?
B.E.

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

 

[MikeHalloran]

When we adopted JIT, the Manufacturing Instruction Sheets, as marked and initialed on the factory floor, became the absolute grand masters. The affected engineering drawings were corrected to agree with them as time allowed.

It seemed backward at first, and I made a hell of a fuss about it, but it worked out okay.

We were making in vitro medical devices, where the FDA surveillance is not so intense as for in vivo devices, and probably a little less intense than FAA scrutiny of aircraft, but it was still serious stuff.



Mike Halloran
Pembroke Pines, FL, USA