Manufacturing to CAD -without Drawings Yes/No 10

[Camille-starr]

I’m not new but an old user that can’t retrieve his login details due to emails change etc.

So, I have always produced full drawing packages with my CAD models and supplied them to the manufacturer for quote and production. I understand that most of the machinists I use programme from my models and use the drawings to check details and then the drawing is used for checking once the item is received.
The drawings I produce are full on manufacturing drawings with all details, I’m an old school draughtsman.

Am I working in the past, can we start to rely more on the model and lest on the drawing?

How do you handle this and what if any standards do you work to?

 

[Metalfixer]

I'm retired now, but were I used to work, Drawings were needed for every
part and operation eg. grinding, plating, heat treat, even raw material.
Drawings were considered legal documents, also needed for engineering changes etc.
signed by the engineer

 

[dgallup]

There is a strong movement toward model based definition but it takes just as long to create a complete set of requirements in the model as to make a drawing and only trained people with the correct software can see it so I see absolutely no value in it.

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

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.

 

[cr7]

... with the correct software can see it so I see absolutely no value in it.

I can open attached file in Adobe Acrobat Reader DC, so I'm not sure what you mean?

Reference

 

[drawoh]

Camille-starr,

MBD (Model Based Definition) will be a functional concept when everybody has the CAD software and portable devices that can manage the files. It will help if a Free[ ]Software file format replaces all the proprietary file formats out there.

A lot of MBD out there is laziness, and an admission that people are too dumb to prepare proper drawings. If you can provide your fabricators will good drawings, they will charge less. Good documentation is part of your DFMA strategy.

--
JHG

 

[Tmoose]

I understand the notion that the value of a CAD model if a machine tool can be very quickly "programmed" right from the 3D part file.

Maybe this is just my inexperience ins such matters, but the efficient handling of tolerances concerns me some. Making everything +/- 0002" is way overkill most of the time, but not good enough sometimes too.

 

[MJCronin]

Drawings become part of both the Contract legal record (work scope, compliance with client demands) as well as the historical record of what is drawn. Unique client requirements often are found on the notes to the Contract drawings.

In the future, as the component is improved, discussed and reviewed paper drawings are dug out of project files and are reviewed in meetings.

There is nothing to prevent a company from keeping information in both a physical and electronic format.

.... and I am suggesting that there is not a single solution that would suit everybody....

I do not view the final step of turning 2D and 3D CAD drawings into paper drawings as much of a burden on industry



MJCronin
Sr. Process Engineer

 

[drawoh]

I understand the notion that the value of a CAD model if a machine tool can be very quickly "programmed" right from the 3D part file.

I missed this one when I ranted. There is more than one way to manage tolerances on the scale model.

[ul]
[li]Model to nominal size?[/li]
[li]Model to median size?[/li]
[li]Let some CAD monkey replaced the <DIM> tag on the drawing with the "corrected" dimension?[/li]
[/ul]

--
JHG

 

[swertel]

ASME Y14.41 Digital Product Definition Data Practices
ASME Y14.47 Model Organization Practices
ASME Y14.46 Product Definition for Additive Manufacturing
ASME Y14.31 Undimensioned Drawings
NIST.AMS.100-1 Summary Report on the Model-Based Enterprise Capability Index and Guidebook Workshop
MIL-STD-31000 Technical Data Packages

Were these the types of standards you were looking for?

--Scott
www.aerornd.com

 

[dgallup]

I can open attached file in Adobe Acrobat Reader DC, so I'm not sure what you mean?

That is not the authoring CAD system file and my Acrobat Reader warns against opening it.

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

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.

 

[SwinnyGG]

I do not view the final step of turning 2D and 3D CAD drawings into paper drawings as much of a burden on industry

I wouldn't describe it as a 'burden' necessarily- but it adds significant drafting/engineering time.

My last employer I was pushing to move to complete 3D information transfer without paper drawings at all. This meant for any given design the design review would be completed in 3D, released to the machinist in 3D, checked against 3D when received, etc. This required a few things to be in place to make it work:

1) Internal QC needs the ability to access 3D data, and the training on how to use it correctly to get the information they need

2) You need to have vendors that you have a high level of confidence in with regard to their ability to translate 3D data into tool paths without drawings as an intermediate medium. Shops that are capable and willing to do this on complicated parts are not common in my experience.

3) You need management that is either comfortable not looking at detailed models at all (because usually they won't have licenses for the tools) or is comfortable looking at things in 3D during design reviews.

4) Your drafting staff (or you if you do it yourself) needs to understand how to incorporate things like surface finishes, tolerances, etc into models so that all required information is clearly communicated to the machinist.

This process, if you can get it right, yields several benefits:

1) In the one-off, quick-turn world where I used to live, not creating paper drawings can save you multiple days on large/complicated/intricate parts and assemblies; in that world, this is very significant.

2) Eliminating drawing review and transferring that manpower into design review in the 3D stage pays dividends. You will spend less time fixing things like 'this callout overlaps this dim line' and more time evaluating things like 'can I simplify this part and shave weight/cost/machining time'. In other words, removing the second check which comes at the drawing release stage means you have more time in design review. In practice this resulted in better prototypes and final designs for us.

3) Change control is extremely easy when there's one file, which is live updated when a change is made and instantly viewable by all stakeholders. There's no individual drawings to revise, and no paper drawings in the field to hunt down, collect, and replace. This again saves everyone time.

 

[mfgenggear]

just in my niche in the world, rarely is manufacturing straight from the 3 d model, but it helps to decipher geometry.
but my processing is broken down for rough turning , machinability hardening, semi finishing turning and milling allowing for grind stock, and other factors.
some body has to draw it out either on cad or print it it on paper. for all manufacturing, coatings, or special information in the related specifications.

edit: also it is a very important to control configuration control, manufacturing changes, to either make better parts, and correct mistakes.

 

[JohnRBaker]

I don't know if this is still an issue or not, but back about 20 years ago one of our customers was using the 3D model to completely manufacture prosthetic implants, such as knees and hip joints, going directly from the CAD model to the hard-part was the most efficient route to go in this very time-sensitive industry segment. However, the FDA required that a paper drawing be included in the patients file documenting the implant, however, they had 30 days or so to get the final information together, so our customers would simply design, manufacture and then document the implants, in that order.

Also, I can remember when our parent company, McDonnell Douglas, when they won the F-15 contract (which was before I was working their, but I heard all the stories), it was a pretty much a sole-source contract for the air-frame (probably the last one like that as after that it was pretty much policy to get as many companies involved as possible for political reasons). Anyway, since most all of the air-frame parts were being both designed and manufactured by McDonnell Douglas, they avoided making fully detailed drawings, depending on the 3D CAD models as much as possible. However, to cover themselves with the Pentagon, they included in the contract a clause that if the government ever needed an actual paper drawing for any reason, they had to give MDC 48-hour notice before a print would be delivered, long enough to produce that needed 'drawing'.

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without

 

[drawoh]

1) Internal QC needs the ability to access 3D data...

MBD is way easier when fabrication is kept in[&#8209;]house. If you subcontract, your vendors must be able to read it. You need to exclude vendors who do not adopt your preferred CAD technology. Sending out 2D[&nbsp;]PDFs plus 3D[&nbsp;]STEP files works well, but the STEP files do not contain tolerances.

2) You need to have vendors that you have a high level of confidence in with regard to their ability to translate 3D data into tool paths without drawings as an intermediate medium.

Wrong! You need vendors who are confident in you. They need to be confident that when they fabricate pieces to whatever data you sent them, that you will accept the parts and pay for them. If your data is crap, either they refuse your business, or they factor the expected drama and free re[&#8209;]work into their prices.

3) You need management that is either comfortable not looking at detailed models at all (because usually they won't have licenses for the tools) or is comfortable looking at things in 3D during design reviews.

Communicating with management is your problem, not theirs. If management knows that is good for them, they will remember that if you cannot explain stuff in clear language, it probably is because you do not understand it. The 3D[&nbsp;]model is an excellent tool for explaining how stuff works. I see no reason why this cannot be your CAD[&nbsp;]model running on your computer. Presenting assembly drawings and assembly process instructions at the design review is good DFMA[&nbsp;]practise.

4) Your drafting staff (or you if you do it yourself) needs to understand how to incorporate things like surface finishes, tolerances, etc into models so that all required information is clearly communicated to the machinist.

Why is this any different than 2D[&nbsp;]drafting?

This process, if you can get it right, yields several benefits:

1) In the one-off, quick-turn world where I used to live, not creating paper drawings can save you multiple days on large/complicated/intricate parts and assemblies; in that world, this is very significant.

I can create 2D drawings of most parts in well below an hour. Since I am the designer and I know how the parts work, and I am skilled as ASME Y14.5, my drawings are correct and of high quality.

I have encountered parts where it did take significant time to draw. These parts were complex and critically important, and again, we needed quality specification.

2) Eliminating drawing review and transferring that manpower into design review in the 3D stage pays dividends. ...

An awful lot of people here dream of any drawing checking and review. My basic assumption is that if the drawing is of high quality, it is attached to a good model. Either way, you are checking the same information. There is a lot to be said for a piece of paper and a highlighter.

3) Change control is extremely easy when there's one file, ...

I regard the 3D[&nbsp;]part and the 2D[&nbsp;]drawing as one entity. Unless the proposed ECR is to correct the spelling of "discombooberate" on sheet[&nbsp;]3 of your drawing, most changes affect all sorts of other stuff.

I regard the 3D[&nbsp;]CAD model as a mechanical engineering and design tool, and I would rather it stayed in the design office. Releasing dumb PDF and STEP is good practise.

--
JHG

 

[TheTick]

Ultimately, any part is as good as the relationship and communication between design and manufacturing. Paper or virtual, there are always ways to not say what should have been said.

 

[SwinnyGG]

a lot of stuff in response to my post

I didn't say that pure MBD paths for transfer of information are the best for everyone- but they worked extremely well for us on very intricate mission critical parts.

You don't have to agree, but we did it successfully and it worked very well.

 

[mfgenggear]

SwinnyGG

it can work if the procedures are set up that way, but ultimately some one has to lay out all the operations, cad models of each step, and some one has to order material, outside processing and on.
so the burden of that will fall on manufacturing. and if the cad model is wrong it will be more difficult to catch it before it does damage. paper now PDF files can be messy as well.
being old school I prefer paper. cad models all it takes is pick a wrong entity to make a mistake. it has been rare to make a full cnc program from a full model with out altering it. breaking down the machining.
if the flow down of the requirements have to have some type of drawing or specification. I take it as laziness on the engineering department. no offense

 

[SwinnyGG]

if the cad model is wrong it will be more difficult to catch it before it does damage

I disagree with this point- in my experience detecting and correcting design flaws is no more or less difficult than on paper. For certain types of features it's much easier, but on the whole you get out what you put in as far as effort.

That isn't to say that every company in the world would be better served to eliminate paper drawings entirely. Clear definition of process, and clear dedication of manpower to design checks are mandatory for this process to work- just like they are for successful delivery of designs via 2D drawings and standard practice.

 

[mfgenggear]

I disagree with this point- in my experience detecting and correcting design flaws is no more or less difficult than on paper. For certain types of features it's much easier, but on the whole you get out what you put in as far as effort.

That isn't to say that every company in the world would be better served to eliminate paper drawings entirely. Clear definition of process, and clear dedication of manpower to design checks are mandatory for this process to work- just like they are for successful delivery of designs via 2D drawings and standard practice.

sure when you are using your program and apps, that is not the same for the manufacturers, if all the processing done in house, but what about the suppliers.

 

[JohnRBaker]

One big advantage of a 3D solid model is that, at least for most systems, it will be 'manifold', which means that it represents a valid physical object. Granted, it might still be impossible to manufacture using traditional metal removal techniques, but at least it represents a realizable part.

John R. Baker, P.E. (ret)
EX-Product 'Evangelist'
Irvine, CA
Siemens PLM:
UG/NX Museum:

The secret of life is not finding someone to live with
It's finding someone you can't live without