Technical Drawing/Drafting

I see this at work sometimes and it could be due to inexperience on the drafter's part or maybe a lot of the parts just don't need that level of complexity. Also hopefully if a manufacturer contracts parts out for machining, for example, their QC/QA people would pick up on any defects upon receipt.

I've seen the opposite problem too, where a drafter will add a bunch of geometric constraints that don't appear to matter. For example, too many decimal places (trailing zeroes with no tolerance given), or parallelism across a 1" part.

I don't know that they're going away so much as things are evolving more towards a model-based approach. ASME Y14.41 (model-based definitions) covers a lot of the requirements in that regard, though it is in constant revision as the standard evolves.

I've always felt that... if you're machining from a model, inspecting from a model and carrying all the pertinent data in the model, then a 'print', be it digital or paper, just becomes a document people not involved in those processes would have need for.

In the model-based approach, the GD&T is still carried within the modeling environment and is still necessary. All of our in-house stuff carries PMI (Product Manufacturing Information) per Y14.41, but that's just us. The mother ship doesn't use it at all yet, but then, we're a lot leaner, with fewer hoops and barrels to jump through.

Typical example: primary-secondary-tertiary established, be it casting or finished part... some form that can only be described as a spline curve on a surface out somewhere in space. Why on earth would anyone want to define that with anything but with a profile feature control in relation to A/B/C? Yes I see attempts to drive dimensions to construction data all the time, as if that will somehow be useful to anyone. :)

I'll have to experiment if the annotation view carries over to a step file but I see your point.

Yeah, check your step export settings. There's usually an option to include the PMI, or whatever your flavor of software may call it.

"I've seen the opposite problem too, where a drafter will add a bunch of geometric constraints that don't appear to matter. For example, too many decimal places (trailing zeroes with no tolerance given), or parallelism across a 1" part."

Yes, I see that a lot as well, especially with young designers but, many older ones too.

This goes back to a lack of training. When I checked designs, I'd typically type something to the effect of, "wrong type of tolerance zone" or "this needs a perpendicularity control", etc.. What I found was that, they'd make the change and push the work back through, at which point I'd sign off on it and off we go.... until the designer's next design... and here I am staring at the same problem again. Why? Because I told them what to change and was never taking the time to teach them 'why'.

Kind of like your significant other nagging you to fix something; you get it done just so they'll leave you alone, so you can get on to the other things you need to do! :)

So I started trying to train people on the fundamental level, using examples that reflected the work 'we' were doing. Datum assignments being the most important and I found that, once they had a good understanding of functional datum assignment, the rest got a lot easier.

Every young designer should be familiarized with manufacturing processes and more importantly, their capabilities. The International Tolerance grades and their correlation to typical machining processes are a good place to start.

Lack of understanding there, leads to a lot of crazy tolerances that aren't feasible for manufacture.

http://www.cobanengineering.com/Tolerances/FundamentalDeviation.asp

I agree, a lot of things are changing in the 21st century but that's not necessarily a bad thing. Many of the drawing standards out there today were developed back in the day when all you had to go on was a 2D drawing, now that 3D modeling has become so much more prolific the relevance of such things is diminishing.

Additionally keep in mind that in engineering we often run into the case where something done a certain way just because that's they way its been done for 50 years but it's not necessarily the BEST way. That is the case with many old drawing standards and as a result many companies (like the one I work for) just create their own standards.

Another thing is that many drawing standards were developed not just for the fabricators benefit but to make it easier for the drafter to create a drawing from whatever bits of information or drawing on a napkin they happed to be working with. Now that drawings are computer generated from a model the engineer makes themselves its easy to pick out the optimum way to show things, and that way isn't always the same way as it would have been done in 1965.

The point of a drawing is to clearly communicate the intent of the engineer to the people making the part, if the drawing doesn't do that it's failing at its one and only purpose. The thing I get onto people about the most is drawings that are convoluted and populated with too much information scattered about in weird places where the fabricator has to play detective to figure out what information he needs.

A good drawing should have only as many views and dimensions as is necessary to build the part. A fabricator should be able to look at your drawing and immediately know everything they need to know to make your part. There should be no hunting around or referencing back to other things or eye squinting or having to look up what some obscure term means, CLEAR-ACCURATE-EFFECTIVE-SIMPLE those are the principles of a good drawing, whatever standards you use.

"But that's the way we've always done it" :)

Just recently had a meeting at our company where I presented a case where acceptable parts by the old standard (pre GD&T) produced parts that did not meet the design requirements. Many due to a lack of form controls.

I was then told that, had I sent the GD&T part out to another shop, they wouldn't understand what they were looking at. A bit presumptuous perhaps but I've never felt it my job to worry about what people may or may not know. If they don't understand something? It may be time to educate one's self.

The print is essentially an insurance policy. It says to the manufacturer, if you build parts for us they must meet these specifications and if they don't, we have the right to reject them. They are free to bid or pass as they see fit and as I see it, if you don't have the knowledge, metrology, whatever to meet the requirements... we don't want you doing the work anyways.

I also see this. But I think this is mostly a good thing. Modern machinery is now days so accurate that we expect a certain quality from them. Also If a companu does not have these machines, they mostly also lack the knowledge to understand the standards.

But big companies still use them a lot, so in automotive or other high quality demanding sectors there is not really a downturn.

I agree....I used to check a lot of drawings and red-line drawings with G D& T and make people cry. They didn't understand it or the manager would come by and tell me that they don't use that shit here.

We have lots of clients that just don't open 2D drawings anymore... I don't know if it is because they don't know how to read a drawing, or lazyness or anything else... But what also bugs me is the fact that when they send us a 3D model to make a mold project, they also send a 2D drawing with lots of pointless tolerances and they want the 3D model to fit them (more than 80% of the 3D dimensions aren't following the 2D standard they require, which forces us to change and adapt their 3D product to the 2D drawing before starting the mold project, consuming a lot of extra time).

With regards to teaching, one of the reasons for the lack of the skills one might associate with the drawing environment is that the educational ethos in the UK is based upon students developing their own learning and in some circumstances the criteria for assessment. Also as education within the further educational sector, that is to say for apprentices and students over the age of 16 is now a business, outcomes are more linked to successes and time. So in the past a full time diploma course would have been 32 hours per week of taught sessions, in the modern era this equates to 16 hours per week of self discovery whilst being guided across a broad range and to meet a target. This has the effect of ultimately giving the learner an insight into the information rather than and in depth knowledge. This is then expected to be topped up when they enter their chosen profession, as which standards do we teach to, and to what level whilst trying to predict the future work of a student.