Speaking of GD&T...

[deadeye]

How many of you use it? We're considering it but nobody here has had much training on it. Has anyone found some free (hee hee) intro training literature on the basics just so's we can get a taste of what we're up against?
Tanx,
Deadeye

 

[ctopher]

Everyone should be using it and know it (drafting/designers/checkers/machinist/engineering). Don't know of any free stuff. We had someone come to our company and train us. It's a little confusing and takes a while to understand it. Your machined parts will be more acurate and consistant. good luck

 

[SBaugh]

You should have them send you to a college course on it. I learned alot on it when I was in college. I still remember a lot of it. I did get a refresher course a a few years back. IMO - To me it's like riding a bike, it might take a bit getting used to it again, but I will soon be back to where I was at.

Regards,

Scott Baugh, CSWP
3DVision Technologies

http://www.3dvisiontech.com

http://www.scottjbaugh.com

faq731-376
When in doubt, always check the help

 

[StarrRider]

I seem to be missing something. I'm not sure what the abreviation GD&T stands for. You are not referring to Geometric Dimensioning & Tolerancing are you? If you are, then ANSI is the place to go.

Lee

 

[deadeye]

Yes I am referring to Geometric Dimensioning and Tolerancing. I had considered posting this to ANSI but I wanted a view from the trenches, and I want fellow SW user opinions. Last I touched it was 25 years ago and I couldn’t even begin to explain it most of it now. There’s been some ‘drafting-bashers’ around my company that feel it’ll fix whatever mechanical mysteries our production department seem to think exist. I’m all for GD&T, but for the right reasons, and with ample thought into the considerable commitment it is. Some think it’s as easy as having Drafting just start using it without anyone else in the plant learning it. Thanks for your opinions.

Deadeye

 

[SBaugh]

Everyone will have to understand it or it would kind of pointless to put it on your drawings!

Welcome to the managers side of it I guess. I just don't understand how people can think that by adding something to a drawing is going to magically fix their problems.

This is in no offense to you in anyway Deadeye Just trying to understand how some managers think sometimes.

GD&T would help out design or production flaws, but if you and others don't understand GD&T then it's pointless. You need to inform them that GD&T is not going to magical fix to their drawings. They will have to fork over some cash to get some of the important people trained, then they can train the others below them there at your facility.

Regards,

Scott Baugh, CSWP
3DVision Technologies

http://www.3dvisiontech.com

http://www.scottjbaugh.com

faq731-376
When in doubt, always check the help

 

[MadMango]

You definately need to know how to apply GD&T, or you will be designing added cost into items you out-source for manufacturing.

I've seen shops tack on an additional 10-20% on quotes just because GD&T symbols are on the drawing, forget the fact what the symbols are defining. And vice versa, I've seen drawings with wrong symboly applied, and shops do extra work to meet the specs, but the specs were inadequate.

For GD&T, get (purchase) the latest copy of ASME Y14.5M. If you've got the experience behind, then the standard is a great refresher.

MadMango
"There is no reason anyone would want a computer in their home."
Ken Olson, president, chairman and founder of Digital Equipment Corp., 1977
Have you read faq731-376 to make the best use of Eng-Tips Forums?

 

[StarrRider]

DeadEye

I have used GD&T on and off over the last 20 years in a number of companies. The thing that you should (and I think you do) realize is that GD&T will only help a company that knows how to use it.

Even if a company does know how to use GD&T, I personally think they should be used in moderation and only where needed. There is a time and a place for everything and GD&T is no exception. If a part can be made without using GD&T – then why add it? On the other hand, if the part is critical and you are getting too large a percentage of bad parts, then you should look at why the parts are bad. Adding a couple of geometric tolerances will probably eliminate the problem. It is a judgment call – at the engineering level.

Another consideration is cost. In one company that I worked for, the machine shop manager did the cost analysis for every part. He explained why he increased the cost of each part by $50 per GT. The main idea was that a drawing without GD&T could be given to any machinist, normally without any discussion. He would only give a drawing with GD&T to his best machinists. Even then, there was normally quite a bit of debate about their meaning and often the engineer needed to be asked what he wanted.

I hope this helps.

Lee

 

[EnglishMuffin]

It is interesting that StarrRider says that GD&T increases costs. It is normally touted as being a way to reduce costs, although this is a common tactic when one is trying to introduce anything new, particularly in the US. The original reason it was introduced almost half a century ago (in a newly cooperative post war Europe) was to make drawings readable by people who couldn't understand written instructions in the various languages, and it was at first little more than a pictographic shorthand. Obviously, this was not such a problem in the US, which has conseqently lagged behind in applying GD&T. It has become much more complex over the years, with the introduction of things like MMC etc, which ironically seems to have been a US led development, and is one of the features that is supposed to lead to cost savings. One of the other issues today is that more and more parts are outsourced, often to vendors with whom one has no prior experience. If the drawing is fully toleranced, one has a legal justification for rejecting the part. If you know who is going to make a part, you can certainly get away without using GD&T, as people have for centuries. Sometimes you can get away without having a drawing at all, especially if the person making the part happens to be yourself.

 

[ctopher]

The only place we try not to use GD&T much is sheetmetal. Most sheetmetal mfg do not understand it, and is too difficult to meaure position on thin material. Sometimes we do it if needed.

 

[StarrRider]

EnglishMuffin

I would have to agree with you, but I have been called on too many times to explain what someone was trying to define using GD&T. To say that it can be confusing is putting it mildly. I learned GD&T at Control Data Corp. about 20 some odd years ago. I remember a lot of friendly arguments about how this GT affected that GT - and in some cases they completely nullified what their author intended. Fortunately, our department manager was on one of the ANSI boards (no idea which one now) and he was our final reference. Most companies are not that fortunate.

I have also witnessed a half dozen machinists argue for an hour while trying to figure out what a couple of simple tolerances actually meant.

I hope this helps.

Lee

Consciousness: That annoying time between naps.

 

[deadeye]

Hey StarrRider, interesting comment "If a part can be made without using GD&T – then why add it?". How do you determine when GD&T is necessary? I've always considered it a clearer way of dimensioning/tolerancing a print. Always applicable as long as users can understand it, whether you're designing shoeboxes or rocket engines.
Deadeye

 

[JNR]

Hmmm.... lots of interesting comments. WE have some very complex geometry with very complex compound angles, in machining, etc. We have to use GDT. But if it is applied correctly it does reduce costs. It certainly allows you to tolerance things much more appropriately so you get what you really need and it fits without paying for additional tight tolerances caused by traditional drafting methods. There are few gotchas, but most of us don't need those odd-ball GDT types. Remember that it both defines your design intent and the method of inspection. One common pitfall is to think the once you start putting GDT on a drawing every dimension and profile has to have it - not so!!!! Use it where youo need it. One other thing that is interesting - there is a trend these days to let the CAD file define most of the part and only dimension critical features (in fact now some of our parts specially casting are so ugly that you would never be able to dimension them by tradition methods anyway). The GDT profile tolerances can work very well in conjunction with this methodology.

3/4 of all the Spam produced goes to Hawaii - shame that's not true of SPAM also.......

 

[ctopher]

Understanding how a part is inspected and machined also helps how to use GD&T on a dwg. i.e, where the part is held in the fixture to measure/machine another feature.
just my 2 cents worth

 

[StarrRider]

DeadEye

That is a very good question. I don’t add GD&T to anything unless it has a critical feature and I am worried that I may not get the result I need unless it is added. To me the issue is the part and the process that will be used to create the part. If the part is small and complex with several critical features then GD&T is probably needed. This is especially true if I do not know the supplier or what process will be used.

On the other hand, adding GD&T to a part to ensure critical features that would be acquired naturally - doesn’t make sense to me. As an example: if you are designing a hub with several flanges, do you need to specify that every flange has to be perpendicular to the center of the part, that the flanges need to be parallel to each other, or that the flanges should be concentric? You would if it was a molded part, but not if it was created on a lathe.

I have found that many people using GD&T almost invariably over define everything. There is a natural tendency to say – “but what if xxxx” – and add another dozen tolerances to control it. This seems to be a never-ending story, but it results in simple parts with 4 or 5 datum and a drawing that begins to look like a GD&T Christmas Tree.

When I made that statement yesterday – I was mainly thinking about modifying drawings to reduce failure rates. I have seen a lot of problems caused by tolerance stack-up and most of the time adding a couple of simple GD&T statements to one or two parts will eliminate the problem without tightening the tolerances on everything. So the question is, should the GD&T statements have been used in the first place? Possibly, but I doubt if the solutions I’ve found would have been originally seen. It is more likely that the parts would have been more tightly controlled and more costly as a result,

I was not thinking of the simple GD&T callouts (like Flatness, etc.) yesterday. Most of that should be added without hesitation and using a note to avoid GD&T doesn’t make sense. Further, I don’t think that there is a machinist out there who doesn’t understand the simple stuff.

Oh yes - the manager that I mentioned. I don’t know what he was before he was the machine shop manager. He might have been a machinist, but then again, maybe not. I have to admit that I have run into people who had strange ideas when it came to pricing parts.

Lee

Consciousness: That annoying time between naps.

 

[JNR]

If do don't have enough staff to inspect parts adequately, why bother to tolerance the drawings in the first place, because you are not verifying them, you are just using the parts as is?

I still say that GDT should REDUCE the amount of unecessary tight tolerances on most parts. Often a simple profile tolerance works wonders to get good castings while only specifying the critical features in more detail. I agree that you should not spray GDT all over the place for the sake of it - but then you should not spray "traditional" tolerancing in the same way. We definitely have less problems getting parts made (outside vendors mostly) to spec. using GDT and the costs are lower. We make heavy use of CMM equipment, as do our vendors. This makes verification of GDT (or traditional dimensioning) less painful.

3/4 of all the Spam produced goes to Hawaii - shame that's not true of SPAM also.......

 

[automationbabe]

My most favorite example of GDT usage, which I have used as well, is for a spacer.

The spacer must be a certain flatness on both sides an a specific thickness. However, I don't care if its made out of any shape stock...

So the thickness is toleranced...and GDT is applied to profile stating how little I care about the outer edges.

I learned GDT from one on the GDT board...best explanation he gave us was use GDT when you want to state how much you do or don't care about a feature.

Fun stuff all around though...

 

[StarrRider]

JNR

There is defiantly a relationship between the size of the company and the amount of inspection that is preformed. At one company we had a 10-man machine shop, 1 full time inspector, and about 130 employees in total. In a much smaller company, we only had about 10 people involved in a project with possibly 60 employees worldwide. In that company, parts were rarely inspected – there simply wasn’t time to do it – we were too busy swapping hats, putting out fires, or trying to accomplish something.

The answer your question, “why bother to tolerance the drawing in the first place”, is simple. Our vendors didn’t know how often the parts were inspected or to what degree – and that lack of knowledge kept them reasonably honest. We did find errors from time to time and had them replaced, which also helped, but inspection was never implemented properly. We were mainly doing proof of concept R&D projects and our goal was always the next trade show. Appearance outweighed functionality – but the machine did have to work (or at least look like it did). If a public reaction was favorable – then – maybe – the project would be funded and more features/functions would be added. Most projects never made it very far, but one made it through 3 prototypes, going to several shows in about 3 years and was finally sold to manufacturing. When that happened, all of the parts got reworked so that the bugs were eliminated and the shims and cheats were removed. It would have been an impossible job in the timeframe we had if we did not stay on top of the documentation. - YES - I LOVED that job.

When MadMango

brought up the cost issue, I remembered the machine shop manager and thought I would throw it in. I do avoid excessive GDT but I use it when and where I need it. Personally, I was surprised when I was informed why my parts were so expensive. In truth, I would have gone to a different vendor if it wasn’t for the fact that he worked for the same company that I did. So – I was stuck with him. Yes - I could have bitched and argued and go over his head – as many times as necessary – but that would only have given me an enemy with 15 years more seniority and a position more influential than mine. Even if he didn’t fight dirty, it was simpler to make him a friend. Unfortunately, that meant that I had to add leader notes stating what I needed (like “Machine this surface flat +/- .003”).

Well-documented parts using GDT may be cheaper – I don’t know, but I am sure there are numerous studies proving it either way (I never trust anyone who “Proves” anything using statistics – it is too easy to get the results you want). I only request quotes – I don’t give them. I do know that “logic” will not provide the answer, no matter how compelling the argument. There is a world of difference between an Engineer and a man who makes a dozen quotes a day – and then has to explain why he is making or loosing money. It does help to know where your parts come from. An internal shop has incentives to keep part costs high. A private shop that is HUNGRY will normally (if they know you are getting quotes from multiple sources) underbid any shop that does not NEED the work.

Lee

Consciousness: That annoying time between naps.

 

[TheTick]

GDT is a lifesaver for us in many cases. We have customers that often attempt to reject parts based on subjective criteria. More than once, we have had these criteria explicitly spelled out with GDT on a print that the customer had approved. That takes us from "Oops, sorry." to "How can we work together to change this." in one short phone call.

If the ladies don't find you handsome, they should at least find you handy.