Who's responsible for understanding GD&T besides the drafter/designer?

[RubenGman]

Drafters and designers are expected to have more than a basic understanding of GD&T, they're expected to have a working knowledge on how to apply it in order to produce fit and function at a reasonable cost. In addition to the basic courses, many pursue supplemental classes that teach advanced techniques that relate to specific problem solving that may have a direct impact on their design projects. In my years of working in the automotive and aerospace (spacecraft) industries I've yet to see another person outside our profession hold the same expertise we do. Sure, some inspectors have the basic understanding of single and multiple datum reference frames, or even understand the basic concept of datum precedence but are they expected to have the same level of expertise we have? What about the machinist? How many have taken advanced courses and maintained current in ASME or ISO standards? My brother is an estimator at a contract shop and confessed to me yesterday that most drawings that come in with a tolerance spec of .002 or under, are simply assigned to their best machine (The Hurco)that will insure compliance and produce the whole part better than required. If the drawing contains anything questionable, they'll call the originator for clarification. They will do whatever it takes to get the part out within specs on time but, they won't spend the extra time to find the virtual conditon of features if they have to take into account multiple variables such as Datum Shift thru size modifiers on datum references. Time is money and if the drawing requires a lot of it they lose profit. So to prevent losses, they'll simply run the part on their best machine....even if it means opening up a new shift. Do the inspectors have the same working knowledge we do? I may be wrong in thinking this way but are we wasting our time with calling out more than the basic GD&T needed to define our parts if we can simply add a note on the face of the drawing specifing to be made on their best machine or an overall tolerance their best machine is capable of maintaining since it that's what's likely to occur anyways?

 

[Heckler]

Our QA department is staffed with two guys that really know their GD&T. They have to the government sends in auditors all the time. And since we're AS9000 certified everything is documented. Even our prime contractors send in their QA engineers to check our documentation.

I've also worked at places that QA was a joke.....I had to explain a weld symbol callout. I even had a vendor not know what NS / FS ment....Near Side / Far Side.

Best Regards,

Heckler
Sr. Mechanical Engineer
SW2005 SP 5.0 & Pro/E 2001
Dell Precision 370
P4 3.6 GHz, 1GB RAM
XP Pro SP2.0
NVIDIA Quadro FX 1400
o
_`\(,_
(_)/ (_)

"Coming together is a beginning, staying together is progress, and working together is success." - Henry Ford

 

[RubenGman]

Heckler,

Half the battle is having your Inspectors know GD&T as much as you do....consider youself blessed if they know more. The other half is having your Machinist know GD&T as much as you. The odds are pretty slim however since this is not the norm. All our contractors are AS9000 certified and yet, still find a large percentage of them asking for the 3d cad models. When this happens, the probability of having the machinist read and interpret GD&T is far less. One would be hard pressed to find both the machinist and inspector responsible for interpreting your drawing and making your part, able to read ANSI as well as you or I. Again, I've yet to see either of them with same expertise as those in our field who are continuously adding to their education of GD&T application.
I don't have an answer for the mis-match between the intended and the interpreted, I'm just wondering if the direction we're heading, unless everyone who's involved participates, is a correct one.

 

[RubenGman]

Tick,

When in doubt, spell it out.

 

[TheTick]

My point, exactly, Ruben.

Personally, I don't wait for mistakes to end up in "GDT court". Before steel gets cut, I'm in touch with vendors or machinists to make sure any arcane points are understood.

 

[ctopher]

RubenGman,
I have used several shops that understand most of GD&T. I have only ever met one person that knows it (or understands it) 100%.
If the shop does not understand something on the dwg, they call me and we go over it together. If we both can not agree with the understanding og the specific GD&T, we find help elsewhere until problem fixed.
I use GD&T on all parts. The more I use it, the more I understand it. Same for the shops and inspectors.
I get purchasing all the time try to change the dwg to remove GD&T off. If I lose the battle, I tell them to have someone else do the change because I will not be responsible for any parts that do not work.
My outlook on this? If you design parts, have some understanding of it. Find a shop that understands it. When purchasing goes out for bids, let them know that the understanding of GD&T is included somehow.
I agree that CAD/CAM makes it easier for a michinist to machine a part without much GD&T knowledge, but it still has to get inspected ... and fit.
What you put onto a dwg is what you get in the part.

Chris
Systems Analyst, I.S.
SolidWorks/PDMWorks 05
AutoCAD 05
ctopher's home site (updated 06-21-05)
FAQ559-1100
FAQ559-716

 

[Heckler]

We don't live in a perfect world and people make judgement calls based on their knowledge and undstanding of GD&T sometimes right and sometimes wrong.

Best Regards,

Heckler
Sr. Mechanical Engineer
SW2005 SP 5.0 & Pro/E 2001
Dell Precision 370
P4 3.6 GHz, 1GB RAM
XP Pro SP2.0
NVIDIA Quadro FX 1400
o
_`\(,_
(_)/ (_)

"Coming together is a beginning, staying together is progress, and working together is success." - Henry Ford

 

[RubenGman]

Ctopher,

Do you recall the big push towards converting our coveted archaic inch system to metric? Back in the mid eighties, I was working for a Class 8 truck manufacturer that adopted the metric system on all cad drawings (we used Cadam back then). The problem we encountered time after time was that most vendors weren't embracing the new system so, as a favor to them, we referenced the inch equivalence in order to ensure no ambiguity occured when interpreting our drawings. The struggle to conform to something as simple as the millimeter never caught on, sad really. This is how I feel with most GD&T....not all mind you. It seems that we have to be there to clarify and interpret for them so there's no ambiguity and possible mis-interpretation.....remember time is money and if there are ANY errors on the fabricated parts, profits are lost! They learn only enough to get by to appease us and say they can read GD&T in order to be certified by the latest ISO. They need to do this to stay in business. We are still the only one's persuing it's advancement. I feel we are still holding their hand. Has anyone of you felt completely comfortable knowing you've applied, not just basic but, advanced callouts and had complete faith they've interpreted your drawing without ambiguity time and time again, year after year? It's been over 40 years....you'd think it would have caught on by now.

 

[ctopher]

Yes, I remember. I am for having both inch and metric on dwgs, but I get a lot of flack all the time for this.
From my experience, there is more mis-interpretation of dim's and tol on a dwg without GD&T. With GD&T, there is only one way to interpret it. The problem is not with the standards, it's with the user’s mis-interpretation of it.
i.e. If you have a part with a hole thru it and a centerline drawn thru it, the hole is theoretically straight, the machinist will machine a straight hole, but is it 90 deg to a surface? It may or not be. But, because there is not any GD&T, it is OK, but may not work with mating parts. With GD&T, you can make sure it is positioned exactly the way you need it and it can be inspected. Without GD&T, inspector’s jobs can become very easy. IMO, if everyone would take the time to understand it more, we would be more ahead than we are now.

Chris
Systems Analyst, I.S.
SolidWorks/PDMWorks 05
AutoCAD 05
ctopher's home site (updated 06-21-05)
FAQ559-1100
FAQ559-716

 

[ewh]

The only experiences where I've had parts made to GD&T that is understood by everyone involved is with in-house parts. Granted, we may not always be 100% correct as to the use of the standard, but we do all speak the same dialect. I have confidence that if I specify something on a drawing that the engineer, machinist or inspector don't understand, they will ask me my intent. Only when that intent is understood are chips cut.
As far as parts made out of house, the majority are tooling with complex lofted surfaces, and the solid model is the definition, so it goes out with the p/o. Since it is usually for tooling, we only specify a profile tolerance and let them determine any other tolerancing. This relieves much of the headache associated with trying to follow the standard.

 

[drawoh]

RubenGman,

I am not absolutely confident in our machine shop's understanding of GD&T, but usually, machine shops are not the problem. Machining is an accurate process. In most cases, a machine shop supervisor should be able to examine a drawing and satisfy themselves that everthing can be done easily with their standard tools. The machinist can be told the use the DXF file to generate the CAM model. Machine shops easily work to within +/-.005".

If they ship you a non-conforming part, you ship it back. They will learn.

Machining does a great job of covering up an atrocious lack of comprehension of drafting by both design and manufacturing.

The fun starts when you punch sheet metal, then bend and weld it for precision electronics. It appears to me that sheet metal shops are more sophisticated about GD&T. They have to be, as do designers specifying this stuff. My favorite shop bends sheet metal to +/-.015".

I do not send enough stuff out to welding shops and foundries to make generalizations about them.

JHG

 

[RubenGman]

Ctopher,

You hit the nail on the head. My concern is not everyone involved, that should be, has jumped on the bandwagon with us....which is why we're not further ahead than where we should be. In fact, we're slipping back every time they ask for a 3d cad file.


EWH,

The people in-house are most likely following your lead(and the people in your department making drawings) in order to be able to understand your language. But, ask yourself this, how many of those machinist and inspectors have taken advanced GD&T courses....how many do you encounter on the forums pursuing their own advancement of ANSI knowledge to further the application of thier use of the language we in our field tend to sing with? As far as sending out a solid model, the shop will unlikely even look at your control boxes let alone extrapolate percentages based on tolerances with or without bonuses. Why should they?.....their 5-axis will ensure your part not only meets but exeeds your specs....time is money.

 

[RubenGman]

Drawoh,

See my reply to EWH. The reason shops are able to cover up the lack of understanding is because their machines can make the parts using the dxf file accurate enough to exeed specs on the drawing. The drawing is most likely just used as reference at the machine shop. I worked at a machine shop when I was 18 as a jitterbug (we made all the Atari enclosures) and witnessed first hand what sheetmetal stretch and deformation really was. There's more geometry involved with bent parts....understanding GD&T can be more critical than throwing a hunk of 6061-T6 on a Hurco and letting it run.

 

[drozic]

I always thought that you define several "critical dimensions". Critical means that they affect form, fit, and function. With that, GD&T ensures that stacked tolerances of critical dimensions will not cause interference.

I laugh when someone tells me that every dimension is critical, and I refuse to approve or verify that drawing.

I really laugh when I get floating centerline to obscure angle as critical. To those people I say, "If you need a coordinate measuring machine (CMM) to verify that dimension, don't make it critical"

 

[Nella95]

drozic: I don't like that 'critical dimensions' callout - maybe for eng, but not manufacturing and quality. I don't know - something psychologically happens with quality and manufacturing where they seem to assume that if it's not 'critical' - it's unimportant.

Maybe saying dimensions are 'less critical' would work better.
Sorry, my soap box is starting to crack and splinter.

 

[ctopher]

If a dim is critical, it has to be labeled that way.

Chris
Systems Analyst, I.S.
SolidWorks/PDMWorks 05
AutoCAD 05
ctopher's home site (updated 06-21-05)
FAQ559-1100
FAQ559-716

 

[ewh]

Per ASME Y14.5 para 1.4(c), while perhaps not critical, every dimension to define the end product is required. The tolerances you give are up to you.

 

[ctopher]

Drawings for military, they usually will specify "critical dimensions". These dim's have to be labeled, tracked and documented by Inspection & QA. All dim's on a dwg are important to the design, but only certain ones are 'critical'.

Chris
Systems Analyst, I.S.
SolidWorks/PDMWorks 05
AutoCAD 05
ctopher's home site (updated 06-21-05)
FAQ559-1100
FAQ559-716