Parallel Requirement On Part of a Surface 2

[aardvarkdw]

I have a round, flat part with a hole in the center. It kinda looks like a huge washer. Datum A is the bottom surface. How can I get a parallel tolerance of .001in to Datum A for theoretical area 12in in dia. from the center of the part and then the remainder of the part has a parallel tolerance of .003in? Right now there is a FCF with a parallel symbol and no tolerance that references a flag note which says, "TOP SURFACE TO BE PARALLEL TO DATUM -A- TO WITHIN .001" INSIDE n12.000" FROM CENTER CUTOUT AND .003" ACROSS THE REST OF THE SURFACE."

Is this alright or is there a better way. Any suggestions are helpfull. Let me know if you need a pic in order to visualize this.

David

 

[ctopher]

aardvarkdw,
It is almost the same as mine.
For most companies and machine shops, they will understand.
MechNorth's description would be correct, but a lot of companies don't use or understand much of GD&T.

Chris
Systems Analyst, I.S.
SolidWorks 06 4.1/PDMWorks 06
AutoCAD 06
ctopher's home (updated 06-21-06)

 

[MechNorth]

Dingy2, You are correct on the use of the chain line vs the phantom line, and the hatching. I was thinking it was the same as for datum targets which use phantom lines. Unfortunately, many CAD systems still make it difficult to hatch a non-section, so it tends to be omitted. The use of the leader to the surface is arguable, but the Y14.5 standard doesn't show any GD&T application to isometric views, only to profile, plan and sectional views. That's why my question re Y14.41, because I had understood that standard was to address GD&T applications on isometric views, solid models, etc. Another problem that is encountered is the inability of some CAD packages to allow you to change the line type to dashed as needed for leaders to show that a control is applied to the back face rather than the front face.

David, there is a note in the book (I had the page last night, but can't relocate it right now) that indicates dimensioning and such are to be done on the profile views ... it may say SHOULD be, but I can't recall. Another reference is made to it in 6.4.1.1 - Straightness Tolerance ... "A straightness tolerance is applied in the view where the elements to be controlled are represented by a straight line." Similarly for Flatness 6.4.2.1, and for Profile 6.5.1(a). This document was originally meant for 2-D drafting rather than the 3-D world we now work in, so some major extrapolations are now needed. Again, I thought Y14.41 was supposed to clarify the application in that environment.
The image you posted on tinypic looks good. I would suggest adding a centerline in the side view and a general tolerance for the other dimensions. I hate to keep saying "problems with CAD", but it's true; in this case, most CAD packages would have a hard time keeping the chain-line associative with the part geometry, so any changes could easily shift the chainline wrt the centerline of the geometry and confuse the user.


Jim Sykes, P.Eng, GDTP-S
Profile Services
CAD-Documentation-GD&T-Product Development

http://www.profileservices.ca

 

[thundair]

Right on Jim... we have been in Model Base Definition for over a year now and it is hard to imagine going back to anything in 2D.
Our FT&A only allows us to pick a view plane and point to the surface we are trying to control. With boundary's like uni-tolerance on profiles or limited tolerance like Chris's example is difficult for now.
We cover most with field notes just like the originator of this thread, David.

Cheers

I don't know anything but the people that do.

 

[ajack1]

Doesn’t this just highlight what we are all up against? The OP asked the question how do you show the flatness tolerance on a washer.

Many people have come up with different ideas and people who post on here regularly and quote standard this or that still get it wrong.

As far as I can see only MechNorth has it right (and I may be wrong about that) and we are talking about the flatness of a washer, not parallelism, concentricity, tolerances of the OD and ID, material specs and all the other bits and bobs.

If we cannot get on simple thing right on the most basic of parts what hope do we all have?

 

[MechNorth]

Ajack1 - No hope at all of gaining absolute consistency...that's why we all still have jobs! I also believe it was set up this way by the founding fathers of GD&T to give engineers something to debate professionally so that their wives can't say "Why are you always arguing with our friends!?"


Jim Sykes, P.Eng, GDTP-S
Profile Services
CAD-Documentation-GD&T-Product Development

http://www.profileservices.ca

 

[ctopher]

We also can not get it right without knowing the design. If he wants parallelism, I try to help with it. If it should be flatness, probably, but wasn't the question.
I try to make my responces short. I'm not a trainer here, just a helper.
We all do what we can and just have fun with it.


Chris
Systems Analyst, I.S.
SolidWorks 06 4.1/PDMWorks 06
AutoCAD 06
ctopher's home (updated 06-21-06)

 

[ajack1]

Sorry ctopher my mistake he does state parallelism and not flatness.

The point I was trying to make (badly) was that even with the simplest of parts and with only one critical feature people who seem to know what they are talking about still get it wrong, what chance do we stand with complicated parts and many critical features?

 

[aardvarkdw]

The best you can do, I guess, is to get as many "opinions" as possible and make an educated decision as to what will best convey the information. As I frequently do by comming here when I am unsure of my own reasoning or can't find any definitive information to validate my reasoning.

 

[ctopher]

Yes, that is all you can do.
Getting a copy of the specs in addition to posting questions here, you will get your answers.
Thanks guys!

Chris
Systems Analyst, I.S.
SolidWorks 06 4.1/PDMWorks 06
AutoCAD 06
ctopher's home (updated 06-21-06)

 

[KENAT]

ajack1

From a couple of conversations I've had with someone who should know even those on the asme y14.5 committee don't always agree on the best way to dimension/tolerance things.

Some have particular favourites, such as profile, while others use the full range of methods as they see appropriate.

I will admit I didn't refer to the spec before giving my original post and didn't think it through, it was just a quicky to give aardvarkdw and anyone else who wanted to something to think about.