67 Replies Latest reply on Jan 12, 2017 6:29 AM by Peter Van De Bogert

    GDT QUESTION

    Scott Cole

      I am updating an old drawing. how would I apply gdt for the holes they want to be .060 tir

        • Re: GDT QUESTION
          Christian Chu

          Pick one hole for datum and apply GDT to the other, similar as the one below (for total run out)

           

          • Re: GDT QUESTION
            Christopher Estelow

            Exactly what Christian said.  I would also consider changing that bolt circle from a radius to a diameter as well but that could just be my preference.

             

            Chris

            • Re: GDT QUESTION
              S. Casale

              The PDF presented does not indicate there are two holes as there is no depth call-out. The assumption when design checking that this 2X is an error. However. 2X or not:

               

              Where are the

               

              1. Need at least one datum
                • Better if you have two trying to accomplish what is desired.

               

              • Re: GDT QUESTION
                Matt Wallace

                I am far from an expert, and I am surprised someone more knowledgeable than me hasn't said something.  My understanding is Concentricity is difficult to inspect and rarely called for, unless you have a really good reason to control median points.  The symbols that folks have been adding are for total runout, which is different from runout and concentricity.  My guess is that none of these would be appropriate for a sheet metal part.  Position is the callout you will want.

                 

                Scott, your aren't being helped by the initial drawing.  You are going to have to do some reverse engineering and figure out what this part is supposed to do.

                  • Re: GDT QUESTION
                    Jim Steinmeyer

                         I have to add that threads like this one is exactly why so many companies and engineers are reluctant to move to GD&T. It just leaves to many people scratching their heads as to what it means or what the "correct" way  of dimensioning a part is. I have been following this thread in order to improve my knowledge and honestly it seems to me that the clearest way would be to extend the bolt hole center lines to show they match and then add a note. If designers and drafters who are supposed to be trained in creating prints have this much trouble figuring out the "correct" way to dimension a hole how can we expect the poor guy on the floor to know what we are trying to convey?

                         Just my two cents from someone who has had very little training in GD&T.

                      • Re: GDT QUESTION
                        Chris Dordoni

                        "It just leaves to many people scratching their heads as to what it means or what the "correct" way  of dimensioning a part is"

                         

                        I'm new to GD&T as well, and there are a few points I believe are true, but I would like some confirmation from someone that works with this regularly:

                         

                        - Is it true that engineering knowledge is required to understand the function of the part in context, so the order in which the tolerances or datums are applied is relevant?

                        - Is it also true that one person might generate a different spec than another, based on their experience?

                        - Isn't the purpose of GD&T intended to remove subjective analysis, so the part will be inspected the same way regardless of who does the inspection?

                          • Re: GDT QUESTION
                            Jim Steinmeyer

                            Chris,

                            My understanding would be that "in theory" statements 1 & 2 are incorrect and 3 is correct. With that said, look at how much effort is being placed in figuring out the correct way to tolerance these holes. Do you expect the shop to have a better understanding than the guys here on the forum? Will they put any where near as much effort into understanding the print?

                            • Re: GDT QUESTION
                              Christian Chu

                              Chris,

                               

                              Its kind of confusing to see diff. answer from this post because, I guess, this part was desinged by someone else and we provide help based on some assumptions from the PDF file. if we know exactly how this part is made and mounted, then the answer would be ended up the same

                              Be definition, GDT specifies how far actual surfaces are permitted to vary from the perfect geometry implied by the dwg. So the designer will view the part in the assembly and decide the Max and Mim conditions for the part functioned properly

                              Jim already answered to your question so #3: GDT can be used for both making and inspecting the part

                               

                              Edit: if I was the one to design this part I'd would want

                              1) Both side plates are parallel or both perpendicular to the base

                              2) holes on both sides perpendicular to the side plates

                              3) if I mount a shaft thru 2 holes, then 2 holes must be concentric in a certain amount

                               

                              The first 2 seem to be straight forward, but I have to think about how this part is inspected for #3. Usually an indicator is used in the machine shop for inpsection so I use TIR  to make sure the shaft is not rotated out of the center line between the 2 holes.  I can generally provide concentric GDT (still correct) but the guys in machine shop might have no idea how to inspect it

                                • Re: GDT QUESTION
                                  Chris Dordoni

                                  "Be definition, GDT specifies how far actual surfaces are permitted to vary from the perfect geometry implied by the dwg. So the designer will view the part in the assembly and decide the Max and Mim conditions for the part functioned properly"

                                   

                                  Christian,

                                   

                                  -This would suggest that it is impossible to provide GD&T for a lone part with no context ...?

                                  -How can the designer know the Max and Min conditions for the part to function properly ... it would be necessary to have engineering input to determine this?

                                   

                                  For example, I reverse engineer a lot parts from scans. I don't have a context for the parts, and if I did, I am not a mechanical engineer so I don't believe I would be qualified to indicate what the tolerance should be.

                                    • Re: GDT QUESTION
                                      Christian Chu

                                      Chris,

                                      Please see the below image with  solid lines for perfect geometry and dash lines for actual surfaces

                                      1) without straight GDT, the part can be bowed like the the shape of dash lines even though the dimension is still correct. If you are willing to spend $1000 for a smart phone, you sure don't want a bowed shape iPhone

                                      2) In the dwg, you add tolerance so the shaft can be sliding fitted with the bearing surface (solid lines); in reality,  the shaft can be bent and not be sliding fitted into the bearing even thought the dimension is still correct. so you need to determine an allowable amount of bend so the shaft can still be fitted into the bearing as shown in the 2nd image - This is what you need of perpendicular GDT

                                       

                                      GDT.jpg

                                      GDT2.jpg

                                  • Re: GDT QUESTION
                                    Rick McDonald

                                    Chris Dordoni,

                                    To add to what Jim Steinmeyer and Christian Chu stated, If you look at the drawing you will see that it is just a portion of the actual drawing and is missing a lot of detail and dimensions.

                                    For instance, there is no defined X dimension for where the holes in question are located on the part.

                                    The drawing would need all dimensions to be a proper drawing. There would also (possibly) be notations to indicate the designers intent or the use of the final part.  The designer is responsible to make the points you raised be clear enough to the fabricator so that the part can be made accurately.

                                    In the case of this part, as it is shown to us, was an incomplete drawing because it did not provide an absolute definition of the tolerances, GD&T  or specific alignments needed.

                                    Scott Cole said he was updating an old drawing - so I would imagine it was never completed or he removed details so he would not be showing the critical details for anyone to copy.  Scott can correct me if I am wrong, but I think he is trying to best define the part so there will be no questions in the future, but that he is also not 100% sure of what is the best way.

                              • Re: GDT QUESTION
                                James Riddell

                                This is a good candidate for true position with a circular tolerance zone.

                                • Re: GDT QUESTION
                                  Matt Wallace

                                  One of the sides that has the holes needs to be a datum.  One of the 1.78 holes needs to be the next datum. One of the patterns needs a true position callout to the the first two, and receives a datum callout.  The 1.78 hole on the other side gets a true position callout referencing the first 2 datums.  The second pattern gets a true position callout referencing the 3.

                                   

                                   

                                  I suspect the position of the pattern relative to the to the 1.78 hole matters, and matters more than the position of the patterns relative to each other.  In this case, the second 1.78 hole is a datum, and the second pattern gets the required position callout to the hole, and then gets a second position callout to the first pattern.  GD&T allows you to make the tight callout to the center hole, and then a much looser callout to the other pattern.

                                   

                                   

                                  Post up a screenshot of what this thing is supposed to do, and a drawing file in a format that can be marked up.

                                    • Re: GDT QUESTION
                                      James Riddell

                                      IMHO, that makes things much more complex and difficult to inspect.

                                      If I were looking at what Scott is saying, my interpretation is that the 1.78 hole is already a clearance and therefore not as important. He's saying that the mount holes are driving the position requirement (right Scott?).

                                      • Re: GDT QUESTION
                                        Scott Cole

                                        like this matt, I'm not sure how tight the tolerance should be

                                          • Re: GDT QUESTION
                                            Matt Wallace

                                            I think that is getting closer. B should have some perpendicularity callout to A since no other datum is present to control its orientation, but what is desired is clear enough.  The positional tolerances need a diameter symbol.  The 45 deg dimension should probably be basic, but it would need a datum to reference.  I also think the second 1.78 hole should be a datum (Say D), and the second pattern gets the same .01 diameter tolerance zone relative to A and D  but gets the looser .06 dia tolerance relative to the first pattern (defined by ADC, I think).

                                             

                                             

                                            I would guess that the tolerancing really applies at MMC.  If so, that can effectively loosen the tolerances and you will still get parts that go together.  If this is a production part, you can in theory have a set of go-no go gages to quickly inspect the part.

                                             

                                             

                                            I am far from an expert, so I keep hoping someone who knows what they are talking about jumps in.

                                              • Re: GDT QUESTION
                                                Scott Cole

                                                OK REV 5

                                                  • Re: GDT QUESTION
                                                    Matt Wallace

                                                    I think that is a lot better than what you started with.

                                                     

                                                    You need a diameter symbol for the second pattern tolerance.  You also need a tighter tolerance callout to A and D (but not C).  I am guessing that there is some kind of flange mounting to the part.  The tighter tolerance controls the pattern to the center hole, while the .060 is the much looser control for the orientation to C.  So you will end up with 2 Feature Control Frames for the second pattern, the first to ADC for orientation, and second to A and D for position to D.  The concept is in ASME 14.5 section 5.4

                                                     

                                                    Thinking more about this, the entire initial 5 hole pattern could be called out as a datum (B, say), and the entire second pattern gets the loose orientation tolerance to A and B, and then the same tolerance to A that the first pattern has.  If you aren't ready to put your fist through your monitor, I think this is the best way- but I don't know how this part is really supposed to work.  If you do this, then the 45 deg dimension becomes basic.

                                                     

                                                    I am also guessing that the tolerancing really applies at MMC, which is a useful concept to get familiar with.

                                                     

                                                    It may be appropriate to point out that number in the tolerance is the diameter of the tolerance zone, so .06 means no more than .03 away.

                                                     

                                                    This exercise is making me appreciate how many of my own parts I make.

                                                      • Re: GDT QUESTION
                                                        Jim Steinmeyer

                                                        Just a wild guess but I suspect that the goal is to obtain something like the attached view.

                                                        Capture.PNG

                                                        The bolt pattern that both bearings are located by needs to be within a tight tolerance for the shaft to be in the bearings without a bind. as the OP has stated the center hole is large enough that even if it is a little off the shaft will clear. Again as with my earlier post, a drawing is an attempt to clearly convey the required dimensions and tolerances to the shop floor to get the part made correctly. I am afraid that since it has taken this many attempts to get the "correct" GD&T tolerance information on the drawing the shop will look at and have no clue what is required. Then one of two things will happen, either a diligent welder will hunt up the engineer/drafter, or more likely, he will shrug his shoulders toss the print aside and weld it as he feels like.

                                                             Here where we don't know GD&T and hire just anyone with a pulse that can strike an arc so our prints have to be dumb-ed down. I would create a detail view of the hole pattern at location A and provide tolerances in that view. Then using the projected views that you have I would extend the center lines and include a note indicating that the bolt holes MUST be aligned to the desired tolerance. Crude, yes. Archaic, probably. But the shop would get the idea and make the parts reasonably close to what was desired.

                                                             I know that isn't what you asked about but I have been yelled at far to many times because the stupid engineers are unable to make clear prints that the shop can easily follow.

                                                        • Re: GDT QUESTION
                                                          Scott Cole

                                                          this hopper has a bearing on one side with a feed screw, the other side is just a spout that's why the .060 range

                                                        • Re: GDT QUESTION
                                                          James Riddell

                                                          Scott, I have had some years experience with GD&T and this looks to be way overkill.  You do have some problems with it even so.  Your 0.060 tol is rectangular, making it circular will improve alignment.  Keeping the frames assuming RFS is usually a better way to go but it requires a little more inspection time/calculation.  Your QC will have a difficult time establishing your BC datums - this means added cost/time.  By calling out TP on the right BC to the opposite hole you are defeating the purpose of keeping both BCs within 0.060 of each other.