Hello, first of all, thank you for your work in your videos, they are very classifying. Now we go to my question, I have a part of sheet metal, 1.6mm thick, in which the only important thing is the relative position of 4 holes, the rest of the part is covered with the general notes of the part list. Can I dimension the four holes with a position tolerance without any Datum? I just want to control the pattern between them. In Asme. Thanks
Very interesting video. 2 questions here, which I think were not clear: 1. While in the nominal case, those 5 cylinders converge to a point, in reality they will not. The axes will kind of miss each other in 3D space so how could a single point be determined? 2. Regarding the orientation in plane (that fixes the rotation of the pattern of 4 holes), you mentioned that you can choose whatever you want. But again, since on the actual part, those 5 cylinders will not have the nominal angles, is the reference orientation considered as an average between the orientation of those 5 cylinders? Otherwise, choosing just one would imply that it is a difference to which one you choose, since the will have different deviations to the orientation. Thank you very much!
I would think the CMM can do the mathematical calculations for the 5 cylinders converge to a point. Every two axis will converge to a point, if we can find a a circle that just big enough to accommodate all the points converged from those 5 cylinders, that center of the circle I would believe be the determined zero point
@@kw2519 No sure this is it. The only "approximation", if you will, is with regards to the point cloud filtration/association, which is the way to approximate the real geometry in the best way. Moving forward from there, all elements are perfect geometrical elements that have a precise mathematical description. There needs to be a precise way to calculate all of these median points, axis intersections etc.
@@AlexLapugean ya and that’s usually defined by the way your point cloud is setup. You can take two measurements of 3 points and get a very rough approximation. Or you could sweep the bore at 10 depths and get a much better result. I’m just saying it all comes down to how the parts dimensioned. As he said, we are the interpreters.
These are great questions. You are absolutely correct in saying that the axis of the holes will never converge on a single point. In order to get the center point of the pattern of holes as a datum we need to simulate the datum. And for a pattern of holes, that simulator (physical or digital with a cmm) is 5 cylinders located at true positions (ie converging on a single point) and those theoretical cylinders will expand simultaneously until the pattern of tangible holes "settle" out on the radial high points of all 5 holes simultaneously. This is simulation at RMB. However, a much more practical real world gage would be if datum B were called out at MMB, this would allow us to have stationary pins that dont need to expand. And the part is allowed to "shift" on these perfectly located pins, resulting in datum shift.
I put GD&T on my drawings to better define my parts and give the machinists more tolerance. But the parts are always quoted more expensive because they have to have more experienced technicians interpret my drawing and they say it requires a CMM inspection now. How would you respond to this?
Thanks for this comment, unfortunately we hear this very often. And while very complicated parts MAY require use of a CMM to speed up the process, the simple use of GD&T does not inherently implement the need for a CMM. In fact alot of times if the GD&T is truly understood by the inspector, many of the manual inspection processes don't change, only the interpretation of the variable measurements is what changes like in the case of diametric deviations. Long story short, complicated parts require complicated measurement methods and that has nothing to do with GD&T. The first step to alleviating this problem is training on tolerances and GD&T across the industry. We are always happy to help, whether that is through our free content online or more formal training options.
As far as I know, the material conditions might be applied only on feature of size. What should be the MMC value of a point, which is an intersectoin of five axes? Are you going to add the MMC of every cylinder? I think - no. Are you going to find the MAX MMC of some cylinder an use it? Or is it the MIN value? Think it is not correct either. Also, cannot think of any real world problem, that will require such dimensioning schema. Of course, if it is going to be a check gauge - then, yes, it might work. But, don't think the CMM will be able to evaluate this.
Thank you for answering my questions. This video is clear. Now I am more confident to discuss with my colleagues. Super appreciated.👍👍👍
Hello, first of all, thank you for your work in your videos, they are very classifying. Now we go to my question, I have a part of sheet metal, 1.6mm thick, in which the only important thing is the relative position of 4 holes, the rest of the part is covered with the general notes of the part list. Can I dimension the four holes with a position tolerance without any Datum? I just want to control the pattern between them. In Asme. Thanks
Very interesting video. 2 questions here, which I think were not clear:
1. While in the nominal case, those 5 cylinders converge to a point, in reality they will not. The axes will kind of miss each other in 3D space so how could a single point be determined?
2. Regarding the orientation in plane (that fixes the rotation of the pattern of 4 holes), you mentioned that you can choose whatever you want. But again, since on the actual part, those 5 cylinders will not have the nominal angles, is the reference orientation considered as an average between the orientation of those 5 cylinders? Otherwise, choosing just one would imply that it is a difference to which one you choose, since the will have different deviations to the orientation.
Thank you very much!
I would think the CMM can do the mathematical calculations for the 5 cylinders converge to a point. Every two axis will converge to a point, if we can find a a circle that just big enough to accommodate all the points converged from those 5 cylinders, that center of the circle I would believe be the determined zero point
That’s what tolerances and bonus material are for. Nothing is ever perfect, so the same rules apply here.
@@kw2519 No sure this is it. The only "approximation", if you will, is with regards to the point cloud filtration/association, which is the way to approximate the real geometry in the best way. Moving forward from there, all elements are perfect geometrical elements that have a precise mathematical description. There needs to be a precise way to calculate all of these median points, axis intersections etc.
@@AlexLapugean ya and that’s usually defined by the way your point cloud is setup. You can take two measurements of 3 points and get a very rough approximation. Or you could sweep the bore at 10 depths and get a much better result. I’m just saying it all comes down to how the parts dimensioned. As he said, we are the interpreters.
These are great questions. You are absolutely correct in saying that the axis of the holes will never converge on a single point. In order to get the center point of the pattern of holes as a datum we need to simulate the datum. And for a pattern of holes, that simulator (physical or digital with a cmm) is 5 cylinders located at true positions (ie converging on a single point) and those theoretical cylinders will expand simultaneously until the pattern of tangible holes "settle" out on the radial high points of all 5 holes simultaneously. This is simulation at RMB. However, a much more practical real world gage would be if datum B were called out at MMB, this would allow us to have stationary pins that dont need to expand. And the part is allowed to "shift" on these perfectly located pins, resulting in datum shift.
I put GD&T on my drawings to better define my parts and give the machinists more tolerance. But the parts are always quoted more expensive because they have to have more experienced technicians interpret my drawing and they say it requires a CMM inspection now. How would you respond to this?
Thanks for this comment, unfortunately we hear this very often. And while very complicated parts MAY require use of a CMM to speed up the process, the simple use of GD&T does not inherently implement the need for a CMM. In fact alot of times if the GD&T is truly understood by the inspector, many of the manual inspection processes don't change, only the interpretation of the variable measurements is what changes like in the case of diametric deviations.
Long story short, complicated parts require complicated measurement methods and that has nothing to do with GD&T. The first step to alleviating this problem is training on tolerances and GD&T across the industry. We are always happy to help, whether that is through our free content online or more formal training options.
As far as I know, the material conditions might be applied only on feature of size. What should be the MMC value of a point, which is an intersectoin of five axes? Are you going to add the MMC of every cylinder? I think - no. Are you going to find the MAX MMC of some cylinder an use it? Or is it the MIN value? Think it is not correct either. Also, cannot think of any real world problem, that will require such dimensioning schema.
Of course, if it is going to be a check gauge - then, yes, it might work. But, don't think the CMM will be able to evaluate this.