Defining a plane from "tips" of multiple curved surfaces, and also chosing appropriate dat
Defining a plane from "tips" of multiple curved surfaces, and also chosing appropriate dat
(OP)
I'm looking for the best way to define a plane, and to choose which of a couple options make the most sense for defining a primary datum plane (i.e. Datum Plane A).
Background: I'm fairly new to a job working for an electricity company (in Costa Rica). My department is fairly new, and currently have the task of designing some replacement parts for steam turbines. Most of the folks here appear to be somewhat new to the concept of GD&T. I am NOT an expert, but I have more experience than they do working with ASME y19.5 (about 4 years in the aerospace industry in the US). Thus, I am evaluating one of their drawings for a particularly complicated part.
The part actually an assembly, and is a complete nozzle stage of a steam turbine. In that stage, the construction method is such that they first weld all the blades (64) of the stage into inner and outer bands of a large disc. The technicians essentially do their best to align the blade tips. From that ring of blade tips, they use a CMM to compute a derived, imaginary plane (see
for a conceptual idea). For sake of argument, we'll say that the plane is shown in the image as the "Base" line (just to have an easy picture in your head, though realistically it will be the opposite side). From that plane, the rest of the stage assembly is machined, including attachments to the casing.
The most critical feature, in terms of next-level assembly (and therefore the as-inspected part), is a plane which ultimately forms a seal (metal to metal) with the casing for the turbine (see
for a cut-view). My confusion comes from the fact that this planar surface is also essentially (and necessarily) one of the last features cut in the machining process. So, in terms of next-level assembly, I'm inclined to label this surface as Datum Plane A, as it is the locating feature for the full nozzle stage assembly. However, in terms of manufacturing, since it's cut last, I'm not sure if that makes sense. In terms of manufacturing, the derived plane is primary.
So, my first question is: Is the final planar surface appropriate to choose as a primary datum plane (Datum Plane A) (i.e. utilizing a Next-level assembly reference frame)? Or is the original, derived plane more appropriate for Datum Plane A) (i.e. utilizing an as-manufactured reference frame)?
My second question is, considering that my derived plane is computed from the contact points of multiple blade tips (ref Image 2), is a multi-surface Profile call-out the appropriate way to control this plane? Although the technicians are currently "doing their best" to align the tips, once we determine what is reasonable, we want to be able to define a tolerance zone for that derived plane. So, ultimately, I believe we'll be defining that plane with a FCF calling out Profile of .150 (mm), and a note above the FCF stating "64 surfaces" (possibly with an additional note defining those surfaces). I would assume that if I choose my final, machined plane as my Datum A, then I would define my derived plane with a basic dimension, and add a reference to Datum A in the FCF.
For the record, from my previous training, I believe the correct answer for Question 1 is to use the Next-Level Assembly (i.e. as-inspected) approach, rather than the as-machined approach. However, I'd still like to have the discussion.
Background: I'm fairly new to a job working for an electricity company (in Costa Rica). My department is fairly new, and currently have the task of designing some replacement parts for steam turbines. Most of the folks here appear to be somewhat new to the concept of GD&T. I am NOT an expert, but I have more experience than they do working with ASME y19.5 (about 4 years in the aerospace industry in the US). Thus, I am evaluating one of their drawings for a particularly complicated part.
The part actually an assembly, and is a complete nozzle stage of a steam turbine. In that stage, the construction method is such that they first weld all the blades (64) of the stage into inner and outer bands of a large disc. The technicians essentially do their best to align the blade tips. From that ring of blade tips, they use a CMM to compute a derived, imaginary plane (see
for a conceptual idea). For sake of argument, we'll say that the plane is shown in the image as the "Base" line (just to have an easy picture in your head, though realistically it will be the opposite side). From that plane, the rest of the stage assembly is machined, including attachments to the casing. The most critical feature, in terms of next-level assembly (and therefore the as-inspected part), is a plane which ultimately forms a seal (metal to metal) with the casing for the turbine (see
for a cut-view). My confusion comes from the fact that this planar surface is also essentially (and necessarily) one of the last features cut in the machining process. So, in terms of next-level assembly, I'm inclined to label this surface as Datum Plane A, as it is the locating feature for the full nozzle stage assembly. However, in terms of manufacturing, since it's cut last, I'm not sure if that makes sense. In terms of manufacturing, the derived plane is primary.So, my first question is: Is the final planar surface appropriate to choose as a primary datum plane (Datum Plane A) (i.e. utilizing a Next-level assembly reference frame)? Or is the original, derived plane more appropriate for Datum Plane A) (i.e. utilizing an as-manufactured reference frame)?
My second question is, considering that my derived plane is computed from the contact points of multiple blade tips (ref Image 2), is a multi-surface Profile call-out the appropriate way to control this plane? Although the technicians are currently "doing their best" to align the tips, once we determine what is reasonable, we want to be able to define a tolerance zone for that derived plane. So, ultimately, I believe we'll be defining that plane with a FCF calling out Profile of .150 (mm), and a note above the FCF stating "64 surfaces" (possibly with an additional note defining those surfaces). I would assume that if I choose my final, machined plane as my Datum A, then I would define my derived plane with a basic dimension, and add a reference to Datum A in the FCF.
For the record, from my previous training, I believe the correct answer for Question 1 is to use the Next-Level Assembly (i.e. as-inspected) approach, rather than the as-machined approach. However, I'd still like to have the discussion.





RE: Defining a plane from "tips" of multiple curved surfaces, and also chosing appropriate dat
RE: Defining a plane from "tips" of multiple curved surfaces, and also chosing appropriate dat
In researching this whole business, I came across the Tangency modifier, which I'd never heard of before. I tried to contemplete whether or not that would be more appropriate for my case or not. On the one hand, it's supposed to control only the high points of the controlled bodies within my defined tolerance range (i.e. a zone between two parallel planes). The problem is, I'm not clear this would control the blade tips from falling BELOW my tolerance zone. I only want to control the tips, which are the high point of each blade, so Tangency would SEEM to be the perfect case for me, but I'm not clear, and would very much like the opinions of experts on the subject.
For the those who don't speak spanish, "superficies" is "surfaces".
RE: Defining a plane from "tips" of multiple curved surfaces, and also chosing appropriate dat
By the way the link in your 3rd post didn't seem to work.
In terms of selecting datum planes, function generally trumps manufacturing process when it comes to that - but that's a generality and you could be an exception because of your weldment nature or something.
What is Engineering anyway: FAQ1088-1484: In layman terms, what is "engineering"?