Style question with ordinate dims, datum surfaces
Style question with ordinate dims, datum surfaces
(OP)
Hi All,
I just started a new job and got handed back a checked drawing with Red Ink all over it. The points at-issue:
1. Ordinate "Zero" aligned with datum surfaces. I don't operate under a rule where zero must be at a datum surface. Sometimes it is, sometimes not. My philosophy is that I dimension the part to be concise and clear on the drawing document, then I tolerance the part (with GD&T) to represent the functional/design requirements of an acceptable production. Red ink on my drawing re-sets my datum references to be all planar exterior surfaces and sets all my ordinate zeros to these surfaces. This is hard to explain with words and I can't post the drawing itself obviously, but my original datums were: A: A mating surface that interfaces with another part's surface via bolted connection, B: A dowel hole on A, C: a second hole with a tight diameter tol on A. The red ink keeps A as-is, sets B to one side of the part exterior normal to A, sets C to another side of the part normal to A and B, and adds a datum D which is the dowel hole. I acknowledge that this is totally acceptable to do, but here's why that layout bothers me: Too many controlled surfaces, and one otherwise unnecessary re-fixturing at inspection (with datum D). Simple fact is that I don't care if the dowel hole is tightly located to the side exterior surface of the part - I care that it is located versus other holes in the part. The outside profile can be comparatively sloppy and function perfectly adequately in the design, hence my original spec.
2. Dimension QTY on center-marked AND connection-lined hole pattern. I've always just kind of assumed that if you have hole center-marks that are connected to each other, I put the dim to one center-mark with no QTY and the connection line implies that dim carries to all holes on the connection line - I've seen this in the Standard and for years at my previous employer. The Red Ink says to put the QTY of holes on that dim, i.e. "4X 7.0mm" but that strikes me as redundant to the purpose of the connection lines. I would say if there are no connection lines, absolutely add QTY. Also if there are different entity types, like an edge that aligns to the connection line or center mark, definitely add a QTY. But requiring QTY AND connection lines simultaneously bug me.
The justification for the "purely external and planar" datum reference, as explained to me briefly by the originator of the Red Ink, isn't necessarily a bad one where you're supposed to be making it easy for the machinist to identify critical dims, and ostensibly this part would have its exterior profile cut prior to the holes... Also of course the inspection setup is easier initially for ABC, but requires two setups because other hole location tols are to ADB because the dowel hole is now D. My counter-argument is that by making the datum references from explicitly-functional geometry, I am more clearly conveying the design intent to the machinist, identifying one surface and two holes as critical and tolerancing non-critical geometry like exterior profile as such. Personally I think I understand the implementation of GD&T better than Red Ink, who reset my reference frame but also leaves Datums B and C controlled only by limits of size - in my understanding, this actually makes it more difficult for the machinist to produce a part that will pass inspection because even if he can hit the tight tolerances from the exterior surface datums as the part is fixtured in the machine, once that part is fixtured to the datum reference according to the location tolerance callouts it will almost certainly fail unless the machinist understood implicitly that those exterior surfaces will have to be very nearly orthogonal for the tight location to be valid on the inspection table.
Sorry for the long post, but am I way off here? Regardless of what's posted back here, I'm not going and telling the senior engineer that my way is better... but for my own personal development, being self-taught GD&T and having almost zero direct fabrication/machining experience, which philosophy is better?
I just started a new job and got handed back a checked drawing with Red Ink all over it. The points at-issue:
1. Ordinate "Zero" aligned with datum surfaces. I don't operate under a rule where zero must be at a datum surface. Sometimes it is, sometimes not. My philosophy is that I dimension the part to be concise and clear on the drawing document, then I tolerance the part (with GD&T) to represent the functional/design requirements of an acceptable production. Red ink on my drawing re-sets my datum references to be all planar exterior surfaces and sets all my ordinate zeros to these surfaces. This is hard to explain with words and I can't post the drawing itself obviously, but my original datums were: A: A mating surface that interfaces with another part's surface via bolted connection, B: A dowel hole on A, C: a second hole with a tight diameter tol on A. The red ink keeps A as-is, sets B to one side of the part exterior normal to A, sets C to another side of the part normal to A and B, and adds a datum D which is the dowel hole. I acknowledge that this is totally acceptable to do, but here's why that layout bothers me: Too many controlled surfaces, and one otherwise unnecessary re-fixturing at inspection (with datum D). Simple fact is that I don't care if the dowel hole is tightly located to the side exterior surface of the part - I care that it is located versus other holes in the part. The outside profile can be comparatively sloppy and function perfectly adequately in the design, hence my original spec.
2. Dimension QTY on center-marked AND connection-lined hole pattern. I've always just kind of assumed that if you have hole center-marks that are connected to each other, I put the dim to one center-mark with no QTY and the connection line implies that dim carries to all holes on the connection line - I've seen this in the Standard and for years at my previous employer. The Red Ink says to put the QTY of holes on that dim, i.e. "4X 7.0mm" but that strikes me as redundant to the purpose of the connection lines. I would say if there are no connection lines, absolutely add QTY. Also if there are different entity types, like an edge that aligns to the connection line or center mark, definitely add a QTY. But requiring QTY AND connection lines simultaneously bug me.
The justification for the "purely external and planar" datum reference, as explained to me briefly by the originator of the Red Ink, isn't necessarily a bad one where you're supposed to be making it easy for the machinist to identify critical dims, and ostensibly this part would have its exterior profile cut prior to the holes... Also of course the inspection setup is easier initially for ABC, but requires two setups because other hole location tols are to ADB because the dowel hole is now D. My counter-argument is that by making the datum references from explicitly-functional geometry, I am more clearly conveying the design intent to the machinist, identifying one surface and two holes as critical and tolerancing non-critical geometry like exterior profile as such. Personally I think I understand the implementation of GD&T better than Red Ink, who reset my reference frame but also leaves Datums B and C controlled only by limits of size - in my understanding, this actually makes it more difficult for the machinist to produce a part that will pass inspection because even if he can hit the tight tolerances from the exterior surface datums as the part is fixtured in the machine, once that part is fixtured to the datum reference according to the location tolerance callouts it will almost certainly fail unless the machinist understood implicitly that those exterior surfaces will have to be very nearly orthogonal for the tight location to be valid on the inspection table.
Sorry for the long post, but am I way off here? Regardless of what's posted back here, I'm not going and telling the senior engineer that my way is better... but for my own personal development, being self-taught GD&T and having almost zero direct fabrication/machining experience, which philosophy is better?





RE: Style question with ordinate dims, datum surfaces
RE: Style question with ordinate dims, datum surfaces
In my opinion, you're not far off and neither is he.
John Acosta, GDTP S-0731
Engineering Technician
Inventor 2013
Mastercam X6
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II
RE: Style question with ordinate dims, datum surfaces
Frank
RE: Style question with ordinate dims, datum surfaces
Drawings (specifications) should be supplier agnostic to avoid this issue. If dimensioned and toleranced for function, the Supplier A versus Supplier B issue is avoided altogether.
RE: Style question with ordinate dims, datum surfaces
Just because holes line up on perfect centers or orientations does not mean they're the same diameter. Indicating the quantity of 7.0mm holes clears that up. Centerlines only indicate a location or pattern. They do not indicate that holes are identical.
_________________________________________
NX8.0, Solidworks 2014, AutoCAD, Enovia V5
RE: Style question with ordinate dims, datum surfaces
Ordinate dimensions can be made Basic.
Because Basic dimensions do not introduce tolerance accumulation, the "zero" point can be made anywhere, including where Manufacturing will like it.
The datums nevertheless can be assigned the way that represents function (as Engineering understands it)
RE: Style question with ordinate dims, datum surfaces
Of course, I believe it can be done or is allowable. Airplane manufacturers and Car companies often used universal origins.
Frank
RE: Style question with ordinate dims, datum surfaces
In MBD, there are no explicit dimensions other than size tolerances. If one wants to interrogate the CAD model, one can get what the basic dimensions would have been.
Auto OEMs are going this route. Most don't even put basics on drawings...with the exception of the powertrain and suspension groups.
RE: Style question with ordinate dims, datum surfaces
Model is basic, you may put your GD&T on drawing to have something reviewed and signed on the paper.
Unfortunately OP's problem is more of company policy / personal preference problem.
"We always did it that way" AKA "5 monkeys experiment"
RE: Style question with ordinate dims, datum surfaces
Model-based is what I'd prefer - actually I asked before I drew the part if I could put only tols on the drawing and send a 3D file for the basic dims. For complex geometry I think trying to dim and tol it just makes the drawing cluttered and confusing. I was told people are still manually programming the machine tools at the shops we use so the drawing should have all required dims and tols. I mean, I'll draw anything but that doesn't necessarily mean it's going to be easier to program... At my last company even the smaller shops had sprung for MasterCAM or similar in the last two years. In New England with all the aerospace up here I think that has been driving it.
In time I think (hope) there will be less oversight on my drawings and I can pretty much do it the way I think is best. Since I just started at this company all my drawings are up for thorough review and are being "corrected" extensively based on personal style and preference rather than adherence to the standard or actual correctness. So for now I'm just implementing someone else's preference and not putting up much of a fight.
RE: Style question with ordinate dims, datum surfaces
Tunalover
RE: Style question with ordinate dims, datum surfaces
I've seen this issue (#2) crop up here as well. My personal opinion is (that and $5 will get you coffee at Starbucks) that if there's only one way the drawing can be interpreted and it doesn't violate standards, it's good. So, if you have holes connected by center lines, to me it's completely clear that the same dimension applies to both and 2X would be redundant (but not necessarily wrong).