Dimensioning NO NO
Dimensioning NO NO
(OP)
We have a senior designer that dimensions parts with a theoretical starting point and dimensions from a 0,0,0 from the center of a part and dimensions everything off the theoretical center line. Is this wrong? No features to dimension from. I think this is the poorest dimensioning that I have ever seen. (Lazy) He says this is what the world is going to. QC cannot measure parts with no features.





RE: Dimensioning NO NO
However, If you want in depth answers I'd post over in forum1103: Drafting Standards, GD&T & Tolerance Analysis. Also state what if any drawing standards you work to.
Of course, if you don't work to any recognized standards then it's difficult to say he's explicitly wrong.
What is Engineering anyway: FAQ1088-1484: In layman terms, what is "engineering"?
RE: Dimensioning NO NO
This is a good question for the Drafting Standards, GD&T & Tolerance Analysis forum.
All dimensions must come from a feature that can be located by fabricators and inspectors. This can be a feature of size. You are allowed to centre 0,0,0 on an outside width or some internal width, which would be called up as a datum.
Locating and fixturing to features of size is not absolutely simple, especially if the FOS is not accurate. I would say that if the FOS is at least ten times more accurate than your other tolerances, you have no problems. After that, you need to think about MMC conditions. You (or he) should consider designing the fabrication and inspection fixtures. This may affect the design of the original part.
When your designer claims that "this is what the world is going to", is he referring to his choice of zero position, or his he talking about datum dimensioning? A lot of people do not like datum dimensioning. Perhaps he thought you were criticising that!
RE: Dimensioning NO NO
Here, not so much.
All I can conjecture is that's how he thinks GD&T works, which in turn implies that he could use some training.
Mike Halloran
Pembroke Pines, FL, USA
RE: Dimensioning NO NO
I'm not saying it is good practice, but that is the way it is.
Cheers
Greg Locock
New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?
RE: Dimensioning NO NO
There are applications where it can make sense, such as top level assemblies especially where there are nominal lines of symmetry, or maybe a C of G or something.
However, for a machined or similar part, it's probably not a great plan.
What is Engineering anyway: FAQ1088-1484: In layman terms, what is "engineering"?
RE: Dimensioning NO NO
Dan
www.eltronresearch.com
Dan's Blog
RE: Dimensioning NO NO
If the rest of your operation, i.e., QC, manufacturing, etc., can't use it (or have to go to extra lengths to use it), his work is, at best, a drain on the enterprise, and possibly completely worthless. Fire his arrogant @$$!
RE: Dimensioning NO NO
I am familiar with station numbers in aircraft, and I have used them on complex opto-mechanical equipment, way back in my drafting board days. It seems like the best way to keep track of the exact configuration of a complex assembly.
None of this wound up on my fabrication drawings. The assembly level station numbers had no relevance at the individual part level. I dimensioned from existing features.
RE: Dimensioning NO NO
Tobalcane
"If you avoid failure, you also avoid success."
RE: Dimensioning NO NO
If those 'center lines' - really center planes - are explicitly derived from a datum feature then it can be correct - as drawoh alludes to with his feature of size comment.
However, arbitrarily picking a 0, 0, 0, probably based off of the 3D model by the sounds of it, that is nominally 'in the middle' is generally bad practice.
While the OP is fairly brief, and maybe I'm reading too much into it, I get the impression this is the case based on the QC inspection comment.
What is Engineering anyway: FAQ1088-1484: In layman terms, what is "engineering"?
RE: Dimensioning NO NO
Usually works like this:
1. 0,0,0 point is the "airplane co-ord system"
2. Drawing uses this 0,0,0 to locate ONE set of datums.
3. Every other dimension is "local" i.e. from the part datum.
J
RE: Dimensioning NO NO
RE: Dimensioning NO NO
If these parts lend themselves to inspection by coordinate measure machines (CMM), which can record and digest hundreds of hard (x,y,z) points, input by a probe touching the surface, then the style of dimensioning you described starts to make more sense. I've never seen it done, but yeah, maybe the world really is going that way.
Otherwise, I agree its poor practice to dimension in this manner because it makes inspection impractical. Generally there should be tangible datum planes or points for 'open setup' in a gauge lab, or for functional Go/NoGo gauges.
RE: Dimensioning NO NO
I work pretty close to exclusively on automotive components nowadays, and the vehicle origin is at some defined place under the hood, with this origin used on most of the individual parts. This doesn't mean any of the dimensions on these parts reference that origin - they are all dimensioned to appropriate features of the part.
RE: Dimensioning NO NO
RE: Dimensioning NO NO
thread404-265463: coordinate dimensions
What is Engineering anyway: FAQ1088-1484: In layman terms, what is "engineering"?
RE: Dimensioning NO NO
As such we don't much use dimensioned lines for inspection, everything is a list of coordinates.
Note this is from my perspective only, where I am interested in establishing the location of bolt holes and surfaces and pins that are used to locate subframes into bodies, I do not know how the general run of the body surface is inspected.
Cheers
Greg Locock
New here? Try reading these, they might help FAQ731-376: Eng-Tips.com Forum Policies http://eng-tips.com/market.cfm?
RE: Dimensioning NO NO
It does amuse me when people say it cannot be done like this when in reality it is exactly how every vehicle is manufactured and inspected and has been more many years.
RE: Dimensioning NO NO
Every part peice is dimensiond frome the center line of the part no mater what the shape or if their is a feature or not.
I like feature to feature, and fit to fit dimensions to check tolerence of fit to the next part.
One other thing this guy outlawed GD&T at our facility.
Sounds like he just has no clue.
RE: Dimensioning NO NO
RE: Dimensioning NO NO
I've worked aerospace stuff where there was some use of global datums.
What is Engineering anyway: FAQ1088-1484: In layman terms, what is "engineering"?
RE: Dimensioning NO NO
On a optical sensor I worked on, I used the front lens of a big telescope as my coordinate zero. Later, the optical designer moved the front lens. At the system level, locating zero ahead of and below the car as Greg Locock noted, probably is based on much painful experience, on drafting boards.
This ought to have no effect on fabrication drawings, and on people running 3D parametric CAD.
RE: Dimensioning NO NO
Peter Stockhausen
Senior Design Analyst (Checker)
Infotech Aerospace Services
www.infotechpr.net
RE: Dimensioning NO NO
RE: Dimensioning NO NO
If you hand your machinists crappy drawings, they will often get the job done anyway. Of course, you will be the butt of jokes, and you will have no control over the final product, and the guy who joins you on the job will post to EngTips about the idiot he has to work with...
RE: Dimensioning NO NO
I agree. Crap in gives crap out. However, that wasn't quite what I meant, but my post wasn't clear on that.
The measure that I have used to evaluate whether or not a drawing is well-drawn is to look at from a machinist's perspective. If the drawing gives me every dimension that I need to fabricate the part with the tolerances either indicated or defaulted without requiring additional calculations or translations, then it's probably usable. If it indicates design intent in the process, even better.
Basically, I should be able to take the print to any machinist in the shop, tell them how many parts that I need and when I need them by, and never hear from them until they drop that many parts on my desk. I should be able to determine, using calipers, micrometers, radius gauges, etc, whether or not each part is correct.
However, from the OP's original description, the senior designer's technique would force me to calculate various dimensions so that I could actually both fabricate and check the part. This would not be acceptable to me (YMMV). If the machinists have to "make it work", then the drawing is not right, because each machinist will "make it work" in a different way, thereby generating a different and possibly incompatible part.
As for the "banning" of GD&T, I cannot express the depth of my feelings on that. While I still to this day have to look up the GD&T information when I need it or come across it, it's too useful a tool to get rid of. The only thing that I can think of is that he so completely cannot understand GD&T that he doesn't want anyone to use it and show him up, so to speak. But I'm just guessing there.
RE: Dimensioning NO NO
RE: Dimensioning NO NO
but IMHO dimensioning from a virtual reference can be done perfectly in digital world, but cannot be transferred to a non-digital world, at least not that easily/intuitive. Most of the time it simply does not exist, as on the drawing.
The quality control person will sratch hit head first and then step in your office with a question..
RE: Dimensioning NO NO
Dimension the part as it would be made and QC'd (re: functional gaging). If one feels the need to dimension to an abstract point, make them reference dimensions in addition to measurable features.
Example - I dimension a slot to be measured length and width with calipers. I reference dimension the center of the slot for our nc machinist.
MechE2
RE: Dimensioning NO NO
Dimensioning for function is primary per American drawing standards way of thinking.
What is Engineering anyway: FAQ1088-1484: In layman terms, what is "engineering"?