Ordinate Dimensioning - When to use 1

[cpdpeckh]

When is ordinate dimensioning appropriate and when is it inappropriate?

My initial throughts are:

APPROPRIATE USE:
-Flat plate
-Single horizontal origin and single vertical origin
-?

INNAPROPRIATE USE:
-Hole patterns
-?

Please offer your thoughts.

Don

 

[ewh]

I would think it proper to use when there is a quantity of holes in a plate which would just add confusion if dimensioned otherwise.

 

[EngJW]

Some people will design a hole pattern with coordinates. I think it is mostly to help the shop, depending on the equipment they use. My preference is to design the hole pattern as a unit, but let its location on the part float a little. For example, an opening with a cover on the side of a casting could move around a quarter inch or so without any harm, but the bolt locations have to be held tight so that it goes together.

 

[xgrigorix]

I think that ordinate dimensions are great for hole patterns. It keeps your drawing from getting cluttered.

GK

 

[ewh]

As EngJW states, the hole pattern as a unit is important, and can be controlled with GD&T while being ordinately dimensioned.

 

[ringman]

EWH,

I believe that method of dimensioning is no longer allowed by Y14.5. I believe that it went out many years ago. It is replaced by composite positional tolerancing.

 

[ewh]

ringman,
I admit that it's been awhile since I've used ordinate dimensioning, but it is still allowed per ASME Y14.5-1994.
While I could not find any examples of mixing GD&T with ordinate dimensioning, the standard does state (para 1.9)"Rectangular coordinate dimensions locate features... from a datum or an origin." If the dimensions are basic and the datums defined, this should be an allowable method of dimensioning.
I do agree however that composite positional tolerancing is the preferred method for this type of control.

 

[MadMango]

I use ordinate dimensions for custom hydraulic manifolds. Knowing how the manifolds are made in-house, on each view I ordinate dimension from one corner, so the origin moves from corner to corner as the block is rotated as it is machined.

[green]"I think there is a world market for maybe five computers."[/green]
Thomas Watson, chairman of IBM, 1943.
Have you read faq731-376 to make the best use of Eng-Tips Forums?

 

[ringman]

ewh,

If you use datums and basic dimensions, the positional allowance has to be specified for the allowable variation.
That is where the composite tolerancing comes into play. It was formerly called 'PLTZ' and 'FRTZ'. Pattern locating tolerance zone and Feature relating tolerance zone.

 

[ewh]

ringman,
It is a given that the positional tolerance has to be specified. And I agree that composite tolerancing would be the preferred way to go. I was just pointing out that ordinate dimensioning is still a viable alternative, not necessarily better.

 

[ringman]

I respectfully disagree that it is a viable alternative to mix +/- ordinate dimensions with basic dimensions for locating a pattern of holes and be in compliance with Y14.5.

 

[Tunalover]

Folks-
If one uses GD&T positional tolerances IT DOESN'T MATTER HOW THE PART IS DIMENSIONED! It's the feature control frames that impart functionality into the design. Granted, while it's useful to dimension the part to imply functionality, when the dimensions are basic then it really doesn't matter. If fact, Y14.5 does not require that dimensions originate from the datums; you just have to have a way "to get there" using the dimensions.

As for me, I mix ordinate dimensioning and normal dimensioning in any way that helps fit all the dimensions into the view.




Tunalover

 

[ringman]

My 'assumption' which may be at fault, is that this discussion is using 'ordinate dimensions' to indicate dimensioning other than basic. If so, I still believe that the standard DOES NOT allow the mixing of such in the locating of a pattern of holes or such. The 'way to get to the datum features' has to be thru the use of BASIC dimensions.

Mixing was allowed WAY BACK WHEN!!!!

 

[TheTick]

ringman: ordinate dimensions can still be basic dimensions. "Ordinate/linear" is a quality separate from "basic/toleranced".

As far as use/don't use: use it when it works; don't when it doesn't. Do ordinate dimensions describe the part geometry in a satisfactory manner? If so, no problem to use, end of debate.

I could be the world's greatest underachiever, if I could just learn to apply myself.

http://www.EsoxRepublic.com

-SolidWorks API VB programming help

 

[Tunalover]

ringman-
Your statements indicate that maybe you are confusing the term "ordinate" with "bilateral."

You also said "I still believe that the standard DOES NOT allow the mixing of such in the locating of a pattern of holes or such" seems to indicate that you believe that bilateral, ordinate (or both) and basic dimensions may not be mixed. This is also not true. I recommend you alleviate all doubt and get back to the "horse's mouth" by revisiting Y14.5M which, incidentally, is "Dimensioning and Tolerancing..." not "Geometric Dimensioning and Tolerancing..." (sorry but I had to get that in there because there are many people who have the misconception that it is only GD&T).

If you take a good look, you'll find that bilateral ordinate, bilateral extended, basic ordinate, and basic extended dimensions are perfectly acceptable (but not necessarily desirable) in any combination on any given drawing.


Tunalover

 

[ringman]

Tunalover,

I see no problem in mixing the various methods on a given drawing. My statement stands that basic and plus/minus CANNOT be combined in the dimensional locating of a pattern of holes.

Stated differently: Datums and bilateral or unilateral tolerances not mix. Do we agree on that?

 

[TheTick]

ringman: Understood. I went back and reread your posts. sure enough, that's what you meant.

 

[Tunalover]

ringman-
You're right. Although Y14.5 never says it directly. A feature control frame (fcf) applied to a hole pattern controls the positional tolerance of each hole within. The spec says that in order for the fcf to apply, the pattern-locating and in-pattern dimensions must be basic. This excludes the use of lateral tolerances.

Lastly, the fact that a part has datums does not rule out the use of lateral tolerances (some features can be located with lateral dimensions and some by basic dimensions). Of course if the drawing had datums but no fcfs then the datums would be unecessary. In this case, while the datums would be redundant their presense would technically not be a violation of Y14.5M.


Tunalover

 

[ringman]

Gee, I sure hope that cpdpeckh got and understands all this. :>)

 

[cpdpeckh]

I have printed out everyone's comments and will sit down with ASME Y14.5M to make sure I understand what everyone is saying. Some dimentioning methods were described above in ways unfamiliar to me, and I need to understand restrictions on mixing dimensioning methods. Thank you for all the feedback.

Don