Dual Dimensioning - Tolerance Interpretation?

[erine]

Hi all,

I'm curious as to what the consensus is on dual dimensioning, when the drawing references ASME Y14.5M-1994.  I can't find any language in the standard that is specific to allowing or not allowing dual dimensioning.  Sections 1.4c and 1.4d both seem to say as much without actually stating that it is not to be used.

"( c) Each necessary dimension of an end product

shall be shown. No more dimensions than those necessary

for complete definition shall be given. The

use of reference dimensions on a drawing should be

minimized.

(d) Dimensions shall be selected and arranged to

suit the function and mating relationship of a part and

shall not be subject to more than one interpretation."

It seems that a drawing listing dual dimensions would require a general note, stating that one of the units of measurement is for reference only. Which in my opinion, would make it useless to dual dimension and only serve to waste space on the drawing.  A drawing without this note would allow a manufacturer to cherry pick tolerance conversions and still be correct.

[TomU]

This is a can of worms.  The best article I've seen is this:  Discussion about Dual Dimensions

These other two may also be helpful:

• Dual dimensioning rounding ?s / is it legal to say "dims in brackets are for reference only&quo - Drafting Standards, GD…
• Dual Dimensioning Standards - Drafting Standards, GD&T & Tolerance Analysis - Eng-Tips

[KenFarley]

If you dual dimension a part, you get the same actual dimensions in both units. This includes the tolerances. For example, a dimension that is +/- 0.010 in inches will have a tolerance of +/- 0.25 in millimeters. You can't have different tolerances for different units of measure, as far as I've ever seen.

The philosophy of having both is often, for us, a kind of courtesy, but also to make checking easier. If a customer uses millimeters, like the vast majority of the world, and we use inches, we will provide drawings for them to approve that have both units of measure. We could just make the drawing in mm, but that would make it difficult for our suppliers and machinists, etc.

[TomU]

If you dual dimension a part, you get the same actual dimensions in both units. This includes the tolerances.

This only holds true for values that perfectly convert from one unit to another.  Everything else gets rounded and by definition the two values are no longer equal.  Remember, all numbers are assumed to have an infinite number of zeros after them (per the standard).

Depending on the decimal place settings, Creo will convert 0.05 mm to .00197 inches, however these two are not truly equal (especially if the 0.05 mm is representing a MAX value.)

[mender]

For reference, this difficulty is why we made the general policy that dual tolerances round 'in', so that you can't measure an out-of-tolerance part in the secondary unit and erroneously find it to be within tolerance.  I'm not sure that there is any good answer here.

[TomU]

The apparent loss of tolerance after rounding causes us trouble.  Some of our customers require us to stay inside the tolerance range for both units.  Depending on how many decimal places the secondary units display, the loss of tolerances can be significant.  Our current workaround is to show the secondary units (inch) at 9 decimal places.  (It's not unusual for us to have dimensions where we need to hold a tolerance of +/- 0.01 mm.)

[mender]

A requirement to measure against decimal tolerances in two different units sounds painful for you.  But it does raise an argument for a detail setup option to say that the secondary units tolerance should be rounded by the same rules as the primary, (or even possibly 'outward' where we have 'inward'), because where someone is doubly constrained, making the secondary tolerance overly broad does not cause potential misinterpretation.  If this is of interest, ask after it in the usual way, and be sure to include the use case.

[TomU]

For reference, this difficulty is why we made the general policy that dual tolerances round 'in', so that you can't measure an out-of-tolerance part in the secondary unit and erroneously find it to be within tolerance.

By the way, this round-in behavior only occurs for limit dimensions.  For all other dimension types the value shown will often be incorrect for one of the secondary values.  This also includes GD&T dual dimension values.

[lyss]

The "round-in" behavior also occurs in Plus-Minus tolerances.

[dschenken · Accepted solution]

This has been a long term discussion in the drawing prep field. I don't have the version of Y14.5 earlier than 1994 handy, but according to the linked page dual dimensions were mentioned as specifically being removed in the 1982 version.

Discussion about Dual Dimensions

In general, I'd argue that dual dimensioning conversion can best be handled at the supplier end rather than on the drawing. At some point they give rise to some objection:

1) Accurate conversion means an excessive number of digits to preserve accuracy

2) If rounded, those using the rounded value eventually have the option to complain about the lost tolerance or that the nominal value has shifted a little.

3) Someone will notice the drawing is cluttered and probably want a separate drawing sheet that is done all in the alternative units. See 1 and 2.

4) Someone will misread the values from the clutter and mis-make the parts, leading to a bunch of finger pointing and lost work/time/sales.

5) They can be reference, and then it's just a waste of time and someone will want them to be not reference. See 1-4.

Short version: Nothing good will happen. At best, nothing bad will happen.