True positon of a group of holes

[Spurs]

I have a customer drawing that was created by a group in Detroit. They did the GD&T. But of course the customer SQA person is in Mexico. They have a different interpretation of how a gage should be built for this - I am caught in the middle although I agree with the Detroit Engineer. Could I be wrong?

Here is a section of the part drawing

http://files.engineering.com/getfil...-4ffe-8569-f6612387e976&file=part_drawing.jpg

The control frame for true positon has two lines in the callout. The engineer and I interpret this control frame to mean that if we use gages to measure this (as opposed to CMM) we need 2 gages.


The first gage would have a pin at the mmc size of datum A 12.065 sticking out of a plate (DATUM B) Then around that pin would be 6 other pins spaced 60 degrees apart at a bolt circle of 40 mm. These 6 pins all have a size of 5.1-.05 (hole size tolerance) - .2 (true positon tolerance)= 5.85 mm. The part must fit over this gage into all holes simultaneously.

The second gage would be similar except there is no datum A pin and the 6 pins have a size of 5.1-.05 (size tolerance) -.05 (true positon tolerance)= 5.0 mm.


The Mexican SQA says it is only a single gage similar to the first one I mentioned except the size of the 6 pins is 5.10 - .05 (size tolerance) - .2 (first position tolerance) - .05 (2nd position tolerance) = 4.8 mm diameter pins.


Which is correct ?


On another note: Does the 6x marking automatically tell you to combine all 6 holes into the measurement simultaneously? one other person suggested that the first line of the control frame means that you build a gage with the datum A pin at 12.065 and only a single pin at 5.85. Then you test each of the 6 holes individually. My assumption is that the 6x implies that the holes are to be treated as one group. Is this true or just an assumption?

 

[Nescius]

I believe it is frowned upon to solely control perpendicularity of a hole relative to a planar datum feature by referencing that single datum.

The rule of simultaneous requirements applies regardless of the 6X notation. Even if every hole had its own control frame, all holes must still conform simultaneously unless it is specifically noted otherwise.

 

[Spurs]

Nescius

"I believe it is frowned upon to solely control perpendicularity of a hole relative to a planar datum feature by referencing that single datum."

If you are referring to the lower box, the intent is not to control perpendicularity. The intent is to control the relationship of all 6 holes to eachother and datum B without tying them to datum A

"The rule of simultaneous requirements applies regardless of the 6X notation. Even if every hole had its own control frame, all holes must still conform simultaneously unless it is specifically noted otherwise."


Is this stated in Y14.5 somewhere?

 

[Nescius]

I see; it's intended to be a composite positional tolerance. The first box of each segment should be merged into a single box containing one positional tolerance symbol.

Section 4.19 of Y14.5-2009 says:

"A simultaneous requirement applies to position and profile tolerances that are located by basic dimensions, related to common datum features referenced in the same order or precedence at the same boundary conditions."

As a counterexample to the notion that the "X" is the indicator for a simultaneous requirement...What if I had 6X holes at one size, and 6X holes at another size, all needing simultaneous control? A more pointed example is in Fig. 4-39 of the 2009 standard, where it's a bunch of holes and a profile under simultaneous control.

 

[pmarc]

Spurs,

The numbers from Detroit are correct, however I doubt that the datum features precedence in the upper positional callout is trully functional for this geometry. I would think B should be primary and A secondary.

You are correct that the lower feature control frame controls not only perpendicularity to datum plane B, but also spacing within the pattern of 6 holes.

As for where it is stated that: "The rule of simultaneous requirements applies regardless of the 6X notation. Even if every hole had its own control frame, all holes must still conform simultaneously unless it is specifically noted otherwise." It is explained in para. 4.19 of Y14.5-2009 (page 76).

With the '6X' notation the holes are treated as pattern (according to the patter definition given in para. 1.3.42), thus 6 pins must be used simultaneously, not one pin individually for each hole.

 

[Nescius]

pmarc,

Am I mistaken about the single position symbol in a common first cell of each segment in a composite positional tolerance?
I believe the datums are referenced in a way that breaks the rules of composite tolerancing, too. What am I missing here?

 

[Spurs]

Pmarc & Nescius
Thank you for enlightening me
I am glad my perception was not that far off of what it really meant. Now to convince a Mexican - He doesn't want to pay for 2 gages - he thinks it is possible to make this in one gage.

 

[pmarc]

Well, you are missing the fact that in the upper callout datum feature references are |A(M)|B|, meaning that if you want to legally use composite callout with reference to B only in the lower segment, you have to switch datum features in the upper segment to |B|A(M)|.

 

[pmarc]

Spurs,
It is technically doable with only one gage. The gage simply has to have a removable pin simulating datum feature A. For more details on gages and fixture design you may want to take a look to ASME Y14.43.

 

[Nescius]

Thanks, pmarc, I edited my last post while you were typing. I'm slowly putting the pieces together...

My apologies to to you, Spurs, for my screwball assumption.

 

[pmarc]

I also have to apologize Spurs, because my last reply was not correct. The pins verifying position of 6 holes would also have to be removable in order to be able use two different sets of 6 pins (with different diameters).

 

[pylfrm]

Spurs,

I agree with pmarc that the datum feature references for the upper positional tolerance are probably backwards functionally. However, unless the orientation error between datum features A and B is fairly large, I don't think it will make much of a difference for the position tolerance. For instance, 20 mm * (1 - cos(1°)) = 0.003 mm or so.

If you're willing to make some assumptions about the orientation errors of the holes, and verify as appropriate, then I think you could accomplish a lot with a single gage as follows:

Flat surface to simulate datum B.
6X pins at (5.10 - 0.05 - 0.05) = 5.00 diameter.
1X pin at (12.075 - 0.010 - 0.200) = 11.865 diameter.

I haven't fully thought it through, but I think this gage would be exactly correct if the holes all have perfect perpendicularity, and would generally become more conservative (reject more good parts, not accept bad ones) as orientation errors increase. Pin diameters could perhaps be adjusted slightly to account for any errors inherent in this approach.

A drawing change could of course make this gage completely correct, and may represent the functional requirements just as well.


- pylfrm

 

[3DDave]

One thing that is missing is the relationship between A and B, so one can't reliably understand the condition of B relative to A(M), making the guage impossible to define.

Other than that the precedence is probably wrong. It says to fix the orientation of the part with a fixed diameter LMC pin that corresponds to the MMC center hole and then let some datum simulator plane perpendicular to the A simulator touch B at some point. With +/-.020 there is a 1/4 degree orientation when the hole is at the maximum size. The Axes of the referencing holes are projected from that plane, parallel to the A simulator/perpendicular to the B simulator, which means those hole true positions can wander a lot from where they are expected to be.

The Mexican guy is completely wrong.

Whoever is in Detroit should probably put down the pencil or check the drawings more carefully before they go out.

 

[pylfrm]

On second thought, I calculated the center pin diameter incorrectly. Because the 6X pins are already reduced by .050 from the hole MMC, the center pin only needs to be reduced by .150 to allow for the total .200 tolerance. It should be:

1X pin at (12.075 - 0.010 - 0.200 + 0.050) = 11.915 diameter


- pylfrm

 

[3DDave]

pylfrm - the diameter of the datum A simulator should be the MMC diameter of the hole 12.075 -.010 or 12.065. It is unaffected by the tolerances allowed to the other holes.

 

[greenimi]

I agree with 3Dave:
Gage 1: Check PLTZF: size: central pin 12.065. and size of six pins 4.85 each
Gage 2: check FRTZF (if composite used) size of six pins 5.00 (no central pin needed)
Gage 1 and gage 2 could be combined (as pmarc suggested) in one set of gage assembly with replaceable pins depending which callout is to be verified. Central pin shall be movable along with the six pins to verify the pattern of the six holes position.

 

[pylfrm]

I realize what I suggested doesn't correctly inspect the tolerance on the drawing. Spurs had the correct interpretation in the original post.

My suggestion would be correct if upper tolerance were removed from the 6X holes, and the the center hole were toleranced with position diameter 0.15 (M) to datum feature B, right? What I'm getting at is that these are actually very similar tolerance schemes, especially if the holes are accurately perpendicular to datum feature B. Both schemes allow 0.2 relative displacement between the 6X holes and the center hole.

No gage is perfect (due to gage tolerances), so I think abandoning theoretical perfection as well can be justified in certain circumstances.

Have I missed something here?

- pylfrm

 

[axym]

Spurs,

You could design one gage to check both tolerances. The concept is to imagine a single FCF that would covers the requirements of both the FCF's on the drawing. This could be done with POS|dia .05|A(M)|B|.

The gage would have a 12.065 pin for datum feature A, and six 5.00 pins for the hole pattern. If the part fits on this gage, then it conforms to both FCF's.

This approach is obviously very conservative and could reject a lot of conforming parts. But that is the price of simplicity - the requirements become tighter. With GD&T there is usually a trade-off between tolerance and complexity - to allow more tolerance, the specification (or inspection method) must be more complex.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[mkcski]

I am a little late to this post. I have scanned through the posts pretty fast. But as I understand it, GDT is intended for product definition and consequently is supposed to be a functional dimensioning tool. I didn't see where "spurs" described the function and fit-up requirements of the design? It would be much better to start from that basis and develop an accommodating schema, then discuss datum/feature relationships, appropriate controls and suggest "cartoon" gauges.

 

[greenimi]

Usually when the threads are posted the people need help on a specific problem and less on a full product definition. (full design intent is for the most part confidential).
That is the designer’s job and the OP’s (in general) are not aware how the part functions in the assembly. There are exception to this too.