Translate into GD and T language: How? Which is the best way? 1

[greenimi]

Design intent: to control Ø.160/.156 hole location relative to the 1-1/16 hex.
What we don’t want end up having is a part with the hole positioned near the corners of the hex-see sketch-
No control to center hole or to Ø1.11/1.09 hole is needed.

I have 3 options to “translate” in GD and T language the above design intent.

Which one do you think is the one (if any) we should/can use? Are there any differences between these 3 options or they are producing similar results and have identical meanings.

Option #2 is to made to be able to have a functional gage built to verify the requirement. Can we achieve the design intent by using a functional gage? I understood Boundary note is optional.


What else do you think is missing from this drawing to get the design intent? In other words, how the part could look like if we have these options depicted on the drawing?

Do you think can be a better way—less complicated in GD&T language—to get the right parts?

Thank you

 

[fsincox]

pmarc,
Thanks, I find my time spent here, very valuable to me, due to the people like you.
Frank

 

[greenimi]

Pmarc,
To answer to your earlier post:
- there as no tangency at points where flats and arc meet, you are right about this.
- nothing on the print is saying what are the basic lengths for the flat or arc portions, you are right on this either, but the question is do we need these basic flats or arc portions dimensioned? If I specify the hex size 1-1/16, shoudn't be understood that the flats are equal = same length, and the angle between them is 120°-- all 6 angles--
I don't think we need that dimension p7 on your sketch (p7=0.34) for the sketch to be fully defined. At least in my CAD software I don't need it---of course with the above assumptions- equal flat lengths and 120°, defined IMHO by the HEX note-

Thank you

 

[fsincox]

CH,
I wish I had been there for that discussion.
The standard has been clear on this issue for a long time, so anyone who is "trained" should know this. They all have been trained.
The problem may be, as you said, spoiled manufacturing, expecting to be spoon fed. I suspect the real issue is the image of perfection; no manufacturing person wants to be held to perfection, I don’t want to either. I certainly never achieve it.
Frank

 

[pmarc]

greenimi,
My sketch was prepared also by assuming that:
- flat lengths were equal,
- arc length were equal,
- all angles were equal,
- everything was centered at the XOY origin,
- no tangency was at the points where flats and arc meet.

Maybe a 3rd opinion is needed?

 

[greenimi]

pmarc,
We might need "arbitration" here.
I just double check my sketch again. Looks like on mine I don't need that p7 dimension. And more than that on my sketch the p7 dimension is .28289 and not .34 as in yours. What's going on here?
Let's rephrase my assumptions ---beside the ones we already agreed upon, see above--
I've used 1.063 flat-to-flat hex dimension, Ø1.10 (radius .55) diameter (OD). That's all I need to make the sketch fully defined.
More than that the flat length for the "uncut" hex is 1.063* SQRT(3)/3 = .6137234
Now, if the flat hex is "cut" by the Ø1.10 diameter (.550 radius) the flat hex become .28289 (and this last dimension can be calculated also: A^2+B^2=C^2, so no additional constrained is needed)

 

[fsincox]

Good point, I never look at this kind of stuff as I tend to assume left off dimensions have just been left off, the standard does it too. Actually your hex note is redundant and conflicts with the dimension specified, technically (1.063 is not 1.0625). I would say the centering is implied by the standard, I still would use 6X and not "all around" unless it is actually defined all around.
Frank

 

[fsincox]

CH,
I also strongly agree on the slippery slope you mentioned. In this case it just seems we already lost that battle in the real world that I work in day to day.
Frank

 

[CheckerHater]

Thank you Frank,
This is why I feel bitter every time I see sales pitch about GD&T being “universally accepted”.
But enough about that. Let’s go back to our hex.
If you have G. Henzold’s book handy, check out Fig.20.65 thru Fig.20.69 and Fig.20.93.
Lot of interesting ideas

 

[fsincox]

CH,
Excellent, I would love to discuss with you the selling of GD&T vs. its use in the real world, I do feel that it is a contributing problem with the standard and that it has lead to some of the differences in ASME & ISO's approach, that is just my opinion, of course.
Frank

 

[fsincox]

CH,
I am waiting for that new book to appear on amazon and I plan to snap it right up. I want to get it from someone I know how to deal with.
Frank

 

[pmarc]

greenimi,
The value p7=0.34 is completely accidental. It could be 0.1 as well if I draged one of the end points of bottom horizontal flat more towards y axis. Then the lengths of all 6 arcs would increase. I just put this p7 dimension to show that the dimension is missing, not to show its exact value.

I must say I don't get what you mean by saying "the flat length for the uncut hex is...". Could you show it somehow on my sketch?

And of course I would love to see another opinion(s) on this.

 

[greenimi]

Pmarc,
If you are using 1.063 flat-to-flat dimension on the hex and 1.10 diameter, then the flat length cannot be any other dimension than .28289

For the uncut hex I mean: Let's say you have a hex with 1.063 flat-to-flat dimension. Do you agree than the flat length cannot be other than .6137?

Now if you put Ø1.10 (radius .55) in play the flat length became .28289 (and only this length)

 

[greenimi]

pmarc,
Sketches are attached.

 

[greenimi]

Here you are

 

[pmarc]

greenimi,
The sketches were not needed as I analized my sketch once again and indeed it seems that no length of flat (p7 dimension) is needed. I must have not defined one relationship originally. Thanks for being persistent enough to convince me.

Coming back to your original print, I would just add for clarity (as thin extension lines) those pieces of circle which are not part of true profile and associate basic dia. 1.10 dimension with the circle, for clarity. Thanks again.

 

[greenimi]

Pmarc,
I agree. I will add basic dimension on Ø1.10
And for this thread history:
I will scrap option #1 (from the original sketch at the begining of this post)--no value added,
keep option#2 as is, with all around symbol there.
On option #3 I will flip-flop the tolerance values (.005 versus .010).
And now, the remaining options (#2 and #3) preserve the design intent and provide similar results.

Thank you for you help