Multiple Angle Dimensions (Equally Spaced) 1

[EidenC]

We have a customer that dimensions gussets welded to circular plates on a regular basis and we have come to an argument on their dimensioning scheme.

I have attached an example and am wondering if the 6X 60 Deg dimension for the 6 gussets is an accepted practice as the stack up doesn't make sense.

If a welder welded all of the gussets going around the plate at 60.5 Deg apart, he would be way out of tolerance when he welds the final gusset on. The dimension would be 57.5 Deg between the last gusset and the first.

-or-

Does this mean that the welder has to find all of the angles theoretically perfectly and he has +/- 30’ off that?

 

[powerhound]

This comes up frequently. In my opinion, the drawing is incorrect as is for exactly the reason you state. The correct GD&T way to do it would be to make the 60 deg. dimension basic and position the gussetts with a tolerance of position callout. However, this is a welded assembly and for the most part welding is not precise, especially for gussetts. What I have done in the past in this situation is go 5X if it's okay that the last and first gussetts are 57.5 deg. apart. If the tolerance cannot stack up then more specific definition through the use of GD&T is probably in order.

Powerhound, GDTP T-0419
Production Manager
Inventor 2009
Mastercam X3
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[PeterStock]

While 6X does result in one of the 60 degrees being extra or reference, I have seen and used 6X for this. I have also seen and used 5X. I would consider either ok for this situation. Or leave the angle dimension off and just state "EQUALLY SPACED".

Peter Stockhausen
Senior Design Analyst (Checker)
Infotech Aerospace Services

http://www.infotechpr.net

 

[ewh]

This subject has been beat to death in the other thread, and I agree with the others here. What including all angles can do is indirectly indicate non-cumulative tolerances on the angles.

"Good to know you got shoes to wear when you find the floor." - [small]Robert Hunter[/small]

 

[powerhound]

Peter,

I have seen 6X used too; that doesn't make it okay. It is incorrect to use it in this situation but like a lot of drawings that are illegally specified, the part can still be made and could possibly still work. Again, that doesn't mean the drawing is correct, because a second maufacturing run will likely yield a different result -but still be compliant to the print- and that may NOT work. Using EQUALLY SPACED is also incorrect because they will NOT be equally spaced, every gussett spacing will vary, the question is how much can they vary and still work. How can you justify both 5X and 6X being okay? One is incorrect and one is not. When you start getting into the "they know what we mean" mentality, you are setting yourself up for a big disappointment eventually.

Powerhound, GDTP T-0419
Production Manager
Inventor 2009
Mastercam X3
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[drawoh]

EidenC,

I agree that the 60[°] angle tolerances are cumulative, however, there are six of them. If the specification was 5X60[°], the last angle could be out 2.5[°]. The "6X" means that all six angles must be within specification.

My point is a pendantic one, and I agree that basic dimensions and geometric position controls would make for a better drawing.

JHG

 

[KENAT]

Hmm, obviously no ones up for much of a fight this time, unlike previously.

thread1103-220265

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

http://eng-tips.com/market.cfm?

What is Engineering anyway: faq1088-1484

 

[cadman1964]

ahh whats the point....lol lol You can interrupt that drawing any way you want. Heck if our shop built it they would still screw up...lol lol

Solid Edge V20

 

[fcsuper]

Ya'no, drawoh's comments are both correct. This general scenario is covered in ASME Y14.5M. Check out page 22 of ASME Y14.5M-2009. The pattern would (in my mind) preclude the scenario you suggest because of the reason you state, the last angle would be out of tolerance, so it would not comply with the print. All angles much comply with the 6X dim, so the print prevents cumulation at a point. I think you are OK as is, though positional tols would give you what you are looking for as well.

Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group

 

[KENAT]

Even with Drawoh's point though there are potentially conflicting requirements.

However, I argued this strongly last time and don't think I convinced anyone.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

http://eng-tips.com/market.cfm?

What is Engineering anyway: faq1088-1484

 

[ewh]

"Good to know you got shoes to wear when you find the floor." - [small]Robert Hunter[/small]

 

[drawoh]

fcsuper,

I was just reading through the discussion that KENAT linked to. I must have taken some time on my responses there.

While cumulative tolerances do not affect the last hole in your series, they do affect holes in the middle. Adding two or three tolerances is not as bad as adding five or six tolerances, but you still are adding them.

For a given linear accuracy, angle tolerances to each of your features must get more accurate as the pitch circle increases. This gets very accurate very fast, even if all your tolerances are from the same reference point. This is the best reason to use GD&T positional or profile tolerances.

JHG

 

[NomLaser]

Typically I would use a basic dimension on the angle and position tolerance on the holes. This way the angle dimension isn't cumulative and you end up with a clear understanding on the drawing.

 

[MartinSr00]

With 6 gussets and 6 spaces, you've double dimensioned the part. That's not permitted by ASME y14.5 because as you've found, the reality depends on how you look at it, and Y14.5 gives you no real guidance.

Why not box the 6X 60 degree angles to make them basic, attach some datums, and use a feature control frame to tolerize location in the tangential direction? That way, your gussets will be controlled for both location and orientation. No stack up at all!

 

[fcsuper]

Actually, Martin, with patterns that's not true about double dimensioning. In fact, the standard states to call out the circular pattern in just that way. It's only double dimensioning if it is the same spec twice. In the case of a circular pattern, this does not happen.

Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group

 

[KENAT]

fcsuper, can you remind me where in 14.5-94 it says or shows that for non basic dimensions. I'm pretty sure it came up before and I thought it was being taken out of context or something.

Second thoughts forget it, we did this to death on the previous thread. To me if you have a circular pattern where you dimension all the angles with +- dimensions you end up with conflicting/ambiguous requirements.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:

http://eng-tips.com/market.cfm?

What is Engineering anyway: faq1088-1484

 

[MartinSr00]

5X on the angles could be OK, but it still doesn't address orientation.

I've done tons of drawings for a fabrication (weld) shop. True, there are AISC standards that leave drawing interpretation "loose". This is a matter of philosophy, and there is no "right" and "wrong" answer.

I would rather err on the side of being specific. Just because tolerances are loose, it's better to say just how loose. If it's rough, put a locational tolerance of .25 inch or so. Or .50. Then if there are problems, everyone is clear about what the standards for success are.

 

[fcsuper]

I looked this issue up in Genium's good ol' Drafting zone book. They address this point as an issue of concern and they do recommend PosTol's. They example they discuss is linear, but I think it still applies here.

Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group

 

[drawoh]

fcsuper,

A linear array does not wrap around the way a pitch circle does. If I show six holes positioned around a pitch circle by 6X the angle, the final dimension value may be redundant, but not the tolerance.

JHG

 

[fcsuper]

Confusing when I cannot tell who's on PRO and who's on CON. I don't know myself now. lol

Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group