Goal R.002 *Pulls hair out!* 5

[ModulusCT]

OK, so I was told to do the following by my supervisor today. I really hate having to tell him that he has no idea what he's doing and that he doesn't know the applicable drafting standard, so I thought that it would be better to simply offer a solution that does jive with the spec.

He wants to indicate R.002 +.003/-.000, but with an emphasis on trying to produce the radius at .002 rather than taking the total permissible error and shooting for the middle. His solution to this problem is to say .005 MAX; GOAL R.002, but I hate this personally because the word GOAL is not mentioned in the spec as an acceptable descriptor for tolerancing a part. In fact, I'm pretty sure there are parts of the spec that say not to do things like this.

Any suggestions? Is there a way that you can say, what my boss wants the drawing to say by using acceptable symbology and terminology?

 

[Walterke]

Well, what if I have one of those "situations" where an unequal tolerance makes sense?

The only time I (am thought to) use unilateral tolerances is when applying fittings. If you want a shaft to fit in a hole you dimension the shaft D+0/-0,5 and the hole D+0/+0,5, depending on the fitting.

In this case, I have no problem with the shaft being -0,25 and the hole being +.25. It's just easier (and in my eyes, more logical) then dimensioning the hole D,25+/-0,25.

I hope this makes sense.

NX 7.5
Teamcenter 8

 

[powerhound]

Waltereke,
Unilateral tolerancing on features of size is the only scenario I can think of that even makes sense. It's the only thing I've ever even had in my mind when responding to these posts so yes, what you have said makes perfect sense.

"Manufacturing will tend to do things in a way that is better for them..."

.......



Powerhound, GDTP T-0419
Engineering Technician
Inventor 2010
Mastercam X5
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[DeanD3W]

Rather than applying unilateral tolerancing on the two parts, I think it is better to make the stated value of each be the center of its respective tolerance zone and use an equal bilateral tolerance on each.

While the person designing the two parts may view the unilateral tolerances on each as logical, those making and inspecting the two parts may be in two different places. Since they're only looking at one of the two parts the tolerance won't look too logical to them. If the size tolerances are made a "critical" or "key" characteristic with Cpk requirements, the supplier will be driven to the center of the tolerance zone anyway in order to make Cpk as high as possible. This is another reason why I think equal bilateral is a better approach.

So, equal bilateral is better if Cpk is a requirement, safer in environments where the 3D model gets to manufacturing a week or two before the drawing does, and a bit easier to work with in most manufacturing environments since the stated value will be centered and that is also likely the target size, unless manufacturing has their own reasons to choose a non-centered target value.

Dean

http://www.d3w-engineering.com

 

[Belanger]

Well, then, can anyone give a logical reason why the GD&T system allows a unilateral (or unequal bilateral) profile tolerance?

If you all think that unilateral plus/minus tolerancing is useless, you should logically champion the same about profile.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com

 

[KENAT]

What's this 'all' business.

I've pointed out that for typical drill tolerance & similar it has merit.

For 'shaft basis' and similar standard 'nominal' size applications it has some merit though also can be problematic.

I don't think I even said never to use it.

Posting guidelines faq731-376

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

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[powerhound]

KENAT,
I don't see where it's problematic at all. The only thing I see as problematic are the +/+ and -/- tolerances that have come up in this thread.

Dean,
Bringing up Cpk takes this discussion in a whole direction that is different from what we've been discussing up to this point. So far we've only been talking about parts that are in tolerance or not. Cpk introduces not only if the parts are in tolerance, but where are they in the tolerance band and how close together are they to each other. So IF a designer was, in fact, trying to communicate a preference by calling out a unilateral tolerance, would he pretty much be SOL if Cpk was a requirement?


Powerhound, GDTP T-0419
Engineering Technician
Inventor 2010
Mastercam X5
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[KENAT]

The +/+ & -/- was what I was thinking of powerhound.

Posting guidelines faq731-376

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

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[MechNorth]

Discussion of +/+, -/-, and unilateral/unequal bilateral profile tolerances raise issues of modelling and drafting. If engineers design a part at its limit of size, then they have to use unilateral tolerances. From a detailing perspective, unilateral and unequal bilateral tolerances make sense as a convenience. That's one side of the fence, though. The other side is what manufacturing needs to work with, and that's a tolerance zone as opposed to a limit; they typically like to split the zone and go there as the target. Inspection just needs to make sure it's in the zone, so whatever the tolerance is on the print, they use.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[powerhound]

"From a detailing perspective, unilateral and unequal bilateral tolerances make sense as a convenience."

Maybe so, but it also makes sense from a manufacturing perspective if you actually care about what's actually being shown on the drawing. If you just don't care...well, you can't teach that.

"they typically like to split the zone and go there as the target."

Not me, and not anyone that I've ever worked with. Maybe it's not even a Texas thing. Maybe it's an Austin thing.

Powerhound, GDTP T-0419
Engineering Technician
Inventor 2010
Mastercam X5
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[DeanD3W]

John-Paul,
I tried to explain the type of situation for which an unequal tolerance would be the best choice in my post on the 23rd at 0:08... Since I addressed it to you I was hoping you would see it. Maybe you don't agree with the example..?

Dean

http://www.d3w-engineering.com

 

[ajack1]

Personally I can see both sides of the argument. But maybe that is because I don’t live in Texas.

I was brought up on the hole and shaft system and it makes perfect sense to me. Say you have a 32mm shaft and on it there are a series of conditions, maybe a press fit, a slide fit, a roller bearing, a phosphor bronze bearing, a vee belt pulley, etc.

You can look up all the fits for these that are pre determined and all manufactures that I am aware of for example that make roller bearing will call these up and show them and the fit required for the mating part, so whilst no one item will be 32 +/- xx they will all be based around 32mm and easily available as off the shelf items.

I have no idea how much you would increase the cost by making all bearings, pulleys etc one offs. Strangely enough all bearings are based around the ID or OD. I would guess if you don’t do this the logic is you would call up a 6.01 dowels pin, or whatever the figure would be rather than a 6mm dowel pin, now that would confuse me.

I can also see that where a complex model is involved to start putting unilateral tolerances on it means a complete remodelling job for someone, where machining directly to CAD is the only option.

 

[Walterke]

it also makes sense from a manufacturing perspective if you actually care about what's actually being shown on the drawing

Can you please point me to where in your standard it says you have to 'care' about the drawing?


Also, on a side note: people, please use quoting, it's really not that hard...
just type [quo*te name]"stuff you want to quote" [/quo*te]
(remove the *)


NX 7.5
Teamcenter 8

 

[fcsuper]

Walterke,

1.2 of ASME Y14.92-1942 specifically states that drawings must be cared for, both in content and use.

Matt Lorono, CSWP
Product Definition Specialist, DS SolidWorks Corp
Personal sites:
Lorono's SolidWorks Resources & SolidWorks Legion

 

[powerhound]

"Can you please point me to where in your standard it says you have to 'care' about the drawing?"

It doesn't say that anywhere. Sorry about the confusion.

Thanks for showing me how to do that with the quotes. I've always wondered how you guys did it.

Powerhound, GDTP T-0419
Engineering Technician
Inventor 2010
Mastercam X5
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[Walterke]

well Fcsuper then I humbly retreat myself from this discussion.
Though it was probably used in a different context.
'caring' for a drawing sounds a bit too much like 'common sense' to me, which is a very dangerous term to use, as most of you here will agree.

Powerhound, click the 'process TGML' link (not the box) under the reply box to find some more code you can use (bold, italic, hyperlinks etc)

NX 7.5
Teamcenter 8

 

[ptruitt]

When I need to fit a pin or bearing, I get out my 20th edition Machinery's Handbook or some bearing design guide on the internet. I will see unilateral and unequal-bilateral values and I will continue my design process thinking in that mode. When it is time to create the CAD model, I will convert it to equal-bilateral because the suppliers I use will usually want my CAD model and I know that there is less chance for tooling error if everything is presented to them equal-bilateral. (I usually design medium-volume products produced from molds.) So, at this point, I am reducing risk. But I will also be wondering if the supplier (undetermined at this point) will utilize inspection technology that may, at no extra charge, generate value-added process capability data that I can use as feedback when I work on the next generation design or use to compare which suppier is using less of the available tolerance. Recently, I have been receiving graphical inspection reports that show how well the actual part compares to my CAD model. I can adjust the length of 'whiskers' eminating from the CMM probe points that show if the part is trending bigger or smaller. Numerical data also shows how much of the available tolerance has been used as a percentage. The software best-fits the CMM data to the CAD model and equal-bilateral feels very, very intuitive. So using equal-bilateral tolerances may be beneficial if you want to posture your design for utilizing automation and other new inspection technology. Certainly there are many fine products that are either made in low volume or at higher volumes but within companies where established in-house methods effectively utilize unilateral and unequal-bilateral.

Peter Truitt
Minnesota

 

[ModulusCT]

Also, on a side note: people, please use quoting, it's really not that hard...

Awesome! Thanks!

I'm not a vegetarian because I dislike meat... I'm a vegetarian because I HATE plants!!

 

[fcsuper]

Walterke,

As I learned from an old mentor, there's no such thing as common sense. BTW, just in case it wasn't clear, I made up the STD reference as a joke.

Matt Lorono, CSWP
Product Definition Specialist, DS SolidWorks Corp
Personal sites:
Lorono's SolidWorks Resources & SolidWorks Legion

 

[axym]

"If you all think that unilateral plus/minus tolerancing is useless, you should logically champion the same about profile."

J-P,

I do champion the same about Profile.

The reporting of actual values and other numerical data for Profile tolerances has been much more difficult than it really needs to be, largely due to disagreements over how to handle unilateral and unequal bilateral zones.

With a unilateral specification, there is a discontinuity where the characteristic suddenly jumps from nominal to nonconforming. There is no way to report data for this type of spec that does not suffer from counter-intuitive or misleading results for certain cases.

I understand that the standard allows the use of unequally disposed zones, but I do not consider it to be a best practice.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[Belanger]

Dean ... of course I saw your post. But while you set up a good reason for using unilateral tolerancing, you admit that it doesn't really matter in the end. Thus my comment: Apparently all the greatest intentions don't change the idea that the unilateral tolerance means nothing different than an equal bilateral tolerance.
You also brought up the excellent point of capability. In this age of lean mfg, six sigma, etc., it's one thing to be merely legal, but there are many other parameters that are important when it comes to stacks; this is why I still hesitate to jump into the camp of saying that unilateral tolerances are meaningless.
I don't disagree with you in terms of what actually happens in the shop. While the max/min limits are really what make or break the part, I'm just saying that a unilateral tolerance communicates something.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

http://www.gdtseminars.com