Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Positioning Irregular Shaped Feature

Status
Not open for further replies.

JLang17

Electrical
Jan 16, 2009
90
I have an irregular shaped hole that is basically located and sized with tolerances (similar to the shape of this "0" if the left and right sides were flattened). I want to say it can be positioned with a dia. tolerance of .001 at MMC, but my book says I can only apply a positonal tolerance with cylinder, width or spherical features. A profile tolerance doesn't make sense because as the hole departs from MMC, it would break the tolerance zone.

I think I could use a positional tolerance attached to the height and one attached to the width, both with the note "BOUNDARY", but if they're the same tolerance it seems there's a better way.

Any help?
 
Replies continue below

Recommended for you

Do you mean you have essentially have a round end slot?

I think you are on the right lines, look at Fig 5-47 in ASME Y14.5M-1994 and paragraph 5.10.

5.10 explicitly says "The diameter symbol is omitted from the feature control frame."

5.10.1C says"...Where the same positional tolerance can be allowed for both {refering to length & width} only one FCF is necessary, directed to the feature by a leader and separate from the size dimensions.

If you need a sample diagram post again and I'll try and find time a lunch.







KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:
 
Why wait till lunch when this is more interesting that checking a half A$$ed assy drawing.

Link is to my understanding of what the standard means at 5.10.1C .

(Assume ASME Y14.5M-1994 1.1.4 regarding figures is invoked)

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:
 
OK, I missed of boundary below the FCF but hopefully you get the point.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:
 
Thanks, that's what I needed. I had actually shown it like that, but I just wasn't sure.
 
Looks good, Kenat. My preference is to use a composite profile of a surface to control the size of the feature (if possible) and then the location. That way it controlls the length, width & radii too. I find too many (engineers, machinists & inspectors) don't understand the use of BOUNDARY.

Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
 
MechNorth, we've had issues with them not understanding GD&T full stop. I believe they'd be more confused by composite profile of a surface than by 'BOUNDARY', but I've been wrong before.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:
 
Quite understand about working with people with limited understanding of GD&T. When I train, I emphasize surface profile & composite surface profile. I've had 2 students now that were using BOUNDARY without understanding what it meant (reference prints had it, and they were told to just use it), and they confirmed that they and the shop were both understanding it to mean the center plane in both directions. Afterwards, one converted to composite profile and the other stayed with BOUNDARY, though he at least understood it now and was ready to use it appropriately.

One of the most interesting things about GD&T per Y14.5 is that it often gives multiple ways to achieve essentially the same effect. Of course that's good and bad.

Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
 
Can you explain how a composite profile of surface works? I don't quite understand how to use it to control the size, orientation AND location.
Can size tolerances simply be converted to a profile tolerance (size +-.005 = profile tol. of .01)?
Is the size controlled in the upper frame or lower? My understanding is the lower frame does not control location...correct?
I also thought a composite profile of surface was only used with patterned features.
 
Simplisticaly (and I hope I get this right) the top line of the FCF controls gross location of the feature WRT the datums. The bottom line is a refiniment in size, form & orientation.

Look at 6.5.9 in ASME Y14.5M-1994 page 172 & related figures.

Also +-.005 translates to an equalateral surface profile of .01, more or less.

Composite profile of surface is not only used with patterns, in fact the example in the standard is a single feature.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:
 
At the moment, composite position tolerances are exclusively used with patterns (don't know if anything is changing in 2009). Is that where the confusion comes in, JLang?

At first I couldn't recall using a profile control on a pattern of features, but on reflection I have used a note underneath the FCF to the effect of "3 LOCATIONS". The composite profile control only controls individual features though, and does not control the inter-feature relationship. The features are part of a group as established by the principle of simultaneous requirements and the note "3 locations".

Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
 
Let's say I want to use a profile control for the feature in my original post, and the size tolerances are +-.005. I want to confine the location to a diameter of .001, and allow size to adjust within a zone of .01. How do I show this?
 
JLang, you're thinking a bit wrong.

With Profile you are no longer locating the theoretical center of the feature, you are controlling the location of the actual faces.

Also I don't think you can really think of location dia in the same way.

Simplistically, you'd approximate what you want by a top line of FCF with .011 and bottom line .010.

In the case you describe initially I'd stick with your original way of doing it with position & boundary.

KENAT,

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

If one has an irregular shaped feature and wanted to control it location including orientation, positional tolerances at MMC with "Boundary" is the best way. It is easily confirmed on the shop floor with a checking fixture with a pin of its virtual condition shape, size and orientation.

Profile of a surface shouldt be confirmed with a CMM and is very costly. If one saw this on the drawing, it would be confirmed at least once in the life of the part while the positional tolerance at MMC with BOUNDARY could be confirmed on an hourly basis and included in the control plan.

The only time I have seen profile of a surface confirmed on a regular basis using a template was when the tolerance was 3 mm on the trim of a stamping. I really don't think that the profile tolerance was of any value in this situation anyway.

Maybe those drawings that had BOUNDARY were from my past trainees since we are both in the same neck of the woods.

Dave D.
 
Dave, the two I referred to were in different parts of the USA, though one was automotive, and had a lot of dealings with Ontario suppliers...so, maybe.

The use of hard gages is diminishing as more shops move from very large / continuous runs to smaller batches and essentially one-offs. The costs just can't be justified in many cases, unless the tolerances are grossly sloppy, in which case I'd use a surface profile anyway. I've dealt with automotive specialty suppliers that do still use hard-gaging, so I'm not saying it's dead, just diminishing from what I've seen. I've been in engine foundries and fabrication lines, worked with component suppliers (including large assemblies) that gave up hard-gaging over a decade ago. They CMM on a statistical basis typically, and with increasing frequency use SPC to minimize / eliminate final article inspection. This allows a plant to detect and correct a problem before it becomes an issue. I'm also seeing more widespread use of vision systems, which lend themselves nicely to profile controls, particularly for relatively thin pieces. As for cost of using a CMM to check a profile vs a fixture to check a BOUNDARY; consider the size/cost and handling issues associated with a half-ton workpiece engaging multiple features on a checking fixture simultaneously ... a CMM (gantry or arm-style) is more cost effective & easier in many cases. Of course, operators & programmers need to understand the CMM and its software ... that's another story...
While I was an early advocate of CMMs, I came to recognize the issues associated with them (establishing datums from points on a selected surface, sine error, averaging algorithms, sampling rate, etc.) and like to point them out to students who rely exclusively on them for their results. Each situation may warrant different methodologies & equipment, and inspectors need to think things over before jumping in with their usual practice.

Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
 
Jim:

One offs or small runs, yes, go profiles although most only check profile (2 dimensional) of a line rather than a surface (3 dimensional).

I would recommend positional at MMC with BOUNDARY for high volume runs for an irregular shaped feature.

My background is in the quality area rather than design so I have been involved in statistical analysis for a while.

As far as CMMs and statistical analysis on GD&T, that becomes a completely different topic. Paul and I have discussed this many times in the past. Using a CMM on positional tolerances at MMC with BOUNDARY would blow anyone away. The results really wouldn't be fully confirmed but one would have data for stats.

The companies that years ago went fully CMM thinking that is was the best way are, in my opinion, wrong. There are applications for CMM and other times, a hard gauge will truly confirm the requirement. The above example of positional at MMC with BOUNDARY is an example.

Dave D.
 
I've rapped a few knuckles for checking a 2-D slice rather than the full depth; of course, even at full depth, only a sampling of points is taken, and hence not the entire surface is usually checked.

May be an example, again providing there are few features associated with it that must be gaged at the same time, and that the quantities justify the cost of the gage. We don't have enough info posted to be definitive.

Jim Sykes, P.Eng, GDTP-S
Profile Services TecEase, Inc.
 
KENAT,
you said:

"Also +-.005 translates to an equalateral surface profile of .01, more or less."

In this case, wouldn't the surface profile zone be .005? If it was .01 then the feature could expand .005 on one side, then .005 on the opposite side resulting in a total increase of .01, which is greater than the +.005 tolerance.
Just thought of this and wanted to be sure I was thinking about it correctly.
 
Sorry I was thinking of a single surface.

See attatched sketch.
The light bulb just came on. You meant +-.005 was the tolerance on the length or width of your slot, I misunderstood. In that case, yes you'd want to use .005 SP see second image.




KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at posting policies:
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor