Question on Perpendicularity 1

[KENAT]

I got a call from a vendor about a drawing I checked a few months ago.

I’ve attached a VERY simplified sketch giving you a rough idea of the situation.

On the prototype, which was copied from a similar older drawing, the central bore was (incorrectly) identified as datum and the bottom face was perpendicular to it. (The real part has some other diameters coaxial with the central bore so it was unclear what the datum was – small iso

http://files.engineering.com/getfil...b7-4751-9435-2e2de688d2e5&file=iso-perp.tif).

When I checked it I changed it to make the bottom face the datum, as this is where it’s mounted, and made the hole perpendicular to it – Rev A.

The vendor is saying this doesn’t suit the CNC program & machining methodology they’d come up with based on the prototype drawing and will cause them trouble inspecting it and possibly meeting the required tolerance and hence cost will likely increase. They’ve asked it be changed back the make the central bore the datum and have the end face perpendicular to it. Proposed rev B.

I’ve looked at 6.6.4.1 in ASME Y14.5M-1994 and am inclined to think that per (a) Proposed Rev B would be a correct callout although I’ve usually seen it done (and most of the figures in the standard are done) it as per rev A.

So, am I missing something, does making the change have some consequence I’m missing? Is my vendor selling me a line? Also am I missing something in thinking it’s virtually equivalent in terms of keeping the bore and surface perpendicular to each other.

Any input/suggestions appreciated. I have to send him something today, preferably within about 3 hours time, so any input would be appreciated. If you read this after that time feel free to put your 2C in as at least it will help educate me for next time.

Thanks.

Your file's link is:

http://files.engineering.com/getfil...3db7-4751-9435-2e2de688d2e5&file=iso-perp.tif

KENAT, probably the least qualified checker you'll ever meet...

 

[CheckerRon]

I would go with your propose Rev B as long as you are locating to real, measurable surfaces, not the centerline shown in your first sketch. I don't see why it would make a great difference unless the vendor has a CNC and maybe a CMM program set up and doesn't want the expense and bother of changing it.

 

[ctopher]

Depending on your design. You want the hole perp to the end or the end perp to the hole?

Chris
SolidWorks/PDMWorks 08 1.1
AutoCAD 06
ctopher's home (updated 10-07-07)
ctopher's blog

 

[KENAT]

Thanks Ron, I'll make sure on rev B that the datum is the central bore which from a function point of view is what it should be. I got the impression he thought he'd have trouble inspecting rev A with CMM. Also he had some manufacturability concerns.

Ctopher, that's what I'm trying to get my head around. What difference does it make if the hole is perpendicular to the end face or if the end face is perpendicular to the hole?

KENAT, probably the least qualified checker you'll ever meet...

 

[ctopher]

It would make a difference with the machinist for steps to machine it.
You show a hole related to once surface. Do you want the hole through the center axis, or perp to the one surface?
If perp to one surface, the hole could not be down the center of the axis.

Chris
SolidWorks/PDMWorks 08 1.1
AutoCAD 06
ctopher's home (updated 10-07-07)
ctopher's blog

 

[vc66]

KENAT-

Think about one of those nifty GD&T blocks, and you can visualize the basic difference. If the bore is the datum, then that is the feature that is used to constrain the part first. The bore will be straight and the plane will be off-kilter. However, if the Plane is the datum, that will be straight, and the bore will be off-kilter. The difference comes into play when you consider other features that may be referenced from one or the other.

V

Mechanical Engineer
"When I am working on a problem, I do not think of beauty, but when I've finished, if the solution is not beautiful, I know it is wrong."

- R. Buckminster Fuller

 

[drawoh]

KENAT,

For the sake of argument, I assume that your drawing is approximately to scale.

My first thought that the .001 perpendicularity tolerance on RevA is not equivalent to the .001 perpendicularity tolerance on RevB. The bore is longer than the face, therefore, the perpendularity calls up a tighter fixturing angle. For RevA, consider opening up the perpendicularity tolerance.

My second thought is that, as per ASME Y14.5M-1994, both RevA and RevB are valid callups. The fabricator is required to fixture to the primary datum first. It sounds like his fixture picks up the bore, and locates the part while he machines the accurate face. Your RevA at the very least, requires him to redesign his tooling.

How does the part work? Datum_A ought to be the primary mounting face.

JHG

 

[KENAT]

Thanks Ctopher & VC66, I've been having those thoughts.

Drawoh, you're right that the length of bore is significant'y longer than the face so that makes a difference, the light bulb has just come up over my head.

I asked the designer about this when I checked it but dont' recall getting a satisfactory answer and at the time didnt' think I could justify digging further. (Now the designer is out on maternity leave.)

It was based on function that I made the face datum A as this is how it mounts on the instrument in service.

This whole issue is confused by the fact the designer has a thing against GD&T and I made the mistake of trying to meet her half way on this one and am now regretting it.

KENAT, probably the least qualified checker you'll ever meet...

 

[CheckerRon]

Yes, that is the trap alright. Is this a certain Swiss engineer we both know?

 

[KENAT]

Might be.

KENAT, probably the least qualified checker you'll ever meet...

 

[ctopher]

People that are against GD&T are the ones that don't know it, or well enough.

Chris
SolidWorks/PDMWorks 08 1.1
AutoCAD 06
ctopher's home (updated 10-07-07)
ctopher's blog

 

[powerhound]

A+ to ctopher!! It's the classic fear of the unknown and one of the great myths of GD&T.


Powerhound, GDTP T-0419
Production Supervisor
Inventor 2008
Mastercam X2
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[desertfox]

Hi KENAT

I see nothing wrong with your REV A or B.
Here in the uk our standards say that the way it is dimensioned should not dictatae how its made, the dimensions datums etc are given to achieve a function of the component.
I have had lots of discussions in my time about altering drawings to suit how its made and its not always the correct way forward.

regards

desertfox

 

[drawoh]

desertfox,

I argued this when I took my GD&T course, and I was told that the datums specify the fixturing procedure. The logic of chapter 4 in ASME Y14.5M-1994 is that you are specifying how to immobilize the part. RevA's fixture is a tube, whose OD matches the MMC of the bore. RevB's fixture is a flat plate with a short, raised face matching the bore's MMC.

I suppose that if I do not watch what the machine shop is doing, and the parts pass inspection at my end, I should continue not watching them. If your tolerances are loose, there are several ways to achieve them, and if the shop passes the cost reductions on to you, all is well. If the tolerances are difficult, then you must prepare your drawings carefully, and be aware of how the fabricator is going to work. Careful specification of datums is part of that.

JHG

 

[dingy2]

I agree with Desertfox

Both Rev A & B are correct and it does depend upon its function which way to go and not how the part is processed. Does the part drop on the face and a pin goes into the hole OR does the part go over a pin and somehow the face contacts another surface.

"Function and Relationship" of the features to the part or mating part are paramount.

Thought I would put my 2 cents worth here.

Dave D.

http://www.qmsi.ca

 

[ringman]

Drawoh,
You had written:.....RevB's fixture is a flat plate with a short, raised face matching the bore's MMC.

After thinking about it for a while I wondered why the raised face would be a short one rather that the full length of the bore which is the datum feature. Should it not extend for the full length of the datum feature?

 

[vc66]

I believe you're correct, Ringman. It should be a gauge pin at the bore's MMC that is AT LEAST the length of the bore.

V

Mechanical Engineer
"When I am working on a problem, I do not think of beauty, but when I've finished, if the solution is not beautiful, I know it is wrong."

- R. Buckminster Fuller

 

[Xplicator]

I believe that Kenat got his question answered and has moved on, but since this is continuing, here is my 2 cents. Both ways are legal as it pertains to the standard, however a functional DRF comes into play based on the order of precedence. Additionally, I believe it isn't the method or order of the specification so much as it is the actual tolerance values specified. When kenat changed the DRF order to represent the actual assembly process (recommended) and transfered the perpendicularity FCF from the flat feature to the bore feature the requirement, he inadvertently tightened because of difference between the length of the bore wrt the flat vs. the flat wrt the bore. The tolerance needed to be changed to reflect the ratio based on the similar angle principle. I believe drawoh has already been starred for this contribution and Kenat acknowledged it.

The manufacturer was right in raising this flag about the processes capability to achieve spec value. However, in regards to the order of precedence driving a cost increase or even suggesting inspection issues, this is not acceptable from what I can tell about this case. This tactic is common with manufacturers whom have had success bullying customers into compromising their drawings, consequently the functional intent of the product to alleviate themselves from challenges. In the long run I really don't believe that fixturing cost would be very different here or even inspection cost for that matter. Not to mention that this wasn't the released drawing they quoted too, that's a procurement process issue.

"Soapbox" Moment-
I highly recommend that customers should require the manufacturer produce tolerance analysis results whenever they want a dimension scheme and/or tolerance change supporting that there proposed changes won't compromise functional intent. Unless it is an obvious oversight. Of course the customers are not absolved from any responsibility here if they haven't even completed their own tolerance analysis to quantify & validate their released spec (Dimension scheme and tolerance values) in the first place.

 

[desertfox]

Hi drawoh

I haven't got access to the ASME standard but I have scanned a document here which is from PD7304:1982 from the BSI educational section titled "Introduction to geometrical
tolerancing"
Look at clause 2.5 on page 3.

To me which ever way you look at it the face and the bore must be square within .001" from any point on the datum and on the face perpendicular to that datum.
Therefore even if KENAT changes the datum to the face and not the bore and the people machining it use the same fixture they used previously there is no reason why the part in my opinion should not meet the drawing requirements.
You state that on your course you were told the datums specify the fixturing procedure does it actually state that
ASME Y14.5M-1994

regards

desertfox

 

[drawoh]

ringman,

You are right. I got my revisions precisely backward. I should have gone back and looked at the original drawings.

RevB's fixture is a tube extending the length of the part, whose OD matches the MMC of the bore. You would slide your part onto the tube, and test the end face for perpendularity.

RevA's fixture is a flat plate with a short, raised face matching the bore's MMC. The end face would contact the fixture at the three points that extend the furthest down. The bore locates the part in X and Y at the end face. You then test the bore for perpendularity. I am starting to see why the fabricator wants to use the bore as a datum.

The ASME standard does not explicitly state that you design your fixture as per the datum specification. My GD&T instructor did. The standard does state that the part is to be immobilized by datums which are specified with an order of precedence.

As I noted above, it is good practise to use your primary mounting surface as your primary datum. We still do not know what this surface is. KENAT has inherited the drawing and the manufacturing procedure. If the process is working, perhaps it should be left as is.

JHG