Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

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

Noobe Question on Design Intent 6

Status
Not open for further replies.

erk1313

Mechanical
Sep 30, 2009
26
This is a question about design intent vs. matching the steps a shop would use to create the part.

The example is a pin used to support a "table". The pin has a shoulder that is cut short to prevent it from showing through the top of the "table".

As the attached drawing shows, i've dimensioned the part from the edge of the shoulder (since that will be the critical reference point in determining the height of the table off the ground).

However, a colleague suggested I redo the drawing to match how the shop would make the part on the lathe: 1) the entire length of the shaft is measured and cut; 2)then the shoulder is measured and turned down. This would suggest an overall height dimension and a shoulder height dimension.

Which one is preferred?

Thank you for the help, and these great forums!

 
Replies continue below

Recommended for you

I would have dimensioned it more as your colleague suggests, though I would probably leave the dimension of the material removed rather than the shoulder height.

"Good to know you got shoes to wear when you find the floor." - [small]Robert Hunter[/small]
 
As I've been taught:

When detailing a part, the most important thing to capture is function. Detail it so that the functional requirements are captured and any part made in compliance to that drawing will work.

Secong is detailing for inspection, so that they can verify function. This usually comes for free with detailing for function but there are exceptions.

Third is detailing it to make manufacturings life easier. Do it when you can but not at the expense of capturing the function.

Also beware, often times design folks ideas of what is good for manufacturing isn't necessarily correct. Everyone around here used to dimension to centers of radii and the like, in part thinking it helped machinists. Then we acually had a vendor come in and say they'd rather have overal dimensions as it better suits their CNC/CAM and is easier to make allowance for tool wear etc. Funnily enough, this more closely matches the style of dimensioning in industry standards etc.

In your specific case there are at least 3 ways you could dimension the part. If I understand it, the length of the smaller dia portion is critical, so you probably want to directly dimension that with an appropriate tolerance. You then have 2 choices to give an overal length or dimension the larder dia. Overal lenght is slightly better for manufacturing but does it capture design intent? Does it lead to too much tolerance variation in where the fat end could finish, or tight tolerances to control this? These are questions only you can answer based on function of the long part of the pin.

How many legs does you table have? If 3 and it doesn't have to be super level then overall length is probably fine. If 4 or more and/or the table has to be very level then you probably become more concerned about the length tol of the large dia part so may be better off directly dimensioning that.

From a functional point of view it's all about tolerance stack up in this case.

At the end of the day, if the person creating the routing or CNC program can't add 3.375 and .230 then you have all sorts of other problems to worry about.

Posting guidelines faq731-376 (probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484
 
Much better reply than mine, KENAT! I think I'm starting to feel a little lazy.
Congrats on tipmaster!

"Good to know you got shoes to wear when you find the floor." - [small]Robert Hunter[/small]
 
I also agree with KENAT. Always design with function in mind. You are detailing the final product (what you hope to end up with), not detailing how to make the product (that is up to the manufacturer). ASME Y14.5 says to make the drawing represent the final product and only cover manufacturing processes if though processes themselves are critic.

Matt Lorono
CAD Engineer/ECN Analyst
Silicon Valley, CA
Lorono's SolidWorks Resources
Co-moderator of Solidworks Yahoo! Group
and Mechnical.Engineering Yahoo! Group
 
erk1313,

There usually is more than one way to do drawings like this. If the tolerances were way, way within the capability of the machine shop, I would be tempted to dimension everything from one end.

KENAT is correct. Your part has to work. Your inspector has to be able to verify that the part will work.

Also, I think that a tolerance stack analysis will show that you have done it correctly. The total length is much less important than the assembled height and the clamping thickness, which is what your drawing controls.

Critter.gif
JHG
 
Kenat offers some good advice. I typically design for function, and use reference dimensions to help manufacturing. So I would dimension the pin as you show, and add an overall length dimension as reference.

-- MechEng2005
 
As I was clicking "Submit" I realized a big mistake in my previous post... I always design for function, not typically. I typically detail the part for function, with refs for manufacturing.
 
Thanks to all for the great advice. As the table height is the primary concern, I'll keep the drawing as is.

I just read the previous post on dimensioning an angled hole, which addresses some of the same questions of function vs manufacturing. I seem to aggree that dimensioning based on function (even when it adds a step to the machinist) helps emphasize the importance of that particular dimension. It also makes my job easier, not having to second guess what the machinist would do.

Great site... and thanks for not shooting down a noobe's basic question.

 
Actually this is a very good question, even if basic. If more inexperienced folks asked these types of questions the average quality of drafting would probably be a lot better. Even recognizing the difference between function an manufacturability shows you have promise;-).

If you ask a manufacturing guy, or someone from a manufacturing background they may emphasize dimensioning for manufacturing more.

However, my understanding is that from a design point of view your priority is function with manufacturing as a secondary consideration, albeit a very important one.

Glad I could be of help, what with my post lask week on CSK's and now this one, I'm on an LPS roll!

Posting guidelines faq731-376 (probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484
 
KENAT is right on. Just read the Y14.5 Standard. One of it's fundamental rules is that the drawing is to describe engineering intent. Other parts of the spec require you to dimension in accordance with function.

The standard also asks that you avoid specifying means and methods. In the rare instances that you must for some engineering reason, you say "non-mandatory".

If you're the engineer, your part is the contract specification -- the drawing. That's how you work your magic. The production person is there to translate your specification into finished part. That's where they work their magic. You use the drawing for yours. They use route sheets or production control documents for theirs. Don't mix them up.

In your case, the next person may not want to use a lathe. Let the machinist get out a calculator. But in the meantime, he/she should manage tolerances in accord with the functional requirements listed on your drawing. Second guessing the desired means and methods of others on your drawings is a sure route to confusion and error.
 
I have to question adding "non-mandatory" when specifying a method. Doing so pretty much negates specifying it, making it a suggestion and not a requirement. The only time to include such information is when it is a requirement for some engineering reason.

"Good to know you got shoes to wear when you find the floor." - [small]Robert Hunter[/small]
 
I'll have to agree with ewh, if it's not mandatory, then process information shouldn't be on the drawing should it? ASME Y14.5M-1994 1.4e

Where it does talk about 'non mandatory' is 1.4f, but this is process dimensions rather than the process itself.

Posting guidelines faq731-376 (probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484
 
I think that's the point of the standard -- that specifying in-process dimensions should be avoided except in certain circumstances. If you're going to use the "non-mandatory" out in Y14.5, you should consider leaving it off. Personally, I think suggestions on drawings is a bad practice.
 
Agreed, but specifying an engineering required method of fabrication is not to be considered a "suggestion" and is mandatory.
Methods and in-process dimensions are two different animals.

"Good to know you got shoes to wear when you find the floor." - [small]Robert Hunter[/small]
 
Just a clarification on my reasoning in including the overall dimension as opposed to two stacked dimensions - purchasing.
While it would make little or no difference in the manufacture of the part to do it either way, including the overall dimension removes an opportunity of making an addition mistake when ordering material. It may not make the process fool-proof, but it will require better fools.

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

While it would make little or no difference in the manufacture of the part to do it either way, including the overall dimension removes an opportunity of making an addition mistake when ordering material. It may not make the process fool-proof, but it will require better fools.

Add a reference dimension to show total length. In properly managed CAD, this is an absolutely reliable thing to do.


Critter.gif
JHG
 
If the process capabilities can easily meet the required tolerances then dimensioning the overall length and which ever stepped dia length is more critical is probably the best way.

However, if tight tolerances are required on both step dia lengths, then this may not be appropriate. In this case dimension both step diameter lengths.

The reference diameter is pretty reasonably in CAD as it will get automatically updated, and for a drawing this simple wont clutter it too much.

However - if you put it in implicitly as a 'cut length' think about this. The reference dim has no tolerance, so really if using it as a cut length they either need to add the tolerance of the 2 lengths to make sure it's long enough and/or always cut a bit over length. Just for purchasing aid fine, but if they start to rely on it in manufacturing - which they shouldn't as it is only reference - then a problem. Reference dimensions, if people don't really understand what they are, can be dangerous. So on something this simple, I'd think twice about it unless you really trust your purchasers or vendor or whoever.

Posting guidelines faq731-376 (probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484
 
KENAT,

I think we are reading a lot into something that is absolutely simple. Almost certainly, I would fabricate this on a CNC lathe, and insert bar stock through the rear. It looks like holes will be tapped at each end, and one of those taps will have to be a separate process.

Someone has to figure out how much bar stock is required for the production run. Much more important than tolerances is the amount of material wasted by the cut-off process. More material is wasted due to the bar stock not being an exact multiple of the standoff's total length. If the person managing the process is too stupid to work out the total length, they are not going to figure out the rest of this. You have to give them some credit for competence.

Critter.gif
JHG
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor