Least bad of two options: Circular edge or Cone as tertiary datum? 2

[2JL]

Hello Everyone

I am currently dimensioning a part machined from a die casting and I find myself in a peculiar situation where I need to make a choice between two bad options. I need a tertiary datum to ensure that a pattern of threaded holes is in a correct angular alignement (no precision is needed) relative to other features. No machined features is available for this purpose so my only option is to use a conical hole or derived a point from a circular edge.

http://files.engineering.com/getfil...f86-bfd9-e17b56656398&file=Tertiary_datum.png

What do you think I should do? Thanks for your time and suggestions

2JL

 

[greenimi]

So, if you are talking about the machined drawing THEN the machined drawing should be defined as you said from the functional requirement pointy of view. I agree with you here.

Therefore, datum feature A (primary) should be the face and datum feature B should be the central bore. If these two datum features are NOT the functional datum feature, please kindly advise. Are they or not?

If they are not, which feature are?
If they are, then the secondary datum feature B must be oriented to the first datum feature A (so, perpendicularity is the correct control).

“A” is the first thing in the universe (the universe for this part = the coordinate system for this part).

“B” is the second thing in the universe and is not located, but is ORIENTED to the first daum (primary) datum feature.

THEN, you are going to locate anything else to “A” primary and “B” secondary. So, you are going to locate the pattern of the holes to A primary and B secondary, and are going to locate the OD to A and B and are going to profile the feature of non-size to A and B, etc.

That would be my opinion.

I am 100% sure CH, you already know this, but you are just trying to have fun. I am perfectly fine with that. Hopefully, the OP will not get confused.

 

[CheckerHater]

I am out of this conversation.

"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future

 

[greenimi]

CH,
Why?
Should I understand that you agree with me?

Anybody else?

Belanger, pmarc, Evan, 3Dave, just to name a few of the MVP's of this forum.........

 

[SDETERS]

how are you holding the part to machine the features and holes? How is the part being held? What are your Raw Material datums? Are you machining these away?

 

[2JL]

Hi SDETERS

My idea is to describe the functional requirements of the final part then the raw material (ie the remaining as-cast features) will be checked against the machined datums. Since we aren't producing the parts in-house, the goal is to let the process expertise to the supplier. My only concern is good parts no matter how they are handle during manufacturing.

2JL

 

[greenimi]

Again the same distinction must be made between the product drawing versus the process drawing. When this is clarified a suitable solution can be proposed

 

[CheckerHater]

greenimi,

Could you please reference standard that describes distinction between product drawing and process drawing?

"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future

 

[greenimi]

There is no standard for product versus process drawings.

If the overly used “functional” word/ functional requirements in Y14.5 and Mark’s statements (above) are not enough for you, I will rest my case.

The question is: are you seeking solutions or are you in love with your problems?

 

[CheckerHater]

greenimi,

I suggest you keep personal insults to yourself.

With all due respect to Mark Foster, he appears to live in imaginary world where there is no need to make parts from his drawings (see his statement: "However, in *MY* company (if we were a manufacturing firm), these drawings would be controlled by Manufacturing (i.e. NOT Design), and they would the words, "PROCESS DRAWING" in big red letters stamped across the drawing")
Still, I would like to see your own personal opinion, as deep, as Mark Foster's.

In real world there is only one drawing - "drawing", and there is only requirement - to provide all the information necessary to make good part, otherwise it's a "bad drawing".

ASME Y14.8 clearly prefers "Separate views drawing method", that is, casting/molding and machining requirements are shown on separate views / drawings.

OP was very clear saying that "we have separate drawings for cast and machined parts"

Now, open Y14.8-2009 and look at Fig.2-1

You will see that cast features are defined using their own, separate set of datum features (A, B, and C on Fig.2-1 (a))
The machined features are defined using their own, separate set of datum features (D, E, and F on Fig.2-1(b))

Accompanying paragraph 2.2 states importance of specifying cast-to-machining relationship. You can see that on Fig. 2-1(b)dimensions are provided to relate cast datum features to machined datum features.

All this is in letter and spirit of ASME Y14.5-2009 Para. 4.8.1 Temprary and Permanent Datum features - cast datum(s) serve as temporary datums for establishing machining datums.

Now, OP has no control over cast features. He is asking for advise how to best relate machined features to cast (datum)features. Your answer is "Simultaneous requirement".

How possibly simultaneous requirement can control both cast and machined features if THEY BELONG TO TWO DIFFERENT DRAWINGS?! And you say I have problems.

Simultaneous requirement is a powerful tool and shouldn't be used to cover sloppy design work. And I agree - the word "functional" IS overused.



"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future

 

[greenimi]

CH,

"How possibly simultaneous requirement can control both cast and machined features"
Because both features (machined and cast) are on the same piece of meat regardless how you make them (casting, forging, machining, etc). They have to work together. They have to have a relationship to one another.

With all due respect:
How would you do it? So far, just dismantling someone else “potential solution” without offering other to replace it does not help CH.

And also offering conflict statements does not help either:

Statement 1
“If you look closely at option 2, you'll see that you don't need secondary either! You just call everything wrt [A] and argue Simultaneous requirement.”
I am not sure you were sarcastic here or not

Versus statement 2
“Not only you will need tertiary, you may need temporary datum to locate your datum to the raw part.”

Are these two statements in conflict to each other or it is just me? If it’s just me again the question, how would YOU do the drawing (the good drawing)

Your quote: “In real world there is only one drawing - "drawing", and there is only requirement - to provide all the information necessary to make good part, otherwise it's a "bad drawing".”

And rest assured I have a ton of respect for you for the time spent here on this forum.

As for “Still, I would like to see your own personal opinion, as deep, as Mark Foster's.” CH, you cannot see that because…….. is very simple…. I don’t have it as deep as Mark’s. Sorry about that. I am still learning. But that does not keep me from getting, from time to time, in conflict with the MVP’s of this forum.

Therefore CH, how would you do it?

 

[CheckerHater]

I am sorry I forgot that Internet filters sarcasm. So, scratch Statement 1.

How would I do it?

First I would produce drawing of the casting. Here everything could be subject to Sim Req, because, after all, it's all being cast together in one chunk.

Second, I would produce final (machined) drawing, paying attention to:
[ul]
[li]Concentricity of central bore. Concentricity is never implied, so two questions: how concentric and concentric to what (I think you already guessed it)[/li]


[li]Position of tapped holes. Techniques may vary, but clocking wrt cast holes is making sense.[/li]
[/ul]

I have to go to do some of my actual job, but I will try to produce couple of pictures, if you are still interested.

And suggesting that you are not deep enough, was unacceptable. Sorry

"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future

 

[CheckerHater]

OK, as I promised, I am including some pictures.

The case is simplified (maybe oversimplified), and I tried to use (maybe overuse) Simultaneous requirement.

Casting is shown mostly for illustration as OP has no control over casting. Suggested tolerance scheme is not the only one possible, but the one using Sim. Req.
I added central hole to casting, as it would reduce amount of subsequent machining.

I like simple elegant solution as much as the next guy, but when there is possibility of misinterpretation, it’s better to explain yourself in finer detail, maybe even verbal note if necessary.

On machining drawing dimension 1.900 ties casting datum to machining datum.
It would be easy for machinist to center the bore with the existing central hole, but after central hole being removed, there will nothing to check against. So, we center the bore wrt cast hole pattern.
Then we position tapped holes wrt the same hole pattern. This creates Simultaneous requirement – the bore and tapped holes are located loosely to casting (per OP’s “no precision is needed”), but the bore and the tapped holes now form one single pattern.

Like I mentioned, this is not the only possible solution, but it uses Sim. Req., takes in consideration both Y14.5 and Y14.8, agrees with common sense and principles of machining and QC.


"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future

 

[greenimi]

CH,

Okay. Agreed.

Just a couple of comments I still have. Please see if you agree with them or not.

On your casting drawing: I do not know if using position for the central hole just to force the simultaneous requirement is the right (or even legal for that matter) thing to do or not. Personally, I would use perpendicularity and that is because orientation is the only thing you CAN do for that hole. The central hole cannot be located to anything else.

Another thing is when defining castings / forgings I prefer to use the last letters of the alphabets. So, in our discussion I would use Z as the primary datum for the casting (I know that is a minor point, but just to make sure). Probably it is an industry best practice to use Z, Y, X, etc. for castings and A, B, C, D, etc for final product.

Sorry, I do not have any capabilities to change/update your pdf at work.

On your machining drawing (and I will call it product drawing) I would have a couple of corrections (again, IT IS my personal opinion, since you asked for it):
- Your shown datum feature A to become datum feature Z
- Datum feature C to be moved from the lower/ smaller hole to the central hole (big change versus what you proposed)
- Remove the positional callout for Ø.2.250 hole and add perpendicularity to B (datum feature for machining) and another perpendicularity to Z (datum feature for casting)
- All the other applicable holes will be controlled by the following two single segments positional (not composite) callouts: upper line position to B primary and C (maybe at MMC) secondary; lower segment position (maybe at LMC since it is casting) to Z.
- Since you made the OD basic, I have no problem with that and I would just add another profile (for machining) to B primary and C secondary.

Thank you for your time and cooperation. Any comments or concerns with “my revised drawing” please let me know.

 

[greenimi]

See attachement

 

[greenimi]

That attachment is from another thread and I use it as a reference to create "my solution". It is from Don Day -- Tec. Ease -- and it is shown how to dimension on the same drawing casting requirements (process drawing) AND machining requirements (product drawing).

Hope helps more than confuse.

 

[CheckerHater]

Thank you greenimi,

I am glad you agree that there is more than one way to skin the cat.

I am a bit uneasy with your way of using perpendicularity, because it still orientation control. I said it before, just because it is shown in the center, doesn't always say HOW central it is. So, concentricity, position, runout - something.

On my machining drawing datum C is actually pattern of three holes, which may be used to derive central axis - I don't know how much that will change the way you see it.

If you still insist on making central bore secondary datum for machined part, I would simply use cast hole for clocking as OP suggested.

I really don't want to start another fight over every tiny detail. My example was to show general idea of separating cast and machined features.

Just like your attachment shows, cast part may see heavy use of datum targets - this is something we both did not even consider yet, but it could be interesting.

So, what will we suggest to OP?

"For every expert there is an equal and opposite expert"
Arthur C. Clarke Profiles of the future

 

[greenimi]

Re:” So, what will we suggest to OP?”
CH,

I don’t know. If the OP will tell us more about how this product functions and also tell us more how this product is made (datums / datum targets used by their foundry/ casting guys, parting lines, fillet radii, gate requirements, draft angles, die closure, flash, etc.) then probably we will be able to compromise and find an acceptable solution.
There are probably 2-3 good ways to define a product. Not 50 good ways. But there are 200 ways to wrongfully define the same product.

 

[2JL]

Thanks guys for this interesting discussion. I pick up a lot of good lessons for my future drawings.

For this particular case, the bore and the flat bottom face are the functionally relevant features. The threaded holes are for mounting an accesory part for which the orientation is not very important relative to the other features. The as-cast taper holes are big enough to provide sufficient clearance for mounting bolts.

3 months ago I would have done it as per Tec. Ease document but now we are exploring different avenues to make our drawings the most process-agnostic as possible. The goal is to shift the focus to the final part functional requirement and let the process expertise to the suppliers. But we are not just shipping them drawings and expect good parts. Before production is started, we have a series of meeting during which we explained the design to the selected supplier(s) and hightligh critical safety and functional features. Then we go through each step they are going to take to manufacture the parts and if needed, make some change to the final design before starting production.
2JL

 

[greenimi]

Re:" the bore and the flat bottom face are the functionally relevant features."

So, in my opinion, I will go with the option I have proposed.


Re:
CH,
"I am a bit uneasy with your way of using perpendicularity, because it still orientation control. I said it before, just because it is shown in the center, doesn't always say HOW central it is. So, concentricity, position, runout - something."

I am not sure why, ON THE MACHINING drawing, you have to locate the center bore? Why you cannot just orient it?
again, why locate, in your opinion, is a must (concentricity, position runout - something)
Why we cannot live only with orientation control and locate everything else from it?