machined casting drawing - datum targets

[AndrewTT]

ASME Y14.8 - 2009 section 4.4 DATUM REFERENCE FRAME ESTABLISHED BY MACHINE DATUM FEATURES

Machined datum features should be established from a cast/forged/molded part datum reference frame for establishment of subsequent datum reference frames. The initial datum targets shall be shown and described on the drawing showing the machined features. See Fig. 4-9.

----------
OK, so here is the easiest question in the world. Is there any way that I can use a note saying "DATUM SCHEME D (PRIMARY), E (SECONDARY), AND F (TERTIARY) SHOWN ON CASTING PRINT" in place of showing the casting targets (D, E, F) on the machined drawing? I know the standard just said that I have to re-show the casting targets on the machined print but my company throws a fit when we put the same information on two different drawings. The fear is that an engineering change goes through to change one print and the other does not get updated and now there is a discrepancy. A note stating that the casting datum scheme can be viewed on the casting print eliminates this possibility.

So, I have a machined casting where I want to use a position callout, on a machined FOS, with respect to the casting datum scheme D,E,F. This machined FOS is positioned to the casting and becomes datum feature A. The rest of the machined features are described with respect to datum A and subsequent machined datum features. Why is a note saying - if you want to view the specifics of the casting datum scheme you should look on the casting drawing - not allowed? Is it because of Fundamental Rule 1.4 (o) from ASME Y14.5 2009?

(o) Dimensions and tolerances apply only at the drawing level where they are specified. A dimension specified for a given feature on one level of drawing (e.g., a detail drawing) is not mandatory for that feature at any other level (e.g., an assembly drawing)

This rule does not mention datum schemes only dimensions and tolerances.

Thanks!!

 

[KENAT]

Simplistically it's because "ASME Y14.8 - 2009 section 4.4 DATUM REFERENCE FRAME ESTABLISHED BY MACHINE DATUM FEATURES" says so. It's not even a 'should' it's a 'shall'.

To generate your machined datum's, the new datum features need to be dimensioned or located from the casting datums.

I'm a bit unclear how you achieve this for all DOF's if the casting datum's aren't identified on the machining drwing.

If the datum structure of the casting is changed, then surely the dimensions/locations of the subsequent machined datums would need to be updated to reflect the change anyway so forcing the machining drawing to be updated.

Posting guidelines faq731-376

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

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

 

[AndrewTT]

I suppose that I could add a note to each drawing stating that the same casting datum scheme is also shown on the other drawing. This should help to avoid one drawing being modified without the other being correspondingly updated.

Datum D on the casting (primary datum) is taken from a wrench boss (FOS). Specifically it is taken from the size dimension across the wrench flats. On the machined drawing do I need to call out the casting wrench flat size dimension again or do I only need to show the datum coming from the distance across the wrench flats and leave the dimension blank?

Thanks!

 

[JNieman]

KENAT covers the rules of it.

To cover the 'practical purpose' of it, I would suggest fully defining the datum points on the machined print as well. You may not have the same supplier doing the casting as is doing the machining. You may be receiving completed, fully machined castings from your casting supplier, but they are probably subbing out the machining. Or you sub it to a machine shop and they sub the casting to their casting supplier of choice. Regardless, it'll be very easy for the supplier doing the machining to not have the casting print.

Additionally, it's very important that the same datum points are being used by the casting supplier and the machine shop. If the casting supplier uses datum points to inspect their part and verify that it is good, but the machine shop uses datum points slightly different, they could very easily make bad parts. That's one of the biggest reasons datum points are being used for castings when-cast and when-machined - so they are always taken back to identical datum reference frames. So make sure the datum points are dimensioned with basic dimensions or included within a supplied CAD model to ensure consistency.

I've seen prints with datum points called out, no CAD model supplied, no basic dimensions, and the customer simply requests we use points "close to as-shown on print" which means they'll, in theory, never be able to exactly duplicate our inspection.

 

[KENAT]

"On the machined drawing do I need to call out the casting wrench flat size dimension again or do I only need to show the datum coming from the distance across the wrench flats and leave the dimension blank?"

I think what I did in similar situations was add the size dimension but as a reference dim then show the datum as I would normally. Think about how this may impact application maximum material condition/boundary to any subsequent FCFs though.

Posting guidelines faq731-376

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

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

 

[AndrewTT]

Datum feature A (primary) on the machined part references datum D (wrench flats) at RMB so I think showing the dimension as references will work just fine. Good point about how a reference dimension would affect the material boundary though. If my datum feature A had referenced datum D at MMB I believe that I would just callout the desired MMB value in brackets in the FCF per 4.11.6.3 (2009). I believe this would solve the issue you were concerned with.

 

[drawoh]

AndrewTT,

It is possible and legal to specify different datums on the machining drawing than on the casting drawing. It is also a dumb idea. Casting design manuals recommend casting in datum target features, for this very reason. Castings usually require thousands of dollars worth of tooling, making most organization reluctant to modify them. There is nothing to stop you from creating a new machining drawing, different from the original machining drawing, defining a new part that meets new requirements, but using the same casting.

[ol]
[li]Using the same datums for casting and for machining is good practice.[/li]
[li]The machined parts are different from the castings. They need their own document package. Machinists should not be referring to the casting drawing for datums.[/li]
[li]There seems to be a fear that someone will change the casting. How likely is this?[/li]
[/ol]

--
JHG

 

[AndrewTT]

"1.Using the same datums for casting and for machining is good practice.
2.The machined parts are different from the castings. They need their own document package. Machinists should not be referring to the casting drawing for datums.
3.There seems to be a fear that someone will change the casting. How likely is this?"
-------------------------
1. I agree with this. What I was talking about was using the datum scheme of the casting to locate the first machining feature and then creating a new datum scheme from that machining feature to control the rest of the machining features. I believe this is also common, accepted practice.

2. I accept this answer also.

3. The fear is a general fear that we have conflicting information. The door is open for this when you put the same information in multiple places. In reality casting drawings get changed very rarely here. If we are going to say our drawing is prepared in accordance with the ASME standard then we need to follow it. We would need to do something in our engineering change process to avoid any discrepancies between drawings. This is on us.

 

[JNieman]

I cannot envision a change to the casting print that would not necessitate an examination of the machined part print, to evaluate fallout/consequences of the changes.

Yes, this means it's completely "on you", the engineering department, to check for it. Would you accept a change to the casting print without evaluating or verifying the machine print, though?

Let's say the datum points on the casting do change. Let's say the machine part is free of direct datum point symbols, and says to reference the casting drawing. Your machine part /has/ changed, though. The DRF for possibly significant features is in a new place, or depends on new features or points. At this point in time, of the casting print being revised, one /must/ evaluate the changes. Is it any more burdensome to update the datum points on that drawing?

Tangential note: Can't you simply create a model of the machined part that is directly and forcefully tied to the casting model, and have the machined part callouts automatically update? This would be possibly using PMI in NX, I believe. That may get you the "best of both worlds" where you eliminate the possibility of conflicting information but also have the benefit of creating separate, stand-alone drawing packages for the casting and machined part.

 

[drawoh]

AndrewTT,

Let's take the case that I come to work for you and I propose to change your castings.

[ol]
[li]You are looking at thousands of dollars in tooling changes just for the castings.[/li]
[li]My changes will affect machining, even if I do not change the datums.[/li]
[li]If your machinist relies on the datum specification on the casting drawing and I change the datums, the machining process is affected by that alone. [/li]
[li]The changes affect your inspection tooling, whether or not I changed the datums.[/li]
[li]The changes affect your inventory and the functionality of your product, whether or not I change the datums.[/li]
[/ol]

Modifying castings is not trivial. Modifying any part in production is not trivial and involves all sorts of process including design change rules.

Datums should be called up on the machining drawings.

--
JHG

 

[AndrewTT]

I don't think the note that I originally suggested is any simpler or better than showing the casting datum targets in the machining drawing. I just know the knee jerk backlash I will get here when I say we have to do it...

If you machine 10 different versions from the same casting and one of those has a very low volume it can be forgotten about when you talk about changing things. Again this is more of a process issue that should be resolved internally.

Currently we have the machined model as an inheritance of the casting model. So yes, if the casting model gets changed the machined model will update automatically. However, our 2D drawings are created in a different software package. So a draftsman could follow the direction he was given to update a drawing and not realize all impacts of those changes. The engineering change committee should investigate that....

 

[JNieman]

. However, our 2D drawings are created in a different software package.

OUCH! - that changes things in my mind, then. I am a little more understanding of the over-caution of your checker, even if I don't believe it to be entirely logical in application for /this situation/.

 

[KENAT]

The Engineering change process should still catch it though.

With ERP systems and PLM/PDM systems becoming widespread doing a 'where used search' is generally a lot easier to do than it may have been long ago and should probably form part of any robust doc control process.

Additionally, if any change to a part that makes it non interchangeable (including things like need for traceablity of changes that otherwise doesn't directly impact form, fit, function...) results in a new part number that will again help address it.

Posting guidelines faq731-376

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

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

 

[AndrewTT]

Yes, there is much room for some continuous improvement in the way we do things. I think my original question has been answered.

Thank you for input everyone.

AT