Redundant datum /reference dimension as datum

[R_M_]

HI,

I am new to this forum also a beginner when it comes to geometric tolerancing. I've read quite a lot of post on this thread found them very helpful.
I have a case where I need to use a cast feature which is FOS and use the derived median plane as a datum. For functional reasons, the other features must be symmetrical to that cast feature. Attached is a simplified model of the part in question.
I do not want to tell the machinist how to machine the part. I only care that the seal groove and the face to face of the parts are symmetrical to the center slot. My questions are :
1. Since the slot is as cast, therefore in the machining drawing the size is specified as a reference dimension (16)with Datum feature symbol added (A). Regardless of the size, the seal pocket (30 0/0,4) must be in symmetry with the center plane of the slot. Is this legal?

2. I added an Orientation tolerance for one of the faces and use that face as Datum B and the other face referenced to Datum B. I marked in red the DRF where datum B is referenced. Is datum B in this case redundant? Could I just use datum A instead?

3. For the 2x holes ø185, I specified that they are considered as a combined feature, so I added CZ (Common/Combined Zone. Would it be more practical that Datum A is used instead of Datum B? I haven't seen any examples/drawings where the datum intersects 2 separate features that are combined or with common zone requirement.

4. Is datum E necessary? Could Datum D be used instead to orientate the M8 threads? A plate will be connected to that surface. It does not necessarily need to be tightly orientated to Datum D. As long as the surface is flat enough I am happy.

This feels like it's quite a lot to ask but the books do not go into such details and won't learn new things if I don't ask.
By the way, I am using ISO GPS/ISO 1101.

Regards,
Richie

 

[drawoh]

R_M_,

I do not like inaccurate (i.e. as-cast) FOS datums. Either you need a fancy fixture that picks up the inaccurate feature, or you need to specify all your features at the datum feature MMC/B, and tolerate the resulting sloppy tolerances.

Is your datum[ ]feature really cast in place? Can a foundry do that?

I have not worked a lot with castings. If I do one, I want the foundry and the machine shop to work off the same datum features. I would apply datum target features on your casting. You do not need symmetry about your datum, you want it around your 16mm slot. A slot is not a good primary datum anyway.

Your datum[ ]B feature is parallel to your primary datum slot. It is not much use.

Go back and review your datums. Your datums must constrain your part.

I observe that you are using first angle projection. Are you working to the ASME[ ]Y14.5 standard, or to ISO?

--
JHG

 

[R_M_]

Hi drawoh,

I myself have not done a lot of castings. Frankly, it is a new area for me.
I not sure if I used as-cast in the wrong context, but what I meant is that the 16 mm slot will not be machined. It would be very hard to machine the slot and for sure would be very expensive.

I've been told by some very experienced guys that it is good enough to dimension from the centerline. But I've realized after the taking some Geometric tolerancing courses that the centerline means nothing. We had a prototype made for a similar part and right from the start the fitters were already at my desk telling me that the part is not according to drawing. The drawing did not have any geometric tolerances, therefore the machinist and the inspector could choose any features as they please to come up with the imaginary centerline.

The sketch is for a gate valve body (slide). The Gate/knife will travel in the 16 mm slot. As I learned, datum selection should be based on functionality and not the manufacturing process, that is why I choose that to be the primary.

I am not familiar with using datum targets but that is something I will have to look into.

I work for a European company, so yes I am working to the ISO standard.

Thanks for the feedback.
Richie

 

[drawoh]

R_M_,

Dimensioning from centre lines is valid according to the standards, but centre lines come from FOS datums, which do not work well on castings. Many foundries have online design guides, which recommend drafting practise, among other things. I suggest you look them up.

--
JHG

 

[jassco]

Hi, Richie:

For cast parts, cold headed or forged parts, you need two sets of datum. It is not a good idea to use datum A twice. You need to think more about functions of your parts and initial setup of the part on your fixture. If you can post an image, it will help us better help you.

Generally, you want to use datum ZA, ZB, and ZC (datum targets) to immobilize your cast part. You machine a your datum B, C and another clocking feature. And then position the rest of features to the datum B, C, and the clocking feature.

Best regards,

Alex

 

[R_M_]

Hi Alex,

Thank you for the feedback. I will upload in the near future an updated drawing of the one I posted earlier.
What kind of image are we talking about, an image of the part?

Regards,
Richie

 

[drawoh]

jassco,

Do you really want two sets of datums? If I define datum targets on a casting, the foundry and the machine shop are looking at the exact same feature on the exact same part. I see no need to machine in datum features, unless I need an FOS[ ]datum.

--
JHG