large size casting parts drawings, ISO vs ASME

[bxbzq]

I'm working with casting part drawings that are ISO based(ISO 8062). I think the designers use similar to "END ITEM DRAWING METHOD" as explained in para 2.4 of 09' Y14.8, to draft these drawings. What I don't quite understand is there is no datum specified on the casting drawings. None of the casting part drawings has datum specified. These casting parts are typically cylindrical and huge, around 20 tons. Imagine pressure vessel. So I figure there are couple of constraints to apply GD&T for large size casting parts:

1. Features supposed to establish datums are too rough to be datums features. These surfaces will be machined later in machining process, and may be specified as datums on machining drawings.
2. It is not practical to use datum simulators or gauges because of the size.
3. It is not practical to use CMM to inspect the part because of the size. Even for machined parts, I was told inspector uses ruler to measure most of dimensions.

However, we also have ASME based drawings for same parts out of casting. These drawings use GD&T. So I'm not sure if it is the constraints listed (may be more) cause different approaches between ASME and ISO, or I miss something. Anyone works with large size casting parts can share some ideas?

 

[MechNorth]

How large is the casting, dimensionally? 20T doesn't mean much.
As to your specific questions,
1) datum targets are commonly used on castings of all sizes
2) using datum targets, a somewhat simple datum simulator fixture can usually be setup, even for large parts; I've concepted some that were over 25 feet long by 10 feet wide by 6 feet high
3) I've seen very large gantry-CMMs, as well as long-arm CMMs that can be moved around to accommodate large pieces

Most likely the designers of the castings just aren't adequately familiar with GD&T, and / or don't see the value in it for a casting. Alternately, the geometries may be considered "simple" enough that they don't see the value of the effort.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[bxbzq]

Jim,
Take one as an example. It has OD up to around 4.6m, ID around 4m, more than 3m long. There are a lot of holes, flanges, struts, ribs on it need to be located and oriented. Not a simple part at all.

 

[fsincox]

The theory is the same and it should work to describe any part geometry. I still want things coaxial and such, it is just the numbers must be realistic for the size and processes.
Frank

 

[MechNorth]

The size of your workpiece isn't that bad. There are gantry CMMs plenty large enough to fit your workpiece (a previous employer had one that would handle your workpiece nicely), and by using the mobility of mobile-arm-type CMMs, you should be able to do it. Fixturing your datum simulators is still a practical option rather than probing the surfaces directly; it just takes some design creativity.

Jim Sykes, P.Eng, GDTP-S
Profile Services

http://www.profileservices.ca

TecEase, Inc.

http://www.tec-ease.com

 

[SDETERS]

How are you holding the part to machine it? How are the machine features held back to the Raw Material Casting? Are you machining away what would have been your cast Datums?

 

[fsincox]

HHave you done many castings using ISO 8062 in the past? Does it work well for you?
Frank

 

[bxbzq]

I'm new in this company. From the drawings, I see GD&T is not well understood or implemented here. I see many features located by coordinate dimensions; I see datum symbol placed on a centerline-like line which turns out to be the center plane of a cylinder; I see many concentricity callouts on stationary parts. Well, I better take an ISO based GD&T training before making judgment.

SDETERS has brought up a good point. I probably should see the machining process and understand how the part is being set up.

 

[fsincox]

The ISO system allowed for datums on centerlines in the past, They do discourage it now, for the same reasons ASME does. The concentricity symbol is coaxiality and it is just a special subset of the position tolerances, like perpendicular or parallelism is to angularity in the orientation tolerances, they do use material modifiers with it.
Frank

 

[fsincox]

What class of tolerance grade do they use for this casting, do you get the impression from others around you it works well for them? I have been very interested in the application of this standard.
Frank

 

[bxbzq]

General casting tolerance CT14. Our casting supplier is ok with this grade.

 

[Guest0527211403]

We make aluminum castings (not as large as yours) and we don't specify casting datums. We do provide drawings for the parts which show major dimensions and processing requirements, but the foundry uses our STEP data to make the patterns. On our drawing we specify things like pattern mismatch and finish.

Our machining drawings establish datums from cast surfaces (i.e. parting surface), but we use regular (+/- toleranced) dimensions to locate our primary machining datums from the casting, and then geometric tolerances for all machined features based on our primary machined datums.

It hasn't caused an issue with our suppliers (yet). We initially had some tolerancing schemes that involved cast-in pads for datum targets but that was thrown out the window when our machine shop told us they were not needed.

fsincox - we use CT12 without issues from the foundry. Part sizes range from "slender" (~200mm x 200mm x 1700 mm) to "chunky and flat-ish" (250 x 500 x 850). Material used is AA356-T6.

 

[fsincox]

Thank you, both
justkeepgivener,
When you say just plain dimensions, I assume you are working in metric? Meaning you are not dealing with implied decimal place tolerances just the "CT" implied tolerances.
Frank