Machining a forging

"base part" is one method... putting the forging into an assembly and then making cuts in the assembly is another

In forging part, under Solid Bodies folder, right click on body and select "Insert into New Part".

How about using configurations: one configuration of the forged part with the machined features suppressed, and a configuration of the machined part with all features unsupressed.

Mike

Cliff,

Yea, it does seem backwards, but from a modelling (construction) angle, it's more straight forward. Having the forging as a base allows you to go from simple to complex. The other way (which makes much more sense from an engineering perspective), is more complex to model. It could be done that way in many cases though.

Both methods can be associative. Whether or not changes in the driving parameters result in desired behaviour depends on how well things were defined.

Mark

"Michael Eckstein" a écrit dans le message de news: thEXe.22219$ snipped-for-privacy@fe03.lga...

That's what I usualy do. Never had to regret it.

I disagree--all question of SW methodology aside, using the forging as the base part corresponds to physical reality.

You can hold the forged part in your hand before machining. I've never yet seen anybody weld additional material onto a machined part to create a forging.

Michael wrote:

Products are not designed as collections of forgings. They are designed as a collestion of assembled finished parts. Forgings, and die catsings for that matter, are derived from finished product designs. Why would you reinvent the wheel?

Cliff,

In Solidworks you'd use a function called configurations. The base configuration is the forging, the other would be the machined forging. This is just a simple case. One part could concievably have dozens of variations.

It's just easier, from a parametric "housekeeping" point of view, to model the forging and cut away all of the machining. This is also the most natural workflow for the software. It's also true for Pro-E and other similar systems.

Mark

Cliff,

When you design piece parts, forged or otherwise, in the context of an assembly, it's pretty hard to make those kinds of mistakes. You can also tie key features, between one part and another, with equations so the proportions remain constant. Not very hard to do really.

Mark

Hmmm... that statement seems odd:

Working in UG, one loads the finished part model into a new part file for the toolpath work. Next, one would load the forging model (as an assembly: lives in a separate part file)into this file. The forging model is then "mated" to the part (ie, centered about the part for best fit). Then, fixtures are either created or brought in (assembly) and mated to the forging surfaces for location. As the part nears completion, other fixtures may be brought in (other layers) and mated to some of the machined surfaces of the part.

The point is (IMHO): the finished model should be the root part. If your concerned about inprocess machined features, one can save an inprocess model (created by the toolpath) to be used as stock for subsequent operations. This can be done via UG's own inprocess stock feature or imported from say Vericut.

Designers design parts. Not forgings. They are a result of the finished part.

-- Bill

Just to update: I see this was a crosspost from comp.cad.solidworks (darn Google). I wouldn't have made the reference to UG.

-- Bill

Bill,

Why not ? Nothing wrong with hearing about how things are done in other systems by other people.

Mark

Bill,

Sounds kind of open ended. Does the forging update when changes are made to the base part using this method ??

Don't know if I like the idea of using a mesh model as a production reference.

This is both a yes and a no,, depends. Often, a machined part is very different from a forged/machined version. You usually have alot of rough, drafted features with only the "working" surfaces and features machined. I think it's important to have this accurately represented in the master assembly. I don't do many forgings, but I do castings. The results are more important than the steps or methods.

Regards

Mark

Ah, now I see what I'm missing in this thread. Your talking about the design stage as well as through manufacture. By all means, one could associate the forging model to update with the part model. Might be a bit trickier than the other way around but I've done. The beauty of it is that the finished part model is what's key. One other issue with ordering forgings is they can and will grow or shrink as time goes by. You'd want to be able to account for forging changes as they come from different vendors. I've seen this lots with old forgings for FAA type parts.

I now see what your talking about. Many time the forging is very much the finished part but for some locating surfaces and a few other things. In that case, the forging is so close to the finished part. Why not just model the forging first to use as the basis of the finished part. In other cases the forging is basicly a "blob" shape around the finished part where everything is machined. I'm no designer but in the case of the former, I model the part then, create the forging based on it. Since most of my work has been programming this type, that's where I'm coming from.

Along this topic, it seems that nowadays everyone just goes for hogouts from solid forged square blocks as forging lead times can be long or not available. This creates a real mess as we must sometime provide all of the existing draft angles even though it's no longer from a forging. Very common with gov't contracts for old aircraft spares.

-- Bill

Bill,

You mean a part that was once made from a near-net forging but is now machined 100%, and they want all of the draft angles from the original near-net shape machined in ???

Seems kinda dumb to me, unless there's a structural reason for these features. But then, your dealing with a big bureaucracy. It'd probably cost

Regards

Mark

It's dumber than dumb! Problem is these are usually older parts from the 70's. They are being made for the Dept of Defense as spares. They supply the origional Douglass drawings. There is very little they will do when it comes to changes.

-- Bill

Bill wrote: ... it seems that nowadays everyone just goes for hogouts

I find this amusing since they aren't likely to get decent grain flow that way, which is one major reason to forge in the first place. I guess the it does compact the metal and close up voids, though.

As a forge die designer I would be very happy if some of our customers would take draft into account when designing parts, but then I'd be a lot less busy and have to bring a pillow to work...

Cordially yours, Gerard P. Lakeview Forge Co.

Excellent. An on-topic discussion, and a guy who's brain I need to pick a little.

I do a lot of 3D mechanical design and modeling;

I realize this is like asking "How far is up?" but here goes;

What draft angles do you guys like to see?

I assume it varies with length and width of feature and such.

Radii?

It's an aerospace part. Deviating from the print is the biggest no-no you can imagine. And how much would it cost to re-engineer the print and get FAA aproval for the new part? I wonder how close you got with your guess?

Later,

Charlie