moulds for a steel frame 1

[beautifulhills]

We need to have a mould for a frame (8' x 6' x 35'), with equivalent W6x 12, but to pour into the mould. This will be a repetitive job, making many of the same frame, we expect that to remove the mould it will (will have to be) thrown away. Clearly the 8 corners will be the most complex, the meeting of 2 beams and one column. Has anyone heard of such a task before, we re figuring what the optimal material to make the mould from is, we re thinking this is a feasible project. Any insights or experience appreciated.

 

[DAVIDSTECKER]

Tooling wise you only need 2 corner molds and 1 mold for each of your 3 long runs.
The mold would most likely be a collection of "Cores" that would be assembled like building blocks.
Your problems are going to first get molten metal to flow such long distances in such a slender cross sections.
As the metal cools it may not fill the voids.
Secondly the shrink is going to cause stress if the casting can not contract while it cools down to room temperature.
For example steel has a shrink rate typically of 1/4" per every foot of length.
So your 35' run is going to want to shrink by around 8-3/4".
What material are you going to use?

Another option may be to use the "Lost Foam" method of casting.
There you would make the part in foam that is easily made per casting and then it is replaced in the sand as the molten metal is poured in.
Most engine blocks are made this way.

 

[beautifulhills]

Hi David, Thanks much for the informative response. Attached is one frame that we must make from molds. It's structural for construction. ( Some of the cross pieces could be added later, I m sure you get the idea). As you say there are corner molds. There could be a mold for the seismic bracing also, a fancy pattern essentially. I ve descended to the casting and molding books on Amazon also - not sure what they're going to do for me. I m studying the sand molding - the voids appear to be sand, and then blocks are added in what parts need to be attached into the finished product, there's a great example of BMW engine blocks. Could you explain more on the foam, this is a completely new subject for me and I m struggling to grasp how where the foam will be? Foam is the void that is filled with steel, maybe.

As to shrinkage, I m guessing the mold will have to be in pieces, so each mold shrinks with the shrinking steel and somehow this pull is built into the permanent mold so that it doesn't 'snap'. (As they say where there's a will, there's a way!)

Any info appreciated

 

[bobjustbob]

There is no way I would try to cast that. That is a fabricated structure for a reason.

To get a good mold to fill you would need to break it up into small sections. With the thin sections you will need lots of heads and chills to get good solidification. The yield is going to be terrible. I would guess 70% of the metal poured into the mold would be heads and gates if you were lucky.

Heat treating is going to bend and warp the sections significantly so you will need a method to straighten them afterward. Also there will be weld repair. With the long flat area there will be lots of weld repair. Or lots of dirt traps.

You will spend more time piecing the parts together and fixing indications than simply fabricating the structure. I doubt any foundry would take this on.

 

[arunmrao]

I agree with bobjustbob's comments. It is not a cast component by any stretch of imagination. Fabrication is the easiest way ahead.

The casting will suffer from hot tears ,warpage, misruns, etc. In fact it will be a classic course material for "Defects in Castings".

AFS runs a very popular course on "Defects in Castings".

"Even,if you are a minority of one, truth is the truth."

Mahatma Gandhi.

 

[DAVIDSTECKER]

WOW!!!
That's way more complicated than I thought.
As cast steel part it would be very difficult to get the molten steel to flow completely through the mold to form the thin members.
A 1/2" wall minimum is what I try to design and even that is for much shorter runs.
The American Foundrymen's Society publishes a book titled "Patternmakers Manual" which is a very good basic "Old School" introduction to pattern making.
The "lost Foam" method is were you would create the part in Styrofoam including the "Gating and Risering" requires to pour the steel into the mold.
Yamaha and others use a loose, dry sand that the mold is covered in.
Lots of You Tube videos out there.

I had a similar machine frame years ago at another company with many Structural steel shapes that were drilled on many sides and lots of welding.
I redesigned pretty much the whole thing based on Laser cutting plate and press braking the parts.
I was amazed at the cost savings that could have been realized. That's another story.
The laser I was working with could hold a 0.0005" tolerance in 1/4" steel.
Because I could put all of my bolt holes and slots into the flat plate up front I reduced the number of steps and set ups it took to machine features in 3 sides of a piece of Channel for example.
Another huge advantage is that you can design the parts to be self fixturing, Tabs and Slots, I believe the Japanese have term for it.
I recently did just that for a set of 4 roller fairleads for a ship and it worked great even with using our water jet.
Bottom line; I would encourage you to take a different approach to your frame, I think that the casting will "Eat your lunch" both in effort and cost.
Best regards, David

 

[DAVIDSTECKER]

Here is a photo of the 4 roller frame.
I'll add some more photos when I can.
The original design was C-channel but I water jetted the holes and slots, even the wrap around slot in the flat plate then had the C shape formed.
You can see the tack welded parts fit right were I wanted them and the welders didn't have to do any layout to put the cross members and gussets in place.
Huge time saver and reduces possible "F" ups and re-work.
You have to remember that you're not building a Rolex so let some dimensions float when you can and also use consistent reference points when you dimension.
Oh, one more trick, make your tabs just a little shorter than the material thickness, that way you can plug weld the slot and just grind flat if you what a smooth look.
Best regards, David

 

[beautifulhills]

David (and Bob)
Thanks much. Yes, this is a frame with no sensitivity to smoothness of finish - it will be covered by sheathing and siding Please see attached - a simplified design - some of the prior beams particularly the thin ones were/are temporary to provide rigidity only for shipping of the frame, and are removed on site. The frame requirements are - up to 22' 6" long. We could, for example, - make 2 11' ers alternately. One of the big things is that this is repetitive for us, which we repeat many many times, thus providing a bigger 'budget' for designing it a mold up front, as the cost will be amortized over many production runs. Surely if it was just one frame, fabricate erect and have done with it yes that is right. The intent would be to put the frame mold up on it's side, and introduce points for steel pour so as to minimize the possibility of voids. We will add the lateral braces later as a 'repeatable kit', in my proposal.
David, I appreciate much the detail you have given on laser cuts, (and your answers in general) and the photo - it looks so well made.


(Arun Rao, I m so sorry you've had so much trouble with your castings - I see your projects that didn't work in Defects in Castings, that must be depressing. When you're going through hell, just keep on going)

 

[arunmrao]

@beautifulhills

You are welcome to hold your opinion about my castings. I have been producing castings for over 40 years and am active member in AFS.

"Even,if you are a minority of one, truth is the truth."

Mahatma Gandhi.

 

[btrueblood]

Nobody casts I beams and shapes (since the advent of steel, you might be able to pour heavy section short I beams in grey iron, but it's not the 1800s anymore), it's just not cost effective to do so. You can't form the long thin sections from a single pour point -the liquid metal will solidify and choke the longer runners before reaching all the gates, and your method of gang pouring a line of gates to fill a single mold is a non starter. You won't like the results of the non-cracked/non-voided parts anyway (warping, distortion). Presumably this frame will eventually have other bits attached to it, and those bits have an expectation of a certain degree of flatness and other form conditions at the mounting interface, and the only way you can provide that with a casting is to machine it, which just added more costs.

Brake formed and tab welded (robotically?) are good suggestions from David, if somebody (like an engineer) designs the frame and its bits to minimize the cost while maintaining the essential requirements that the frame was to provide.

 

[DAVIDSTECKER]

If you really want to incorporate castings then one option would be to just do the corners similar to the Steamer Trunk corners. See photo.
You can put reinforcement gussets into the design and make them so that it can accept straight cut beams to simplify the connections.
You could even cast in bolt holes, just be generous with the hole size.
If you can keep it simple then the pattern can also be rather simple, maybe 1 positive on a plate and 1 core box, maybe a match plate if it's not too large.
Then either green sand or No-Bake molding process.
Another option would ISO shipping container corners, a company I deal with TandemLoc has them.
If you are using the frame for shipping then they might be perfect for you especially if you can match the frame size to standard shipping container requirements.
The base unit of size for a container is 20'.
Best regards, David

 

[beautifulhills]

True blood, David, thank you.
Attached, roughest of sketches. The corners David somehow like your stream trunk corners. As you say with bolt holes, to get the bolts in, once the HSS posts, and horizontal I beams were in. The insert for the I beam needs some thought beyond my sketch. The simpson post base is really nothing like it (for the sketch, turned upside down) but one wonders how simpson make those? But part of the corner for the HSS posts would look something like that - of course it would be one piece with I beams - so three insert points. (As somebody pointed out, yes one the ideas have completed an engineer needs to design it). Then there's the lateral brace - maybe that could be molded. Thoughts appreciated

 

[beautifulhills]

Simpson post base

 

[SwinnyGG]

But one wonders how simpson make those?

No mystery. Those post bases are stamped from a single piece of steel sheet, die-bent into shape, with a single corner welded to close the section. In the photo you linked, the corner on the right side of the label is the welded corner.

I think the posters urging you to consider fabricating this design as-drawn in your first post are correct.

All the castings that you're designing do is add cost. That large casting you've drawn in your last post will be heavier, more expensive, and no easier to assemble than a bolted joint would be.