Hot Metal Processing Log

From Open Source Ecology
Jump to: navigation, search

Thu Jul 2, 2020



Maybe you can help us get to the casting part of steel using re-usable molds.

Pouring 200 lb/hour of stainless steel shaft makes for a robust business case on a small scale. I'd like to look at practical experiments for getting there.

Can you fill me in on the best technique that you can suggest - either to cast steel, or soft work steel - to rod form? We're interested in rod from 8 mm to 3" diameter.

I'm thinking, an easy but powerful experiment for small scale rod production. Casting or other hot metal work sounds good to me. You mentioned hot working material into shape. But I'd like to be starting from scrap steel, adding ferrochrome to it as needed to make stainless steel. The OSE strategy is to shift from rusting steels to non-rusting steels as general purpose steels, for the sake of lifetime design. When we make stainless, the value of the product goes from $5k/day value to triple of that.

Thus - what would you suggest for the simplest reusable cast experiment for making billet or bloom that can be processed to precision shaft from there?

The workflow I see is:

1. hot metal bloom or cast 2. rolling for good dimensional properties - for structural steel 3. CNC machining for finished precision products




In industry, as you probably already know, large forge presses are used to hot or cold work stainless steel to pound it into rough shape and obtain increased integrity and required grain structure. These finished billets are then used in secondary processes such as machining. If high tight tolerance diameters are needed, as the case of the 8mm 3D printer rids, centerless grinding machines are used to grind the rods to final diameter.

All of these processes are run by people who rarely question if there is a better less energy intensive way to process these metals.

Here is the thixotropic state process for casting that I think is under utilized:

However if you are going to make your own stainless steel, you will have to melt scrap steel to a liquid state. In doing so you will want to take that liquid stainless steel and processes it right away into the cooled shapes you want. Letting it cool and the re-heating it will cost a lot of energy. So having your process steps from liquid stainless steel to the final part right next to each other would be ideal. This is what lean manufacturing principles would dictate. Lean manufacturing is another under-utilized set of techniques. Of course lean manufacturing can use some hacking to improve it as well. -- Daniel N. Meyer Fab Lab Senior Manager



I'm thinking for the 10k sf and under facility, it could look like 3" billet casting. These could thus be 8 feet long at 3" rounds. This meets need for heavy shafts, and can be processed to any other such as 1"x6 plate rather easily.

So what do you think: start with 200lb melts with a 200kW induction furnace for at least one heat per hour. For the cast billets - can we cast 3" rounds? How to do that?

Then we take these into metal rolling. If you say don't let them cool - we can take these out of the molds and process them right away? How to do that?

What I would like to see is a CNC adjustable, reciprocating (moves back and forth between rolls, rolls are adjustable in 2 dimensions. So we take the billet, make it into plate such as 1/4", or up to 1" or even 2". The rolls have infinite flexibility, so we can get out any cross section that we want. In this process, there could be local induction reheating - so we could have a flexible or adjustable array of induction coils that the billet would go inside of.

With this, we have the ability to make any steel profile. 

The 3" can be a ready feedstock for a screw machine. So it's both a feedstock for rolling and milling, and is nearly what we need for shafts. We can play with the metallurgy, maybe we can develop a general purpose stainless that is easy to work. I'd like to make the tractors and all else out of this for lifetime design.

Any insights on how to get to the above or what would be another route? I'm after toolchain degeneracy here - reducing material choice to a smaller multipurpose set, allowing for lifetime design, allowing for structural and precision steel.