The Shop > Metal Stuff

Casting a Model Westinghouse-Type Twin Steam Engine in Iron

<< < (11/11)

Good job! So.....did the plan B work or is this allready plan B23? :D

I always though that brake disks are worst cast iron to work with, but what do I know, I never cast cast iron, very little aluminium and other low meltpoint stuff.

Thanks Pekka!  :beer:

When it comes to iron, I'm probably at Planet B -- a world of unknowns! With aluminum or zinc alloys. I know what I'm doing and what to expect. Brass, eh, a few more wrinkles. But iron! It is a different world.

You don't worry about chill hardness with aluminum. Casting thin sections? No problem. Facing sand? Just finer than the backing sand -- no need for reactive additives to control sand fusion. Time to melt? About as long as it takes me to ram up a mold, and I often do while the aluminum is melting, because the furnace is a quiet thing gently heating instead of a roaring oil smoking jet engine, intent on destroying refractory linings.

The noise level alone in an iron melt has the adrenaline rushing, and meanwhile you're covered in leather no matter how hot the day and proximity to your homemade inferno. And the radiated heat and light are intense, actually dangerous. Cooling time? Even a small iron mold takes 4 -6 hours vs 1 for aluminum, and even then the iron and sand are still hot. Wooden flasks? Short life with iron -- decades with aluminum.

And the results? For me always iffy, because I'm constantly changing variables -- often by necessity. Iron is not easy to come by, and varies greatly in qualities. Disk rotor iron is one fairly available source for that, so that is why it is of interest.

It does have good qualities for use: high strength, high temperature resistance, high wear resistance (hardness) -- all to be expected in its original purpose. I think it has reasonably low shrinkage. It seems clear to me now that it almost certainly is annealed/stress relieved by kilning after casting. And I'm sure there are very specific heats and times for whatever particular alloy a brake mfr is using.

For the home caster, however, yes it is a difficult metal to cast with compared to older traditional massive forms of gray  iron. Rotors have lower carbon content, and a higher melting point, by, as a guess about 3-400F (~200C). That nearly doubles melt time, especially noticeable if cold iron is added during the melt to fill a crucible. My furnace seems to be operating with a relatively low margin of temperature above rotor iron's pouring point, so heating is not linear over time. The last few degrees take the longest.

Not so with aluminum or zinc alloys in even a charcoal furnace -- the temperature difference of the furnace interior and the melt is high. And even with non-rotor grades of cast iron, melting in my oil furnace can begin in as little as ten minutes.

I do think that rotors are a good source of iron, but that additions of other lower melting point iron will probably help the home caster. Radiator iron is a fast melting type, and very fluid, but produces a fair amount of difficult slag, and also has a high shrinkage rate on cooling. For me, I would like to find a good mix of the two, since I have both on hand. How much of each, I don't know. Ideally less of the radiator iron, and more of the rotor iron, because the radiator iron is in finite supply.

I'm guessing that a larger than necessary crucible would help with speeding up melting rotor iron, so that you can avoid adding metal after the initial full charge. Or, in a smaller crucible, using mixed metals, adding radiator metal to a molten rotor metal charge, rather than rotor metal to molten radiator metal charge. But these things are to be worked out in practice.

Same thing for annealing to get machinable metal from rotors. Or, and I haven't reached that point, learning just how much ferrosilicon will correct it. So far, it's been too little.

Pouring, slagging, riser and gating methods are critical to getting impurity free castings. Again, slag quantity and type varies according to a lot of factors, including metallurgy, scrap cleanliness, furnace flame mixture control, time to melt, etc.

So yes, large number of variables, I guess I'll probably always be at least plan B, or C or Q with iron. But it keeps you coming back to try to figure out what you did wrong and could do better.

As a follow-up here is the rectangular bar cut open -- nice clean metal inside. The outside was rough because it was simply poured into a rammed vertical open topped single-flask mold -- like a sprue. No facing sand and no parting line.

Shrinkage measures a total of 1/32" across both wide sides. The thickest section at the corners is 1-1/16". the thinnest near the center is 1-1/32". That's roughly 3% shrinkage due to differential cooling. The metal was roughly 1/3 rotor iron and 2/3 radiator iron. Remelted from a failed casting and sprue and riser.

The original pattern (a steel bar) was a true 1" across by 2" wide. The casting is thicker and wider than the steel bar because rapping makes the mold cavity larger.

If the finish is better it is sometimes possible to machine back to true size using this molding method..

That is pretty small shrinkage. Last bar looks good industrial quality. It's really nice to read your stories, there is a lot of experiment and though there.


[0] Message Index

[*] Previous page

Go to full version