The Shop > Metal Stuff
Oil fired crucible furnace
vtsteam:
Sorry Pat, not interested in anything but traditional greensand. If it takes years to master, so be it.
Sand type doesn't cause slag inclusions. Pouring slag does.
Gating can catch it. But better not to pour it in the first place.
vtsteam:
--- Quote from: tekfab on July 24, 2013, 05:36:18 PM ---Just out of curiosity have you thought about casting it the other way up ?
Mike
--- End quote ---
Mike, both surfaces need to be machined to mating surfaces, so it would just transfer the problem.
It could be cast vertically as awemawson once mentioned, and the upper end trimmed -- allowing a generous machining allowance and assuming slag inclusions weren't the problem.
But a new pattern would have to be made and draft would have to taper along the length which would be pretty wasteful of material. I guess one face could be parallel to the true datum, and that one would have to be the channeled face.
I guess it's do-able as a last resort. But the other mating piece is a more complicated shape, and has lettering, so it must be cast horizontally. I do have one of those already cast -- the first piece I ever did in iron -- beginner's luck on that one.
But also, I need to learn what is wrong here, and how to handle it. So even though it's frustrating to spend so much time on this simple piece, I want to learn how to cast it and have it work.
awemawson:
Steve
I expect that like me, the journey is as important as the end result. Most of my workshop activity involves finding new skills and learning new things. The end product in my case is really a byproduct!
Andrew
vtsteam:
That's right Andrew, if I move on now then I'll never understand molding iron in greensand.
I know it CAN be done, the evidence is all around us in a couple centuries worth of superb iron castings. Not to mention ironman's videos.
The radiators I broke up are fairly amazing in themselves. Making one of those would have been a heck of a lot more demanding than this simple little rectangular block of iron I'm trying to figure out. Some of these tube style radiators are 5/8" ID cross section and 1/8" skin thickness, with ribs, and a central parting line. Must have been some real interesting cores there -- and I think I've even found some chaplets in the scrap.
Just getting those cores out is ship-in-a-bottle-stuff. They must have had the mix just perfect for them to fall apart easily, yet strong enough to hold those intricate long narrow shapes, perfectly positioned. No way did they get scrapers inside after the pour. :scratch:
Now that was iron casting. :bow:
vtsteam:
Today's casting:
Decided to try a different cast iron stock -- I have a rail or old automotive frame section that is cast iron -- can't tell which. It was found in an old abandoned auto junkyard long since bulldozed over. Looks like a narrow train rail, but definitely cast iron. I broke off 7 pounds of it -- it's about 1/2" thick stock when broken up. Nice solid chunks compared to the radiator stock. I made it up to 8 pounds with one of the short pour castings I had from earlier attempts. Figured that would add a little phosphorus to the melt, and it was clean.
I also retired the second crucible -- it is just completly crudded up with slag inside, and the upper walls are getting thin. I think I got 5 melts out of that one. Both were purchased 7 years ago and were stored ever since, awaiting construction of an iron melting furnace.
I just received a 3rd crucible from the same source. Same size but looks a little different. Wondering if it would be any better. So I put the new metal in the new crucible. Decided not to use flux this time since I figured there would be much less slag.
Molding sand seemed a little damper than it was last pour -- that time it was on the borderline of unworkably dry. But I don't think I had blow holes-- just slag inclusions.
Anyway, rammed it up, melted the metal -- that went really well this time. Nice full pot, not much slag, and it was removable. Did a careful job of that. The metal was nice and hot. I poured the mold perfectly, though there might have been the slightest pause at one point where the sprue wasn't choked.
I thought this would really be a good one because everything went so well. But no there was a really big blow hole -- biggest of any of the pours by far. Maybe it was the very slight pause in pouring. I'm thinking the sand isn't permeable enough, though.
So trying to think what could be the matter with the sand. It is brand new U.S. Silica F60 (foundry grade 60 mesh) sand, bonded with ceramics grade bentonite @ 7%. Water was originally added at 4%, but after molding it has been adjusted by feel.
Previously for aluminum I used 100 mesh sandblasting sand with fire clay @ 15%. I used that sand for 11 years, and was very used to the feel of it when properly tempered.
I am not used to the feel of the new iron sand -- it's quite different. I tend to judge the moisture content by how well mold edges hold up when cutting gates and lifting the pattern. Last pour before this one when the sand was very dry, the edges tended to crumble when cutting gates. It was what I would call borderline unworkable. But it did seem that permeability was adequate finally and the blowholes weren't a problem (slag was).
What has occurred to me this evening is that maybe I don't have enough bentonite in the mix. Maybe I had a little more clay in the mix the corners wouldn't crumble with even less water -- in other words, more green strength with less moisture and I hope, more permeability as a result (also less steam released). It's possible that I am compensating for too little clay binder by adding more water than is necessary.
This is all probably because I don't have any prior experience with bentonite. I don't know its proper feel. Also there are more than one type of bentonite -- so an exact recipe like ironman's might not fit my particular sand and greensand mix. I might need 8% instead of 7%.
So that is what I might try next. A little more clay, and a further reduction in moisture if possible.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version