Sciencemadness Discussion Board

High Temperature Argon Filled Arc Furnace

kilowatt - 15-1-2008 at 19:08

I have now been getting into metal casting and have done some aluminum and brass stuff in a #7 crucible furnace which I made last thanksgiving. It is time to take it to the next level now, albeit at a rather small scale.

My latest creation is this small high temperature crucible furnace, which I built mainly for casting titanium and other high melting metals which are reactive in air. It has a 50mL zirconia crucible which can operate up to 2200°C on a rotating jig inside so it can be turned and poured out the bottom into an argon filled mold box below. It operates from my welder and develops about 3500W continuous (50V 70A). Initially I tried to have a graphite crucible holder which doubled as the anode for the arc. Even under argon it sublimed too much before I could even get a load of iron to fully melt (it was only partially melted before I had to shut down. The refractory of the furnace walls also melted somewhat and will be adding a few layers of ITC 100 high temp coating in the future. Being under argon where it could not burn up, much of the sublimed graphite from the anode recrystallized onto the cathode as a strange lump. Next time I will be making a tungsten crucible holder and the arc will be just below the crucible between a pair of arc carbons, as per the original plan which I did not initially proceed with due to the difficulties of welding tungsten rods together with no flux.


[Edited on 15-1-2008 by kilowatt]

P1000595.JPG - 439kB

garage chemist - 15-1-2008 at 19:20

Amazing! What did you make the furnace out of? Is it some kind of castable?
Can you look inside the furace while in operation, and do you have to run a constant stream of argon through it?

kilowatt - 15-1-2008 at 19:50

The outside of the furnace is made of an inexpensive castable refractory that was claimed to be good to 3200°F. I would tend to doubt that claim based on its performance in my larger aluminum/brass melting furnace as well as this one. It works alright when coated with a higher temperature lining like ITC 100 though. For this little furnace the inside form was made from the tops of two 2L plastic bottles cut off and joined together so it is like an ellipsoid and goes out through each end. It just burned out the first time I fired it up.

You can look inside the furnace with the crucible in as long as you've got at least a shade 9 welding lens. For looking at the arc directly a shade 13 is more comfortable. I think if you looked right into it unprotected it would burn your eyes pretty quick, like any multi-kW carbon arc lamp. Even the indirect glare from the top is pretty unbearable especially if the crucible is not in it. I'm sure it could sunburn you pretty easily too, so I wear long sleeves and gloves while running it.

I dump a couple cubic feet of argon through the thing before firing it up, then keep a really slow stream going just to offset what diffuses out the top and through the cracks between the lid and body and between the furnace and the mold box. Later I will be working on getting it sealed up better. I want to get a little fused quartz window (maybe a UV blocking one) for the top to help keep it closed up, but they are expensive. I think a lot of sublimed carbon from the electrodes would build up on it too and it would need cleaned every few minutes. The larger arc carbons like what I had been using for the anode before I made the crucible holder emit a lot of white smoke; I think they contain some kind of binder. At least the decomposing electrodes leave a reducing atmosphere inside which like the argon will be unreactive toward the metals I want to melt in it.

12AX7 - 15-1-2008 at 20:33

Cool!

Note that Ti forms carbides (and nitrides, and oxides, and just about everything else under the sun), so you may have to consider better argon flushing, say, when you get around to that particular beast.

Tim

kilowatt - 15-1-2008 at 20:37

Yes, I will need to keep a significant upward argon flow for titanium to try to keep the carbon away. I think hydrogen could also be a problem. I have also considered putting a tungsten or tantalum lid over the crucible just for keeping shit out of there. It seems like titanium is one of the more difficult metals to cast, given all those little technical problems. It should also be one of the more rewarding.

This thing should also be nice for doing platinum (I have about a troy oz in powder and flake form that I would like to consolidate into an ingot), chromium, small batches of nickel super-alloys (though I am working on a larger gas fired furnace with an alumina crucible that can do those too), and small batches of lower temp reactive stuff like magnesium. I may also be able to use it for making small amounts of calcium carbide, reducing silica to silicon for alloying, and some other high temperature chemical reactions like that. I'm not entirely sure at the moment if any of that could be a problem for a zirconia crucible (by wetting to it or dissolving it), but I'm sure someone here would know.

I would love to make a larger version but can't really afford such a thing.

[Edited on 15-1-2008 by kilowatt]

chemoleo - 16-1-2008 at 18:57

Very nice!
I wondered how you move the anode/cathodes, are they fixed while you are running the arc? When I ran my much cruder arc, without Ar atmosphere, I just tapped the electrodes now and then to bring them closer together. A CaO/C mix sitting just below it easily formed calcium carbide after only seconds of running it, small brownish lumps that produced gas in water that burned with a luminous sooty yellow flame.
Also, what do you mean that the crucible sits on a rotating 'jig', how is this done? I'm also not clear on how you then, under these conditions, manage to rotate the crucible to empty it into an argon-filled mould...do you have a couple of pictures showing this?
I'm also curious as to your refractory material, where did you purchase it? The zirconia crucible? - or did you make any of the refractory materials yourself?

The_Davster - 16-1-2008 at 19:10

I have used commercial arc melters. Even with such a system involving several vaccuum and argon purges, there is still enough oxygen in there to cause colours to appear on titanium when melted. In fact, the operating procedure for a commercial arc melter calls for a small ammount of titanium to be melted before you melt your sample, as melting the titanium first removes, for all intents and purposes, all oxygen from the argon.

The ones I have used, the crucible was made from a pure copper block, that had a network of cavities inside it such that coolant water could be run through it. Samples came out with such a purity that no detectable copper was present. They don't even stick to the copper.

The ones I have used would light up a dark room if unshielded, and special shields were used, which I think were UV blocking glass, with two way mirror character, and the mirror forming coating was a fine layer of gold.

kilowatt - 17-1-2008 at 08:46

Yeah I just tap the electrodes closer together while running. I didn't get any pics of the current crucible holder setup, mainly because it doesn't work very well. The anode which you can see the jumper cables hooked to is connected to a thick U-shaped graphite piece which is hollowed out to fit the crucible rather well. This piece is suspended by the current carrying carbon rod at one side and a tungsten TIG rod at the other side which are each inserted tightly into holes drilled in the anode/holder, and rotates on these. The problems with this are anode erosion, and resistive heating where the rod presses into the holder. The computer render below shows the tungsten jig that I wish to make, but I have been having problems welding tungsten rods together without burning through them. I could get some thicker rods maybe, or use some kind of flux. I can bend the thinner rods by resistively heating them with my welder.

I got the castable refractory on ebay and the zirconia crucible here. http://www.mtixtl.com/index.asp?PageAction=VIEWPROD&Prod...

Couldn't I use magnesium in the furnace as a "getter" to remove oxygen before the titanium has heated up enough to oxidize?

furnaces.jpg - 31kB

Fleaker - 17-1-2008 at 11:07

Very nice, a much cheaper alternative to induction heating. Not quite as convenient, or high quality in pieces produced, but still...extremely hot!

Do you plan on using this for making metal buttons? It is very difficult to use these for casting (particularly for platinum...you will melt it, but not cast it).

kilowatt - 17-1-2008 at 11:35

For platinum all I want is a button really, although I see no reason why I couldn't cast it in a little mold using alumina or zirconia sand. Metals don't seem to wet to the crucible anyhow (at least the small blob of iron I got didn't) which is good. I would like to try casting actual small parts out of titanium, nickel and chromium alloys though. A sand mold using alumina or zirconia with a small amount of plaster or clay binder ought to do just fine. It would be dried out prior to pouring.

[Edited on 17-1-2008 by kilowatt]

Fleaker - 17-1-2008 at 11:39

Oh you could cast it, but you must maintain your heat source on it during the casting process. Just a few seconds of no energy input and it will solidify rapidly. What I am saying is that it can't easily be poured, and it doesn't really flow well either, since it will immediately freeze upon hitting your mold.

Additionally, you do not want to use a sulfate (plaster) as binder because it will readily decompose due to the heat and would not serve its function as a binder. Kaolin may work. Instead I suggest a phosphate. Most platinum investment mixes are phosphate bound, and are very strong physically and my experience with them is that they work well.

kilowatt - 17-1-2008 at 11:51

Oh yeah, I should not need to even turn off the arc at all during the pour; a tiny amount of carbon might be absorbed by the metal stream (especially with titanium) as it goes by the arc but I think that is something I will have to deal with in my final pieces. It would be convenient if I could have a preheated mold for pouring to help the hotter metals flow, but my best bet is probably to heat the metal charge a great deal higher than its melting point to begin with, which is quite possible with titanium and nickel alloys, but not so much with platinum groups or chromium stuff. The metal only has to travel a minimal distance into the mold once it exits the bottom of the furnace, and the argon atmosphere is more insulating than air (though I realize the vast majority of heat loss is through radiation at these temperatures). Obviously a larger batch would help there but at the time that is not within my means.

What phosphates exactly are used as these binders? I am not aware of any phosphates that form a hydrate, nor any that have a particularly high melting or decomposition temperature.

Fleaker - 17-1-2008 at 15:15

Platinum is highly prone to carbide formation as well as the other PGMs.


I won't do everything for you. Query "phosphate bound platinum investment for casting" on google and you should find plenty of references to phosphoric acid (or P2O5 dry mix IIRC, just add water) ''catalyst''. Kerr is one such maker of these products.

The thermal stability of some phosphates is addressed in several threads here on SMDB, with regards to phosphorus production. H3PO4 itself is pretty stable thermally too, if you recall some of the attempts members here made to dehydrate it.

kilowatt - 17-1-2008 at 15:40

Earlier I tried to search for "phosphate bonding" and only got links to some sort of bone therapy. Anyways now I have found the information I need. The reaction between ammonium dihydrogen phosphate and magnesium oxide binds the mix by forming trimagnesium phosphate, as far as I can figure. I would imagine phosphoric acid or its anhydride could be used also as you mentioned, but it might have a different set time. I believe any acid phosphate will work.

Thanks for the help, looks like it should be a good investment mix.