Miniature Arc Furnace for Melting any metal

[Frosty]

The black discoloration is the electrode?

Using the @ symbol with names causes the os problems for Admin and we like keeping on their good side. 

 Frosty The Lucky.

[CGun]

oh sorry. The black discoloration is around the underside of the lid, that is the hole that the electrode enters the crucible from. The arc is about 2-2.5" (about 6cm for you metric guys) below that hole depending on how full the crucible is. The convex shape of the lid is to try to keep the refractory as far from the arc as possible. There is a lot of smoke from impurities burning as well as the electrode slowly subliming away and I am wondering if that is getting into the refractory making it black and "crispy" looking. It is still very hard and doesn't appear to be flaking away, it just doesn't look all that appealing. 

[Frosty]

The black IS at least in part the vaporized carbon from the rod? Impurities being picked up by the refractory would probably alter it's properties maybe significantly.

Glenn sells Plistex for reasonable, I'd give it a try it fires like a coffee cup and should provide some barrier to infiltration.

 Frosty The Lucky.

[CGun]

Thank you. Do I need to remake all the refractory and apply plistex while the refractory is still drying or can it just be applied over the surface of the other fired refractory?

[Frosty]

Apply it to dry and cured refractory in thin coats. It's like painting, lots of thin coats are more durable than one thick coat. Thick coats tend to shrink check and flake more readily. Remember, butter before applying. Buttering is a mason's term meaning wet the surface to which you're applying: mortar, cement, grout, etc. If you put mortar on a dry surface the surface will flash dry the mortar preventing it from setting and curing. It will have a layer of dry dust between the mortar and brick or whatever. Buttering means the mortar or plistex encounters a wet surface and dilutes on contact so it flows into every nook and cranny, the slowed set and cure time bonds it more strongly to the contact surface. 

Remember, butter, Butter, BUTTER. Butter the Kaowool before you spray the rigidizer. Butter the rigidized kaowool before you trowel or cast the hard water setting refractory flame face and butter it before kiln washing it. Plistex is a kiln wash. I'd just paint it over the lid now and see. Its a small area and if it'll prolong the life of that much damage it'll really improve new refractory. Worth trying?

Mix it a little thinner than latex house paint consistency. Let it dry completely between coats and butter before every application. Applying to a layer that isn't dry is worse than applying thick coats.

 Frosty The Lucky.

[ThomasPowers]

The MOB, Mid Ohio Blacksmiths, once got to see an electric arc furnace running at a steel casting company; it used 3 14" diameter graphite electrodes, they ran it only at night when the power rates were lower and the meters on each electrode were calibrated in kilo-amps!  It was a fun tour, one of our members was a metallurgy student at the OSU and met the metallurgist at the steel casting company and asked for a tour.  We went in around 11pm and were out around 2am IIRC.  The Company, in various incarnations, had been there since the American Civil War and the old buildings were chock full of old stuff; somewhat reminiscent of that last scene in "Raiders of the Lost Ark".

[CGun]

A lot of what I applied to this furnace I actually learned by doing research on larger furnaces. A typical DC Arc furnace capable of 300 Ton melts runs at about 600V and 44,000Amps for 26MW of power! It can bring 300 Tons of steel to casting temperature in about 30 minutes with the help of some oxygen lances which blow pure oxygen into the furnace to help heat it (by igniting and burning the steel, releasing a large amount of stored chemical energy). 

But a lot of the concepts actually hold true even at the micro level that mine is. For example, a large arc furnace starts its cycle at a low power rating during what is called the "boring through" phase. It involves the electrode literally boring a hole down through the metal to develop a small molten pool at the bottom. Once the pool is established, the power is turned up until the rest of the melt comes up to temp. This is ubiquitous in melting steel or other metals which are bad conductors of heat (as opposed to copper and aluminum which are phenomenal conductors), but happens sometimes with copper too if there are large gaps between the scrap in the crucible. The Steel directly beneath the arc melts almost instantaneously, without time to transmit that heat energy to the material around it. Thus, the electrode bores down through it rather than melting it as one contiguous block.

For this reason, I built a preheat function into my furnace. It is a sub-routine that runs when a "preheat" switch is turned on. Basically what I have the program do is bring the electrode down, strike an arc, then draw the electrode back up until the arc is extinguished. Then it delays for 200 milliseconds and continues that cycle until the preheat switch is turned off, then it resumes it's normal program, which is what is shown in the videos I uploaded, mainly the machine trying to maintain a constant arc length. I am just waiting on my new control PCB to get here, I design them and have them printed by JLC PCB and then I have to populate it when they arrive... but it takes a while to ship....

This allows the steel a little time to transmit the heat that was just applied by the arc through the rest of the crucible more evenly. I am hoping to do a casting of some single bevel arrowheads I built a master for a while back. 

 

Should be a fun project!

[CGun]

So I ran a test with my new Alumina crucible. It cracked almost immediately. The crack wasn't big enough to allow molten metal to escape so I continued with the test. I had about 5 oz. of steel to run in a quick test. Below was the crucible, loaded and ready to be put under the arc. 

So I ran the furnace for a couple minutes and began to smell a faint sulfur odor. I just assumed it was the refractory drying out and curing so I continued. Total run time was about 5 minutes.

What had happened is that the electrode had bored through the steel, and been arcing against the bottom anode. This had incinerated the anode and the bottom of the aluminum oxide crucible and was evaporating the material... hence the sulfury smell. I knew something was wrong when I looked down into the top and the bottom of the crucible was so bright, I couldn't look at it without a welding mask on (without the arc going obviously). Anyway... I have some Al2O3 glass if anyone is interested in it... ....

 

Back to the drawing board on ferrous metal crucible design...

[JHCC]

Wow.

[CGun]

 So just coming back to this after the alumina crucible catastrophe. So watercooled anode it is... here is the new anode which will be mounted through the bottom of my new 1.6kg silicon carbide crucible it is made of steel and mounted into a water jacket.i fully expect it to melt flush with the bottom of the crucible when in use but if a small puddle is left each time I pour, it should last a good long time.

A real arc furnace anode is made from copper sleeved steel. The copper being useful to both conduct current from the molten pool as well as carry heat away from the steel portion of the anode, keeping it from degrading too quickly. It must work as they claim a service life of about 1-2 years.

Anyway, I just finished refractory for the crucible housing:

 

Once that sets and dries I am going to slowly cook it in the oven, gradually raising the temp up to 450°F to drive off water. I am going to do that 2 times to ensure all the water is gone before firing up the new improved furnace.

I got my new PCB in the mail. This one has inputs for the preheat function which starts an arc for a moment, then draws it up and extinguishes it, then delays for 100ms, then repeats until the preheat switch is shut off. This allows gradual heating of both the steel and crucible to reduce the "boring through" cycle and minimize the chances of the arc operating right over the anode and the crucible bottom. 

 

Arc temperature decreases hyperbolically from the arc contact center. There the arc can reach 13,000- 20,000°F when running at 100 amps. Reference here:

https://www.google.com/url?sa=t&source=web&rct=j&url=https://app.aws.org/wj/supplement/WJ_1976_08_s222.pdf&ved=2ahUKEwj_5qKDlI_qAhXGZc0KHX68AggQFjANegQICBAB&usg=AOvVaw3MwRLCe3xxUJalRaiNIiMf

 

This decreases several thousand degrees just a few millimeters from the epicenter. So the whole goal here is to keep the arc only in contact with the steel and nothing else. We will see how that goes. In any event, silicon carbide in a graphite matrix (I know, graphite in contact with liquid steel is a no-no) should be able to deal with the thermal shock and the extreme temperature better because it is about 3-4x as thick as the alumina crucible and it is more refractory than alumina.

 

Anyway, that's where the project is currently at. I have a ceramic shell mold made for the arrowhead... now I just need to build kiln to fire it into dead and set it...

[Frosty]

Let me see if I remember what you've said earlier without rereading all the posts. The preheat cycle either melts helps melt a puddle to cover the anode and prevent anode to cathode arc? 

Might it help to lay a disk of steel or perhaps a layer of steel powder or swarf over the anode? The disk to approximate a puddle and or the fine particles melt more quickly to form a puddle? 

Just some Frosty thoughts unpolluted by knowledge for your amusement.

Frosty The Lucky.

[CGun]

It's funny you say that. I just found some copperhead bb gun bbs that I am going to use to cover the anode. This will allow good electrical contact and the copper plating on the bbs is no thicker than the copper on TIG rods and shouldn't mess with the chemistry too much. That may be the startup procedure from here on out, add about 30 -40 bbs to the bottom of the crucible, then scrap steel on top, then preheat till the whole mass is glowing orange (1500-1700°F), then run the normal program while incrementally increasing current to the max for this furnace... about 60A.

[Frosty]

I wasn't completely off base!? Woo HOO!  

Little copper plated spheres. Wouldn't the copper plating provide some Faraday-like effects, conducting electricity sort of evenly through the whole mass? Mass of BBs that is. Heck, they even look like little seeds. Hmmmm, 30-40 BBs whats that a Tsp +/-? 

Frosty The Lucky. 

[CGun]

I dont know about the faraday effect... there probably hasn't been any research done on the subject "the effect of electromagnetic flow profiles through an entropic mass of copper plated bb gun bbs." There!someone is free to use that as a PhD. Thesis if they would like... I would just like credit for the title.

[Frosty]

Don't forget me! It was my almost utter lack of knowledge other than some casual lay reading that lead to the thought! I want due credit too!

Hey wait a second. Isn't writing a thesis mostly coming up with something plausible about something someone else hasn't thought to write about? Is there a doctorate requiring a PhBBD? Might have to invent one of those too. Hmmmm.

Frosty The Lucky.

[ThomasPowers]

Research sponsored by Red Ryder and Wham-o.

[CGun]

So I was talking to a friend of mine yesterday and he suggest I make one of these to sell once I get it working and reliable for multiple melts of steel. Do you guys think something like this would actually be sellable? Is there enough interest in ceramic shell casting of steel at home or would people simply buy a larger induction furnace if they were into this? Thoughts? I would probably offer a furnace like this initially for like $1,500 complete, or potentially a kit where I provide the micro-controller and the control system (electrode holder, cooling system, etc. basically the entire upright assembly and crucible in the videos) and the buyer provides the welder/power supply. Thoughts? 

[ThomasPowers]

What's the going rate for a multi million dollar liability insurance policy?

[CGun]

about $40 a month. 

I would need to sell quite a few of them I guess if I wanted to do that.... 

[Phil K.]

https://www.youtube.com/watch?v=BHRIFVubO60

This guy does it and gives step by step instructions. I think he's a little nuts.

[Frosty]

Oh come on Phil, how nuts can he be? He's based the build on "King of Randumb" plans. 

Interesting but I think I'll pass on making one of those arc melters.

Frosty The Lucky.

[ThomasPowers]

Ouch; he must have Darwin on speed dial!

[CGun]

That seems like a recipe for Disaster. Handle holding AC welding clamps in each hand (most of those welders will put out 80+V OCV, which is enough to kill you if your hands are a little sweaty and you happen to be holding the contacts in both hands so the current can pass through your heart). I mean... I can just put a bunch of black powder in a soda bottle with a fuse and make it go bang... but that doesn't make me Lawrence Livermore Laboratories.The steps seem to be, and mind you, I started this way so I guess I can't make too much fun without sounding dumb myself but... well... here it goes:

1. Get a welding power supply... ( doesn't really matter, Microwave transformers will obviously do)

2. Touch the stinger and ground to your tounge... if you get shocked, you know it is working (this is the quality control, you don't want to build the whole thing to find out it doesn't work.  ).

3. Control the current of your homemade transformer using a DEATHPOOL (Design Encompassing A Total Hack-job Proliferation Of Online Lore) conductive water resistor (for the record... no welding machine EVER has controlled current directly using a variable resistor because it is a very dumb way to go about it.... making one you could inadvertently stick your hand into, is Darwin award material). BTW, PWMs are cheap and infinitely better in this regard. 

4. Spark the ends to get the uncontrollable arc to light.

5. Hope that your metal melts before your 2,300F Insulating Refractory brick under the heat of the 13,000F arc. 

6. Make a video about it and your smug, yet ignorant, confidence that nothing could ever go wrong with having a puddle of molten metal in a container not designed for it, being heated by a DIY electrical device with no fail-safes and propose publicly in your video that for a minimal investment you too could make this deathtrap!

[JHCC]

On 6/20/2020 at 8:22 PM, Frosty said:

Isn't writing a thesis mostly coming up with something plausible about something someone else hasn't thought to write about?

A friend working on his doctoral thesis mentioned the other day that one of the first steps in doing the dissertation is to give an overview of the current state of research on one's chosen subject and explain how one's own thesis either fills a gap in that research or approaches one aspect of it from a different angle.

[CGun]

See? So the thesis idea above would totally work... no one has probably ever done research on the electromagnetic properties of a mass of .177 caliber copper coated steel bbs designed for bb guns! Whoever writes that thesis had better have a hyperlink to this page in their appendix.