Hi This thread will trace the feasibility, design and hopefully construction of a 5kW electric furnace capable of melting cast iron. Dazz
I was thinking of how silicon carbide crucibles are not suitable for molten iron due to the solubility of silicon carbide in iron, so there may be some need to protect the element from iron splashing.
Resistance. I am starting with a self-imposed limit of 5kW. 5kW is the standard power rating for a domestic electric oven. I already have a suitable 20A power connector set I can connect to the main switchboard. From this starting point, I will need to figure out if the performance is good enough to melt enough iron. I used to work with high powered radar systems so I know I could design and build an induction furnace. A mains powered induction furnace would produce a combination of lethal voltages and hot insulation. An induction furnace would be the option of last resort. Dazz
I can already see that minimizing thermal loss will be important. I am thinking more along the lines of an inverted furnace. Dazz
This is why I asked in your original post if your aim is to build an electric furnace or to make castings with it.......not trying to be a wise guy...I'll explain. Presuming you're able to achieve iron metal temperatures, you won't be able to melt much metal with 5kw.....maybe a kilo or so and it will take a long time. If all you want to do is make a small puddle of Fe, then that's ok, but your casting size will only be a fraction of your melt size with feed/gating/risering, and that small of a melt will lose heat very rapidly. Is casting a 1/2kg part useful to you? Resistive versus Induction: To me resistive wins for a hobbyist because it doesn't care how metal type or charge size varies. Induction does. If you set out to build an induction furnace you quickly accept it becomes an exercise in building a high frequency, high current/power density, power supply with a complex feed back loop to adapt to charge size and metal type......ferrous versus non-ferrous inductive melting are different games. If you are just going to melt a few ounces of metal then the small commercially available induction systems are probably viable. If you want to make multiple kilo castings, take a look at what a commercial induction furnace power supply looks like for doing that........not within your available power limit or budget. Induction may provide higher ultimate furnace temperature than resistive which is limited by the operating temp of the heating elements, but it's far more complex. Dedicating it to a given metal and charge size helps complexity and hurts versatility. Building a resistive furnace with metallic elements is very inexpensive and simple.....but you're limited to perhaps bronzes being the outer limits of your melting capability. If you damage a metallic heating element it's no big deal. Just twist up another $15 worth of Kanthal. If you damage a SiC or MoSi2 element (and you will), you may be out $1000s. As I mentioned, ceramic heating elements can approach 1800c operating temps but they behave much differently than metallics, namely a very large change (increase) in resistance from cold to hot. The element manufacturers design for low resistance high power furnaces. For example, I inquired about a single element from a Chinese source because at 10pcs they were $58 each. They were ~1500watts each. When I further inquired the resistance was .068ohms, the suggested operating voltage was 10.2v.......that's 149 amps! See what I mean? When I told the manufacturer I thought that was impractical, he said just use a transformer. I said oh, the kind of transformer that step down 10x or 20x, are 10s of kva, and can handle hundreds of amps? Seriously? -Moving on. Now a very long high resistance spiral element could be made but it would most certainly be too fragile. So then, the clever back yard guy says, hey, I'll just take a bunch of those low resistance elements and connect them in series. That'll get the voltage up and the current down to a more practical and manageable level and if I do have a mishap, hopefully it would just require replacing one element. This is the approach taken for most smaller power SiC or MoSi2 furnaces. But even so, 10 elements only gets you to ~.6 ohm. I'd want more like 12-14 ohms......no worries....that's only 200 elements LOL! Now if you could get a single rod that was about 1-ohm, I liked the idea of a cylindrical array of rods with the top of one rod connected to the top of the adjacent rod, same on the bottom so the whole array is connected top to bottom in series. They have to be mounted non-rigidly so they float otherwise thermal expansion will break them. They have to be electrically connected with flexible (possibly high temp) electrical conductor for the same reason. You still need to build at least a 3-step voltage power supply or the initial current inrush will break the cold elements, and you also have to come up with a temperature sensing scheme because you are going to have to actively control the power supply with it. This is no problem up to say 1400C, but above that, your choices for something that can live at 1800c will be few. The last thing, if you've ever used a foundry crucible furnace, during the heat you will have to access & assess the melt to add charge, inoculants, gauge temp, remove slag.....etc. Unless you have a lift off furnace, you'll have to insert tools to grip and extract the crucible. If you have exposed heating elements on the interior wall.....you will damage and break them with tool or metal contact. A lift off (or bottom load) furnace helps but adds it's own set of complexities. So circling back, what is you want to do? If it is just melt iron, and for those of you doing it, I don't mean to minimize the task, but you can realistically expect to do that with a few hundred dollars worth of refractory materials, some pipe fittings, a forced air propane burner, and have multiple to say 5kg-10kg capability in a crucible furnace. Alternatively, for not much more, you can build a resistive electric with metallic elements and maybe reach bronze temps. I melt 20lbs of aluminum with my 8.5kw resistive unit in about 30 minutes when the furnace is at temp. Food for thought! Best, Kelly
The only guy I've seen on Youtube with a resistively heated brass furnace is the guy linked below, He's using modern switchmode welders to drive his heating elements (cheaper and better than a transformer) and has built an electrically heated crucible to melt brass. I'm pretty sure the insulation wool shown in the videos is asbestos too . Silicon carbide can be heated directly with microwave energy and it's possible to use several cheap microwave oven magnetrons to excite a single chamber as they will tend to phase lock to one frequency (according to one experiment that I read about) but even then it's only good for non ferrous use as a silicon carbide crucible will get soft and dissolve in iron at iron temps. Also the idea scaling up to 10+ magnetrons for 7 KW is going to be spectacularly dangerous. Some scientific companies sell tiny SiC crucibles for use in common microwave ovens to melt stuff like zinc.
Hi Kelly Thanks for taking the time to write your post. It makes good reading. I accept that a diesel/oil furnace is the fastest/cheapest/easiest solution but I live in a quiet residential street with nice neighbours who are unlikely to display much tolerance to a roaring furnace and bad smells. The clean quiet electric option is worth pursuing in my situation. If I was going down the induction path (no intentions), I would start with multiple cheap induction cooker elements each connected to a one or two turn coil and synchronised. The exposed voltage at any point would be a lot less than seen on a single multi-turn coil. Microwave excitement is too dangerous. By the time you feel a burn, serious injury has already occurred. Non-lethal burns affect deep tissue and can take months to heal. Victims exposed to lethal levels of RF usually take a couple of days to die. A welder power supply looks ideal for driving a SiC element. I happen to live two hours from the engineer that designs welders for Jasic, UniMig and RazorWeld. The recommended method of driving SiC elements is individually. It will be relatively cheap and easy to design and build a networked driver with the capacity to power one element. The advantage of that is it makes any design scalable. Need more heat, just add more elements and drivers. I have not committed to building an electric furnace and I am not ignoring your advice. I am beginning to undertake a desktop engineering study to figure out if it can be done. Dazz
Found this. Keith Company Electric Furnace They use Molybdenum-disilicide elements at 13kW / cu.ft to get to iron melting temps. Dazz
Melterskelter was able to engineer an extremely quiet oil fired furnace by having a refractory lined chimney attached to the furnace lid: it smooths out the rumble pulsations drastically. He has a description of it on this forum: https://forums.thehomefoundry.org/i...ng-noise-from-an-oil-fired-furnace.503/page-4
Oh it can be done, without doubt. Has been done....many times. There are many manufacturers like the one you posted. Just search MoSi2 or SiC muffle furnace and you'll get a bunch of hits. Commercial units are well into the 5-figures $ and most all are laboratory and process ovens not particularly well suited for foundry work. They may not like the door being opened at 1800c to extract a crucible but that's life for a foundry furnace. It also doesn't have to use diesel or oil for fuel and strong odor is a misconception for a properly tuned oil burner. CH4 or C3H8 will get you there too. Noise can be addressed too as Melterskelter has shown. Happy to help either way. In a former life, we built such ovens for processing highly engineered metal oxides. But in this case, very close temperature and atmosphere control were important so carbon fuels were not a consideration. A metal melting furnace by comparison is a blunt instrument. Like I mentioned, I went quite a ways down that path (high temp electric furnace) myself and wouldn't rule it out in the future, but until I want to cast something that requires an 1800c furnace temperature, I won't invest the time. Being a casting forum, most here would approach such a problem from I want to cast "X" object, and therefore I need a furnace capable of "Y". You've come at it from I want melt iron with 5kw of electricity, and cast what you can with the resulting a capability. Beyond that it's just a matter of the time and money necessary to accomplish the task. Best, Kelly
One additional factor to consider in a resistance furnace is damage to heating elements due to drips of iron and slag. These are unavoidable, at least in my hands. Skimming slag is a messy process. This is especially true when removing largish chunks of slag. Toward the latter stages of each melt 5 to 10 percent of the total melt weight of slag has to be scooped out of the crucible. I am as careful as I can be, but I still drip a few drips of iron and drop an occasional bit of slag into my furnace. The material is hard refractory, but would immediately disable coils. They would either short out with spectacular vaporization of the coil or the slag would eat them up in fairly short order. It is also easy to splash a bit of iron when adding metal to a melt in progress. Some sure-fire method of keeping slag and iron off the elements would be necessary. That sounds like a challenging requirement. Denis
I worked on large induction furnaces for melting copper to make copper pipe. The furnace had an E shaped passage in the floor with the 3 legs vertical with a coil around each connected to 3 phase 480 volt power. Essentially a transformer with the secondary shorted. They had to pour molten copper in to create the circuit on startup.
I'd read that there were 60Hz induction furnaces that ran directly off three phase mains and had to have molten metal to start them before running 24/7 days a week.
How many kilowatts are you looking for? 5, 10...50? 5 kW you're looking at 20ish A, how many amps do you have to give??? Sorry, I want to do this backwards. What is your max amperage and how many phases do you have and at what voltage? I guess, I want to get the cards on the table. At 5 kW you might have or be able to produce 100 mL of cast-iron in an hour. 1kw=3412 btu. As a comparison my barbeque puts out 20,000 BTU. I had a 6000 W heater in my shop that just would not stop. I now have a 30,000 BTU natural gas heater that runs only three times an hour.
Hi Kelly It would be more accurate to say that I haven't fully documented my requirements on this forum. 5kW is a convenient starting point. I have enough power capacity to drive a 22kW load. I already have a 4kW VFD for my lathe.