Open Source Induction Furnace

You may have heard us talk about recasting civilization from scrap metal. Metal is the basis of advanced civilization. Scrap metal in refined form can be mined in abundance from heaps of industrial detritus in junkyards and fence rows. This can help us produce new metal in case of any unanticipated global supply chain disruptions. This will have to do until we can take mineral resources directly and smelt them to pure metal.

I look forward to the day when our induction furnace chews up our broken tractors and cars – and spits them out in fluid form. This leads to casting useful parts, using molds printed by open source ceramic printers – these exist. This also leads to hot metal processing, the simplest of which is bashing upon an anvil – and the more refined of which is rolling. Can we do this to generate metal bar and sheet in a 4000 square foot workshop planned for Factor e Farm? We better. Technology makes that practical, though this is undeard-of outside of centralized steel mills. We see the induction furnace, hot rolling, forging, casting, and other processes critical to the fabrication component of the Global Village Construction Set.

We just got a $5k commitment to open-source this technology.

Furnaceicon

The program for delivering an induction furnace involves a high frequency, high-power supply (between 20 and 50 kW (the latter can be gotten for $1600 on Ebay), and the melting chamber proper. Well, we could buy a turnkey system perhaps for $5k total used, and run it from the LifeTrac generator. The only disadvantage to this route is that if it breaks we’re dead-in-the-water – either with the impossibility of fixing closed-source technology, or a high repair bill. A single component which blows and is inaccessible for fixing could in principle turn a working power supply into worthless junk. Thus, it is worthwhile to tame this technology by open-sourcing the design.

It is more robust and cost effective to open-source the high power induction power supply. We would like to exactly this – which is mature technology. The question is finding a consultant or developer. We are thus making a call for bids. We are offering between $100 and $3000 for a design – depending on its level of completion. We will post this on eLance and other outsourcing sites.

If you are qualified to produce a design, please submit a bid by emailing opensourceecology at gmail dot com. We’re open to other suggestions on tactical and strategic aspects.

The bids will be judged on the demonstrated competence of the bidder, and upon the extent of design completion. The complete design would include all of the following:

  1. Design of an induction furnace circuit scalable up to 50 kW in units of 1 or 5 kW
  2. Design allows for power and frequency selection
  3. Power source may be either 1 or 3 phase electrical power.
  4. Specifications of a cooling or heat dissipation system
  5. Adaptable design specifications for primary coil windings
  6. Geometical design of melt chamber and basic power transfer calculations
  7. Melt chamber includes provisions for loading and pouring.
  8. Complete bill of materials
  9. Fabrication files for circuit and other components
  10. Sourcing information for components
  11. System design and process flow drawings

Serious bidders only, please. Naysayers on feasibility of this proposal will be either dismissed summarily, or our design criteria will be modified accordingly.

With up to $3k allotted for design work, we believe that the remaining $2k would suffice for parts for the actual device: about $1k for the electronics, and $1k for the furnace chamber.

No related posts.

24 Comments »

  1. Alan Millar said,

    December 15, 2009 @ 3:54 pm

    In addition to ceramic 3d printing, there are other computerized ways to make metalcasting molds. I recently read about a reprap experiment to use a PLA plastic 3d print to make a plaster investment mold. The lost wax casting method, except PLA instead of wax.

  2. Alan Millar said,

    December 15, 2009 @ 3:56 pm

    See http://objects.reprap.org/wiki/Casting

  3. Justen said,

    December 15, 2009 @ 4:48 pm

    I’ll spread the word around, good luck guys :)

  4. Rasmus said,

    December 16, 2009 @ 9:24 am

    I have a few suggestions for integrating this with other components. This is where product ecology can work its full magic. Read on… it’s less complicated than it looks at first. Elance is a good idea (Innocentive being another one, possibly better), but the people there may not be familiar with the OSE concept of product ecology (design integration). The first requirement for the job should therefore be to familiarize onself with the entire construction set.

    - the electricity for the induction furnace will probably come from an internal combustion engine that runs an electric generator – OK. Rather than running on biodiesel, this engine could be driven with “woodgas” from a biomass gasifier (pyrolysis unit). The process could also produce some biochar, making it actually sequester carbon every time that steel is recycled in this way !

    - Since induction melting works simply by elevating the temperature, considerable energy savings could be realized by pre-heating the scrap metal. Waste heat from pyrolysis (400-500C) can be used to pre-heat the metal parts, which are then thrown into the induction furnace.

    - other ways to pre-heat the metal can also be integrated, such as solar concentrating power when available (e.g. using a heliostat or solar funnel). This might shave another 25% off the heat difference needed to start the melting.

    - These processes obviously give off a lot of waste heat. This could be captured with water that is then put into a large underground tank for seasonal heat storage. The internal combustion (woodgas) engine also benefits from a cooling mechanism and should have water pipes running through it, transporting away the waste heat. Buffering away the heat also makes sense from a workplace safety standpoint.

    - other possibilities for integration (again, pre-heating metal) are with as well as the production of fired bricks (for heavy-duty applications where CEBs may not be sufficient; temp. of 900C).

    - the possibility of a mobile unit should be considered. Scrap metal is bulky and it may sometimes be better to transport the unit to a particular site to melt steel there, then transport the cast products away.

  5. Marcin said,

    December 20, 2009 @ 1:04 am

    one note recommended – “One place to look in academia is the guy that has tested out “microwave steel” where he heats with microwaves plus induction. The induction gets the whole crucible and
    scrap going with resonant currents, the microwaves are absorbed by the special crucible material.”

  6. fenn said,

    December 20, 2009 @ 12:07 pm

    i’d like to point out that frequency is inversely proportional to heat penetration depth; since you’re melting a big pot full of metal the frequency doesn’t have to be very high. this means you don’t need unreliable/impossible to repair switching transistors; instead you should use an alternator, three phase motor, or stepper motor acting as an electrical generator to drive the coils. the motor has to be geared up to reach high enough frequencies of course.

    why are you outsourcing the design to e-lance instead of the usual open development process?

  7. Marcin said,

    December 20, 2009 @ 7:26 pm

    We are using an open development process, with the addition of funding for potential acceleration. Funding opens up certain development options, which may otherwise be inaccessible.

  8. Rasmus said,

    December 21, 2009 @ 12:02 pm

    I looked up “microwave steel”: http://www.imp.mtu.edu/information/microwave_JAN_04.htm
    and wondered whether replacing the coal in this experiment with simply carbon monoxide might work. The CO could be inexpensively regenerated with solar concentrating power, as has been demonstrated in this device:
    http://www.youtube.com/watch?v=8tt7RG3UR4c

  9. Benjamin Gatti said,

    December 22, 2009 @ 5:06 pm

    This is a good primer

    Capacitor reference

    I note the DIY schematic is oldschool (no microcotroller). I suspect that the coil excitation and system monitoring and control can and should be done via Arduino.

    The use of Arduino would dramatically simplify the front-end electronics. It’d be fun to work out a microcontroller version of the design above.

    The natural progression might be smaller power units, working up to larger power. I note that the above DIY design has fairly modest power output.

  10. Marcin said,

    December 23, 2009 @ 7:36 pm

    @Rasmus: Very interesting on the microwave steel. This is along the lines of http://openfarmtech.org/index.php?title=Humanity_Plus_Presentation#Slide_4 – namely, that advanced industrial processes are done readily on the scale of your desktop – as long as you have access to high quality energy (electricity), as discussed in the slide previous to the link above.

  11. Marcin said,

    December 23, 2009 @ 7:37 pm

    @Benjamin – it is worthwhile to consult with the author of your first link on insights of power scalability of the induction furnace, and on adaptation of the circuitry to Arduino. Excellent link.

  12. John Griessen said,

    January 8, 2010 @ 12:30 pm

    I really like the idea of tailoring the system to your use by driving the induction coils
    with a motor generator that outputs maybe 5KHz
    instead of the usual 20KHz of a cooktop. For melting steel, white heat is needed, so learning about refractories is very high on the list. Having a lid on your crucible is required to keep in the heat. Besides microwave-steel another area to research is how the big recycler of steel in the US uses three phase power to three big carbon or Si-C electrodes spaced farther apart than
    they will arc without having steel below them. They have adjustable height
    and a controller that keeps the current going steady by changing height
    of electrodes. They melt many tons at a time, but maybe it would work smaller — scale could matter much and small scale could require small chunks of scrap to work well. Induction has no such limitations on scrap shape. That and a three hole lid might do everything you want with a motor-generator electricity source, and some crucible design.

  13. Ram said,

    January 25, 2010 @ 1:05 pm

    Hey Marcin y’all keep continuing to amaze me, have y’all tried to open source “Thermojet” its a wax based 3d printer used for rapid prototyping of investment castings. Investment casting would be very suitable as it is very accurate and extensive knowledge exists about it. It is also used in aerospace and other high tech sectors(open source aviation ;-) )?? I also had left a comment on your open source lathe page do check it out. How is the solar energy project going? I also wanted to know the whats the main issue y’all are facing about solar thermal energy- is it just about storage of energy? Have y’all looked at any other working fluid instead of water like co2,ammonia etc??

  14. Marcin said,

    January 25, 2010 @ 1:22 pm

    Ok, investment casting sounds useful. Please start a page on the wiki – http://openfarmtech.org/index.php/Investment_Casting

    The main issue on solar energy is the low-cost heat engines. We found that turbines are much less efficient than modern steam engines on the small scale, so we’re developing an open source steam engine. This is a universal engine for cars, tractors, solar electricity, industrial power – based directly on the sun or stored sun in pelletized biomass. The first step to the modern steam engine is the lathe, to be reported in an upcoming blog post by tomorrow.

    Subscribe!

  15. Elifarley said,

    January 25, 2010 @ 6:28 pm

    I’ve started a wiki page about this Open Source Induction Furnace Project.

    I’ll add some comments from this blog post to the relevant sections there. If you think there’s something important missing, feel free to help!

  16. Initial Steps to the Open Source Multimachine | Open Source Ecology said,

    January 26, 2010 @ 2:24 am

    [...] runs on electricity. We have also secured seed funding to opensource the process, for which we are currently soliciting bids. If we can succeed in building an open source induction furnace, than we have created a low-cost [...]

  17. Marcin said,

    January 27, 2010 @ 12:33 am

    Ok, I posted the project up on Freelancer, so you might want to take a look – https://www.getafreelancer.com/projects/Electronics-Mechanical-Engineering/Open-Source-Induction-Furnace.html . My experience so far is that Freelancer doesn’t reveal their fee structure anywhere, such as in their terms and conditions. Only at the end of the registration process they showed that they charge 3% commission based on bid amount.

    This looks good. I wonder if we get a large number of responses as they suggest. That would be great.

    We’ll see how the process goes – because I think the project is complex enough that for some reason I doubt that anyone can fill in all the requirements. We might have to end up getting only a small part of the project done, but we’ll see. This is definitely new waters.

    I bet you that those guys in China have stores of relevant engineering knowledge – such as circuit design for just about anything. It would be good to tap into that – especially if these guys were really good and knew what they were doing – and could customize their design on demand to meet our specifications.

    I bet you that’s how China is able to do well these days – they just copy proven designs. Well, that’s exactly what we’re doing, except we develop products with the explicit aim of open-sourcing them.

    In any case, it may be that the freelancing business could be a great asset towards developing a scalable, open source product development pipeline. At best, a formal procedure can be created to fund, develop, and deploy open source products along the lines of the roughly-defined Open Source Design Rationale (http://openfarmtech.org/index.php/File:Slide1Linz46.jpg) process. Experience has shown that qualified project developers are perhaps the hardest part of scaling open source product development. This may be a solution

  18. Marcin said,

    January 28, 2010 @ 12:06 pm

    It looks like we found an excellent candidate for designing the Open Source Induction Furnace at getafreelancer.com. The current bid is $2500, from an individual specializing in induction heating. This person appears to be well-versed in control and power circuits, as well as the design of the induction coil and of the balance of system. Here is my response to him, as we’re interested in pursuing this option:
    —–
    Hello, you appear to be most capable of delivering what we need because of your specialization in induction heating. Are you also able to document the design well so that it becomes replicable to others? Our mission is to open source different technologies to make them accessible and understandable by people in general.

    First of all, some questions:

    1. What is the power of the largest induction heating system that you designed?
    2. Have you built any of these, or have they been built for real?
    3. If the heaters have been built, can you send me pictures or further documentation of their performance?
    4. Do you have experience with design of the cooling system?
    5. Do you have experience with induction heaters applied in particular to metal-melting applications?
    6. Do you have experience with the metal melting chamber design and materials issues involved?
    7. Can you break down the project and estimate how much time it would take you to accomplish each of the steps of the project: (1), control circuit design and key calculations; (2) scalable power circuit design and heat dissipation calculations; (3) inductor construction design and design calculations, (4), cooling system design and integration; (5), melting chamber design for effective loading and pouring
    8. Is it possible to utilize the open source Arduino platform for the control part of the circuit?
    9. When you design the circuit, do you typically specify the exact components that we need to use, or do you just specify general values? We need to know the exact components, so the job is easier for us. The design of the control circuit should be such that we can outsource its building directly to a PCB manufacturer or such that we could build it in-house if the circuit features are sufficiently large so that they are doable with inexpensive circuit production methods. We need suggestions on how to outsource the fabrication of the power circuit, or instructions on how to build it in-house. Are you able to provide this?
    10. Presently, we have 20 kW of single phase, 240VAC available from a tractor power generator. Is this a suitable power source, or should we upgrade to 3 phase power?

    If you design the circuit:

    1. Can you design it such that we can scale it in power from 1 kW to 50 kW (or any amount) by adding on additional power modules? Modularity is important for our work, because we are interested in very flexible, and adaptable systems. Our interest is melting metal for casting/hot rolling, as well as preheating metal for things like forming or threading, among many others. We are interested in opensourcing this design for the world, consistent with the goals of Open Source Ecology.
    2. Can we break the project into phases so we can build/test the particular component so we make sure it works? Do you have suggestions on how we organize the project so we can do tests and know that your designs are actually working? What is your track record of producing design that works, and can we come up with a contingency plan if things do not work? We are using money from investors, and we need to be accountable for our performance.

    I look forward to continuing this discussion.

    Thank you,
    Marcin Jakubowski, Ph.D.
    Open Source Ecology

  19. Marcin said,

    January 28, 2010 @ 8:45 pm

    I just talked to another power electronics guy from Texas, Jacques, and he’s much interested in the open source project. We began discussing the controller part of the power circuitry. We hope for good progress soon.

  20. Jason liu said,

    January 29, 2010 @ 6:59 pm

    Our company supply different sizes of induction heating power and induction furnaces. Please send to [email protected] for more details.

  21. Marcin said,

    January 30, 2010 @ 1:20 pm

    Response on the getafreelancer.com bid:

    Hi Marcin. Thanks for your letter. My name`s Sergey Cobin. I am a head of engineering group from Ukraine, “Sigma electronics laboratories”, Research Department of State Donbass University. Our group works at the area of power electronics. For present we develop three chief directions. 1 AC and DC Drives for power range from 1kW to 125kW; 2. Induction heating invertors for power range 5kW to 250kW; 3 AC Stabilizers for power range 10kW to 50kW. I try to get you answers for your questions. 1.I have been developing induction heating systems for power range from 5kW to 250kW for frequency from 2.4 kHz to 66kHz . 2 All systems is real systems, not papers. I finished at October 2009 similar project for power range 50kW and frequency 4 kHz by the order private firm “Ukraine export” , for melting of silver scrap. Melting processes take about 30-40 minutes for 25 kg for high-scrap charge, and 70-90 min for 40-50kg high-scrap charge. Preferably ratio for output power of invertor to weight of scrape is 1.5-2kW/ kg. 3. I send you some promotion material in attachment. 4. I can develop and calculate of cooling system, as for this project it will be water forced cooling, water discharge is 3 liters /minute. 5. I have experience as for inductance heating for metal-melting applications, usually for melting of ferrous metal used packed heater. 6. I haven`t break down project to present, all my projects has success. As for spent time, I can divide it next manner – control circuit design, constructions and key calculations- 20%, power circuit design and constructions- 30%, inductor construction design- 15%, cooling systems design- 15%, melting chamber design- 10%. 8. I don`t use Arduino platform for building of control but I use Atmel controllers for design of control systems. If I will get the Arduino I can do this development. 9. I will propose you exact components and materials. If you can`t get it in USA, I can propose alternative variants. As for manu facture of printed boards for project I will send you gerber files for printed board manufacture. In addition as for printed board I can send you finished goods by air post. I`d like to say that control systems is inexpensive systems, components and materials costs about $50. But power part of the project has much more cost. I will get you all necessary suggestion for manufacture it at home, but you will need some experience of building of power electronics systems, experience of welding and furnace, experience of using of measurement instrument as for oscilloscope, voltmeter and ammeter. This experience is 50% of success at realization of the project. 10. For 50kW output power, single phase 240V input power supply is incorrectness selection. Most probably your generator will get overload. Usually we use a three phases input power supply from 10 kW to higher load. Modularity will be OK!. Before starting of project you will need to clarify of purpose of your project. I t means that heating for melting differs from heating for pressing and other induction heating technology. It`s differ technologies and differ tasks. If it`s possible than please choose one of it. At first for calculation I will need to know maximal and minimal weight of charge of the heater, kg and desired heater capacity, kg/hour. Do you plane to do heating of ferrous metal or non-ferrous metal? It is clear, you can break the project into phases for test of circuits but it will take more time than 25 days. I suppose that for installations work you will need about 20-40 days, manufacture of inductor, reactor and melting chamber takes about 7-15 days, starting up works takes about 3-7 days, maybe delivery of materials and components takes 10 days. In other words, realization of the project takes about 2-3 months. You will need oxy-acetylene welding and cooper-silver solder for manufacture of cooling systems, inductor, reactor, cab-tire cables and other power elements . I will plane to divide the project on stages. After realization of an each stage I will send you information for realization next stage, step by step we will do all project. Yours sincerely Sergey Cobin

  22. Marcin said,

    February 1, 2010 @ 11:10 am

    Second bid response and my questions from Getafreelancer.com:

    I’m an electrical engineer working in the field of designing induction heating systems. I already designed a 250kg unit based on swept frequency technique. I’ll deliver you as soon as i can a full design for your unit based on load resonant technique which has the advantage of low melting time and good power utilization.
    —–

    Thanks for your response. Questions:

    1. Does your bid include design of the induction coil, cooling system, and melt chamber in, or just the power electronics?
    2. With load resonant techniques, what power/melt mass/time ratio can you deliver for scrap steel? Our primary goal right now is scrap virgin steel and sscrap used steels (broken tractors, old implements, fencerow scouring, etc)
    3. Can you make a general power unit, which we could then apply to other purposes (other induction coil systems, for things like heating rod for threading or forming, etc)

    More questions:

    1. Do you have experience with the metal melting chamber design and materials issues involved?
    2. Can you break down the project and estimate how much time it would take you to accomplish each of the steps of the project: (1), control circuit design and key calculations; (2) scalable power circuit design and heat dissipation calculations; (3) inductor construction design and design calculations, (4), cooling system design and integration; (5), melting chamber design for effective loading and pouring
    3. Is it possible to utilize the open source Arduino platform (Atmel microcontroller on a breakout board)for the control part of the circuit?
    4. When you design the circuit, do you typically specify the exact components that we need to use, or do you just specify general values? We need to know the exact components, so the job is easier for us. The design of the control circuit should be such that we can outsource its building directly to a PCB manufacturer or such that we could build it in-house if the circuit features are sufficiently large so that they are doable with inexpensive circuit production methods. We need suggestions on how to outsource the fabrication of the power circuit, or instructions on how to build it in-house. Are you able to provide this?
    5. Presently, we have 20 kW of single phase, 240VAC available from a tractor power generator. Is this a suitable power source, or should we upgrade to 3 phase power? We are considering a 50 kW 3-phase generator.

    If you design the circuit:

    1. Can you design it such that we can scale it in power from 1 kW to 50 kW (or any amount) by adding on additional power modules? Modularity is important for our work, because we are interested in very flexible, and adaptable systems. Our interest is melting metal for casting/hot rolling, as well as preheating metal for things like forming or threading, among many others. We are interested in opensourcing this design for the world, consistent with the goals of Open Source Ecology, our organization.
    2. Can we break the project into phases so we can build/test the particular component so we make sure it works? Do you have suggestions on how we organize the project so we can do tests and know that your designs are actually working? What is your track record of producing design that works, and can we come up with a contingency plan if things do not work? We are using money from investors, and we need to be accountable for our performance.

    I look forward to continuing this discussion.

    Thank you,
    Marcin Jakubowski, Ph.D.

  23. dmitriy said,

    April 21, 2010 @ 12:32 pm

    What is progress now? I’ve techinical documentation to inductioanl furnace (it came with that furnace) marked as ???0,06/0,1-?3 (translated as inductional steel crucible, 60kg, 50kw, can melt copper alloys) with principal electrical schemes and full parts. But it all in russian and old (1984). Contact me if interested.

  24. Darren said,

    February 14, 2011 @ 7:39 am

    Hi I am an electrcial engineering student and I am required to design a small induction furnace suitable for melting materials such as bronze. The internal dimensions of the furnace are 39 cm dia and 40 cm high and the furnace is to be able to heat about 6l of molten material at a time. Any help will be greatly appreciated even if its just fowarding of relevant documents. Thanks. [email protected]

RSS feed for comments on this post · TrackBack URI

Leave a Comment

*