Building a 5,000W Remote Controlled Arc Furnace to Melt Steel and Ceramics
Hyperspace Pirate 25:29
254,010 views · 13,436 likes Watch on YouTube ↗
Enjoy FlexiSpot's Black Friday sale - up to 80% off! lowest prices of the year and with a chance to get your order for free! Use my code ''YTE7P50'' to get extra $50 off on the E7 Plus standing desk and ''WP30OFF'' for $30 off on treadmill.
FlexiSpot E7 Plus standing desk:
USA: https://bit.ly/4hIo07a
CAN: https://bit.ly/47ImQUR
Auto Incline Walking Treadmill:
https://bit.ly/4nEmRPl
In this video I'll go through the process of building a 5-kW electric arc furnace that can get hot enough to melt steel and ceramic materials, and do other miscellaneous high temperature chemistry like making rubies or generating Calcium Carbide from Calcium Oxide and Carbon. The whole system is remote controlled for safety and more precise adjustment without needing constant attention from an operator.
The power supply is built from the transformer out of a 5kW 220V/110V converter with a new secondary winding consisting of 36 turns of 10 strands of 12-gauge magnet wire in parallel. The open circuit voltage is 27V AC, but under heavy load it drops to about 21V AC. Peak current was measured at ~325A so the RMS value is approximately 230A. Power comes from a 220V outlet on a 50-A breaker (so over 10kW is possible). Even without conductive or infrared heating, the high current gets the connections hot enough to melt the 3d-printed brackets that the electrodes are mounted on, so a large 12V blower is used to force air over them.
The electrodes are 1/2" diameter carbon gouging rods. Both electrodes move together in the Z-axis with a 200-RPM gearmotor driving a T8 leadscrew (same as a typical 3d printer) for direct heating (where the melt is part of the electric circuit). For non-conductive materials, indirect heating is used, and the electrodes are turned at a slight angle, with one electrode being movable on an X-axis using a 30-RPM gearmotor driving a T8 leadscrew.
The controller is a wired remote control that simply uses momentary-contact DPDT toggle switches to drive the servo motors forward/backward. However, I found that 200 RPM on a T8 leadscrew made the Z-axis turn too fast for precise control, so it was easy to accidentally plunge the electrodes too far into the material and cause a short circuit that produced very little heating, or pull the electrodes too far out and lose contact altogether. For better vertical control, it would make more sense to have a lower-RPM motor, or a PWM controller to limit the motor speed, since fine-tuning the position of the electrode is critical for striking and maintaining a good arc.
If this were to be use exclusively for metal melting, a better configuration would be to have the graphite crucible serve as one of the electrodes, then only one carbon rod would need to be used, and it'd only need to be moved on the Z-axis. Another improvement (which i'll add in the future) would be a high-voltage pilot arc like a TIG welder's high frequency starter to initiate the arc without touching the material to be melted.
Overall it's a really useful device to have, since it allows me to melt any metal without having to get fuel, but it needs a lot of insulation around the crucible to really work well, and I think connecting it to a DC stick welder power supply might allow better efficiency since the heat is distributed unevenly with DC between the target/electrode, so the polarity could be configured to have really good performance.
Music Used:
Kevin MacLeod - Lobby Time
FlexiSpot E7 Plus standing desk:
USA: https://bit.ly/4hIo07a
CAN: https://bit.ly/47ImQUR
Auto Incline Walking Treadmill:
https://bit.ly/4nEmRPl
In this video I'll go through the process of building a 5-kW electric arc furnace that can get hot enough to melt steel and ceramic materials, and do other miscellaneous high temperature chemistry like making rubies or generating Calcium Carbide from Calcium Oxide and Carbon. The whole system is remote controlled for safety and more precise adjustment without needing constant attention from an operator.
The power supply is built from the transformer out of a 5kW 220V/110V converter with a new secondary winding consisting of 36 turns of 10 strands of 12-gauge magnet wire in parallel. The open circuit voltage is 27V AC, but under heavy load it drops to about 21V AC. Peak current was measured at ~325A so the RMS value is approximately 230A. Power comes from a 220V outlet on a 50-A breaker (so over 10kW is possible). Even without conductive or infrared heating, the high current gets the connections hot enough to melt the 3d-printed brackets that the electrodes are mounted on, so a large 12V blower is used to force air over them.
The electrodes are 1/2" diameter carbon gouging rods. Both electrodes move together in the Z-axis with a 200-RPM gearmotor driving a T8 leadscrew (same as a typical 3d printer) for direct heating (where the melt is part of the electric circuit). For non-conductive materials, indirect heating is used, and the electrodes are turned at a slight angle, with one electrode being movable on an X-axis using a 30-RPM gearmotor driving a T8 leadscrew.
The controller is a wired remote control that simply uses momentary-contact DPDT toggle switches to drive the servo motors forward/backward. However, I found that 200 RPM on a T8 leadscrew made the Z-axis turn too fast for precise control, so it was easy to accidentally plunge the electrodes too far into the material and cause a short circuit that produced very little heating, or pull the electrodes too far out and lose contact altogether. For better vertical control, it would make more sense to have a lower-RPM motor, or a PWM controller to limit the motor speed, since fine-tuning the position of the electrode is critical for striking and maintaining a good arc.
If this were to be use exclusively for metal melting, a better configuration would be to have the graphite crucible serve as one of the electrodes, then only one carbon rod would need to be used, and it'd only need to be moved on the Z-axis. Another improvement (which i'll add in the future) would be a high-voltage pilot arc like a TIG welder's high frequency starter to initiate the arc without touching the material to be melted.
Overall it's a really useful device to have, since it allows me to melt any metal without having to get fuel, but it needs a lot of insulation around the crucible to really work well, and I think connecting it to a DC stick welder power supply might allow better efficiency since the heat is distributed unevenly with DC between the target/electrode, so the polarity could be configured to have really good performance.
Music Used:
Kevin MacLeod - Lobby Time
Category (YouTube): People & Blogs
Playback is via YouTube's official embedded player. Data from YouTube; Exumo is not affiliated with YouTube.