Uploaded October 2024 | Updated September 2026, 2 weeks ago
Added another CT to give me feedback again so now this is running with PLL feedback, has the LC OCD, and also has the line input OCP.
Barely melting a small aluminum heatsink here at about the 450-500W RMS region. It took about 10 minutes of running at that wattage before the fat crucible and small heatsink started to turn red hot. Another 5 minutes or so before the aluminum started to sink down on its own. I'm guessing it would take at least a couple more hundred watts of power for that to turn into a molten pool over some course of a few more minutes.
I can imagine based off that, some input closer to 1kW would easily melt larger amounts of aluminum for me without needing tons of supply amps.
I still have a test rig here with caps I don't want to push too hard and some 10A switches but otherwise everything seemed to run great with this setup. I'm glad I added the current trip because my tank OCP is not really tuned yet and there are times the input current will start shooting up pretty quickly. I haven't tried yet but I imagine I could detune the PLL quite a bit and be able to crank the input voltage a lot higher. It's nice to be able to somewhat melt aluminum at only about 10A but I bet I could get that 10A at closer to 100VDC input with detuning and actually melt it.
When I add the legit resonant cap and retune everything I'll need to then play with the windings on the coupling transformer. It looks like I'd be hitting close to 20A+ at only 120VAC. That's the only way I'd ever push over 2kW on 120 at home but I'll try to limit to around 15A for 1800W since that seems like it would melt a lot of things no problem. I don't really have a need for melting steel aside from being tempted to make evil security bolts. Luckily I would like to melt some copper and that should be easier than steel.
I saw the crucible on scamazon for like 10 bucks so I grabbed it, but it has a strange shape which tapers in toward the bottom and it barely jams into the coil with the ceramic wool around it. I went ahead and got another one which is perfectly cylindrical all the way through and isn't quite as a big. That should make melting aluminum a lot easier. When I get my resonant cap in I still actually won't be able to run it cause I don't want to add like 2 more turns. I just need to dbl check all the dimensions etc then figure out how to stack 2 in parallel for 4uF. These random bricks works but it's already sort of an undertaking building this thing and I'd like it to be able to easily do a couple hundred khz. Not really sure how I'll pull that off without stacking a bunch of 247's but I need to make it all work well enough to cram into this old computer case I'm going to use for the enclosure.
Added another CT to give me feedback again so now this is running with PLL feedback, has the LC OCD, and also has the line input OCP.
Barely melting a small aluminum heatsink here at about the 450-500W RMS region. It took about 10 minutes of running at that wattage before the fat crucible and small heatsink started to turn red hot. Another 5 minutes or so before the aluminum started to sink down on its own. I'm guessing it would take at least a couple more hundred watts of power for that to turn into a molten pool over some course of a few more minutes.
I can imagine based off that, some input closer to 1kW would easily melt larger amounts of aluminum for me without needing tons of supply amps.
I still have a test rig here with caps I don't want to push too hard and some 10A switches but otherwise everything seemed to run great with this setup. I'm glad I added the current trip because my tank OCP is not really tuned yet and there are times the input current will start shooting up pretty quickly. I haven't tried yet but I imagine I could detune the PLL quite a bit and be able to crank the input voltage a lot higher. It's nice to be able to somewhat melt aluminum at only about 10A but I bet I could get that 10A at closer to 100VDC input with detuning and actually melt it.
When I add the legit resonant cap and retune everything I'll need to then play with the windings on the coupling transformer. It looks like I'd be hitting close to 20A+ at only 120VAC. That's the only way I'd ever push over 2kW on 120 at home but I'll try to limit to around 15A for 1800W since that seems like it would melt a lot of things no problem. I don't really have a need for melting steel aside from being tempted to make evil security bolts. Luckily I would like to melt some copper and that should be easier than steel.
I saw the crucible on scamazon for like 10 bucks so I grabbed it, but it has a strange shape which tapers in toward the bottom and it barely jams into the coil with the ceramic wool around it. I went ahead and got another one which is perfectly cylindrical all the way through and isn't quite as a big. That should make melting aluminum a lot easier. When I get my resonant cap in I still actually won't be able to run it cause I don't want to add like 2 more turns. I just need to dbl check all the dimensions etc then figure out how to stack 2 in parallel for 4uF. These random bricks works but it's already sort of an undertaking building this thing and I'd like it to be able to easily do a couple hundred khz. Not really sure how I'll pull that off without stacking a bunch of 247's but I need to make it all work well enough to cram into this old computer case I'm going to use for the enclosure.
