It has to do with several things about mosfets. Some have really low "on" resistance, like a few hundredths of an ohm, and some have a lot more, like half or even a full ohm. This makes a big difference in power dissipation in the mosfet, and also in the current in the load. Also some are more sensitive to gate overvoltages and gate-source voltage difference.
It always seems that the ones I need are the most expensive ones. But this thing works OK with the 740s, and even with 640s if careful.
It's amazing. I showed it to the local TC expert and he asked where the caps were. The whole resonator is only 44 cm tall including the toroid, and if I want to waste mosfets I can get 20 cm gouts -- veritable _fountains_ of yellow-white flame.
It only likes certain mosfets, though. It works well with IRF 640s (for a few seconds), IRF740 (stable if heat is watched), IRFP252 (best overall results) and it does NOT work (without mods) with IRF830, IRFP250 (real surprise), IRFP26x, IRFP450.
You will note that I am no longer fooling around with sticking hand-held stuff into the arc.
The RF burn on my left index finger is basically a pea-sized sphere of cooked meat that needs to heal from beneath and will eventually make a divot-like scar. It will need to be protected from infection for weeks while the body figures out where the live stuff stops and the dead stuff starts, and gets it walled off. Until then it oozes and is painful, also ugly to look at.
Yep, the basic gap is 120 mm straight-line from the point of the ground wire to the toroid, and convection takes the flame and stretches it quite a bit in the draft. Six inches easy, for a short time.
But I don't think it would strike a power arc across six inches between honest smooth spheres.
Yet...
If it threatens to do anything more than this, I'll have to move it to the living room floor. Then things get serious, and I'll need a spotter, in case of pooch-screwing.
It's a lot louder live. Sounds like a .22 going off. At least this one stayed in one piece.
Seven Dollars a Pop!
It took out the Lowside B corner mosfet, and the two Zener gate voltage regulators for that transistor.
I tried higher-rated IRFP264s but they did not work well, I would have to modify the circuit, so I pulled them and tried IRF830s, similar result. IRF740s work well but are unforgiving: blew 2 more of them and another two pair Zeners before I realized that "Stop" means STOP.
Nice hot arcs. Something that I've noticed is different part number mosfets pop like crazy and others don't pop at all haven't figured it out why :P
plumcrazzzy 2 years ago
It has to do with several things about mosfets. Some have really low "on" resistance, like a few hundredths of an ohm, and some have a lot more, like half or even a full ohm. This makes a big difference in power dissipation in the mosfet, and also in the current in the load. Also some are more sensitive to gate overvoltages and gate-source voltage difference.
It always seems that the ones I need are the most expensive ones. But this thing works OK with the 740s, and even with 640s if careful.
TinselKoala 2 years ago
Holy Crap! Looks like something you would get off a bank of MOT's and caps. That is a cool output.
jupy921 2 years ago
It's amazing. I showed it to the local TC expert and he asked where the caps were. The whole resonator is only 44 cm tall including the toroid, and if I want to waste mosfets I can get 20 cm gouts -- veritable _fountains_ of yellow-white flame.
It only likes certain mosfets, though. It works well with IRF 640s (for a few seconds), IRF740 (stable if heat is watched), IRFP252 (best overall results) and it does NOT work (without mods) with IRF830, IRFP250 (real surprise), IRFP26x, IRFP450.
TinselKoala 2 years ago
You will note that I am no longer fooling around with sticking hand-held stuff into the arc.
The RF burn on my left index finger is basically a pea-sized sphere of cooked meat that needs to heal from beneath and will eventually make a divot-like scar. It will need to be protected from infection for weeks while the body figures out where the live stuff stops and the dead stuff starts, and gets it walled off. Until then it oozes and is painful, also ugly to look at.
Be careful, please.
TinselKoala 2 years ago
wow !!!! thats flame must be six inches long !!!
outstanding.
harpbloke 2 years ago
Yep, the basic gap is 120 mm straight-line from the point of the ground wire to the toroid, and convection takes the flame and stretches it quite a bit in the draft. Six inches easy, for a short time.
But I don't think it would strike a power arc across six inches between honest smooth spheres.
Yet...
If it threatens to do anything more than this, I'll have to move it to the living room floor. Then things get serious, and I'll need a spotter, in case of pooch-screwing.
TinselKoala 2 years ago
Never seen that happen before.
janne808 2 years ago
Which, the weird power arc or the exploding MOSFET?
I blow parts every day; I don't feel like I've accomplished anything unless there are a string of smoking corpses along the trail...
That upward-leaping power arc coming out of the lightning storm is rather attention-grabbing, isn't it?
TinselKoala 2 years ago
The weird arc. Exploding semiconductors I've seen a plenty.
Smoking corpses usually equates to a ticket to Stockholm in this context so you're on the right track.
janne808 2 years ago
I heard that baby pop!
onthecuttingedge2005 2 years ago
It's a lot louder live. Sounds like a .22 going off. At least this one stayed in one piece.
Seven Dollars a Pop!
It took out the Lowside B corner mosfet, and the two Zener gate voltage regulators for that transistor.
I tried higher-rated IRFP264s but they did not work well, I would have to modify the circuit, so I pulled them and tried IRF830s, similar result. IRF740s work well but are unforgiving: blew 2 more of them and another two pair Zeners before I realized that "Stop" means STOP.
TinselKoala 2 years ago