Mjolnir DRSSTC reloaded.
Steve Conner, Tue Aug 21 2007, 10:56PMHi all,
My Mjolnir DRSSTC was not performing too well, and people keep asking me to do demos with it. So I decided to try and fix it up.
The main problems were that the "Hivotronics Rapid Discharge Capacitor" was overheating badly and leaking unpleasant waxy goop from its terminals, and the bottom 2 inches of the secondary winding were missing completely, destroyed by flashovers.
Due to the lack of coupling caused by the missing windings, I had to tune the primary current and burst length right up to 400Apk and 300us, and even so, the sparks weren't as impressive as they used to be. And the tank capacitor just got hotter and leaked more goo, and started making crackling noises, so something had to be done.
I started by making a new MMC bank using two strings of 10 0.47uF 1000V yellow polypropylene caps from RS. This made 0.094uF @ 10kV, which is twice the original capacitance. I had to pack them a little tighter than I'd have liked, to make it fit into the space occupied by the original cap, so flashovers were something of a worry.
Next I had to deal with the missing secondary windings. If I just wound them back with the same gauge of wire, I'd have the same flashover problem again. So I decided to wind them with thicker wire, hoping that it would give enough extra coupling to get things going, but not so much that it flashed over. I used 1.2mm PVC insulated wire, compared to the 0.315mm that the rest of the resonator is made of. I don't know if anyone has tried this before: I guess it's like a magnifier but with the secondary and tertiary both on the same former.
The results were pretty good, as you can see from the pictures below Mjolnir is back to its old fiercely sparking form. The sparks shown only needed a burst length of 150us. The new MMCs didn't even seem to get warm, although I only did short runs.
Re: Mjolnir DRSSTC reloaded.
Marko, Tue Aug 21 2007, 11:59PM
Hey, your coil is insane Steve... I mean, almost every part of construction is absolutely perfect. Your circuit never blew and IGBT in life. I'll never be cool like you.
Pity of the Hivotronics cap, it must have been expensive.
Have you thought of redesigning your primary to inverse cone? Lots of people are happy with helical primary, but I see you are using really small tank capacitor and relatively big primary.
Everyone considers high impedance tank bad, now you are pumping large sparks with 50-100nF capacitance.
Marko, Tue Aug 21 2007, 11:59PM
Hey, your coil is insane Steve... I mean, almost every part of construction is absolutely perfect. Your circuit never blew and IGBT in life. I'll never be cool like you.
Pity of the Hivotronics cap, it must have been expensive.
Have you thought of redesigning your primary to inverse cone? Lots of people are happy with helical primary, but I see you are using really small tank capacitor and relatively big primary.
Everyone considers high impedance tank bad, now you are pumping large sparks with 50-100nF capacitance.
Re: Mjolnir DRSSTC reloaded.
Bennem, Wed Aug 22 2007, 07:21AM
Great work Steve!
Its good to know that Mjolnir is still out performing many a coil
for its dimensions!
I was lucky enough to see Mojolnir at a 2005 Derby Teslathon
and took loads of video footage.
I'm still in awe with its perfomance....
Bennem, Wed Aug 22 2007, 07:21AM
Great work Steve!
Its good to know that Mjolnir is still out performing many a coil
for its dimensions!
I was lucky enough to see Mojolnir at a 2005 Derby Teslathon
and took loads of video footage.
I'm still in awe with its perfomance....
Re: Mjolnir DRSSTC reloaded.
thedatastream, Wed Aug 22 2007, 09:00AM
omg dood that is like so cool i used your circuit but i keep killing igbtz should i just give up and become a hermit help pleez?
* thedatastream is still feeling funny after giving blood yesterday
thedatastream, Wed Aug 22 2007, 09:00AM
omg dood that is like so cool i used your circuit but i keep killing igbtz should i just give up and become a hermit help pleez?
* thedatastream is still feeling funny after giving blood yesterday
Re: Mjolnir DRSSTC reloaded.
Steve Conner, Wed Aug 22 2007, 09:32AM
Hi all,
For all I know, it could just be sheer luck that it hasn't blown up yet.
About the capacitors, I bought a bunch of them second hand from another UK coiler, and I think I know now why they were cheap -__-
Tank circuit impedance, I was never too sure about. I'm running a full bridge off a 320V DC bus, so I won't need as low an impedance as someone who runs a halfbridge off 160V. Also, only 5 turns of the primary are in use now. But I think I've always used impedances that are fairly high compared to what everyone else does. Steve Ward probably knows more about this than I do.
The inverse cone primary would have been better than my dual-gauge secondary bodge, but I couldn't be bothered making a whole new primary. If I trash the secondary again though, I will rewind it completely and make a new primary.
thedatastream, my circuit is kind of experimental and I don't know whether it works for all types of coil. I can't guess what's wrong without knowing a lot more about your setup, but I wouldn't throw away a promising career in engineering just because it doesn't work :P If you post more details on your project thread, we could maybe help you debug it.
Steve Conner, Wed Aug 22 2007, 09:32AM
Hi all,
For all I know, it could just be sheer luck that it hasn't blown up yet.
About the capacitors, I bought a bunch of them second hand from another UK coiler, and I think I know now why they were cheap -__-
Tank circuit impedance, I was never too sure about. I'm running a full bridge off a 320V DC bus, so I won't need as low an impedance as someone who runs a halfbridge off 160V. Also, only 5 turns of the primary are in use now. But I think I've always used impedances that are fairly high compared to what everyone else does. Steve Ward probably knows more about this than I do.
The inverse cone primary would have been better than my dual-gauge secondary bodge, but I couldn't be bothered making a whole new primary. If I trash the secondary again though, I will rewind it completely and make a new primary.
thedatastream, my circuit is kind of experimental and I don't know whether it works for all types of coil. I can't guess what's wrong without knowing a lot more about your setup, but I wouldn't throw away a promising career in engineering just because it doesn't work :P If you post more details on your project thread, we could maybe help you debug it.
Re: Mjolnir DRSSTC reloaded.
Marko, Wed Aug 22 2007, 01:47PM
Hi Steve
I don't think conical primary is a large problem to build.. all you need is some Plexiglas and a meter of nichrome wire..
Still other guys just raise their secondary up and still make enormous sparks.
(...)
Where do you get those beautiful stubby rifa caps? Those at RS we have here cost like hundreds of $, a decent cap would be like whole coil budget for me..
I ended buying 2 biggest (1000uF 400V) solder terminal caps and they appear to suck.
And the large anodised heatsink, it's so beautiful, where can one get new heatsinks like that today?
Marko, Wed Aug 22 2007, 01:47PM
Hi Steve
I don't think conical primary is a large problem to build.. all you need is some Plexiglas and a meter of nichrome wire..
Still other guys just raise their secondary up and still make enormous sparks.
(...)
Where do you get those beautiful stubby rifa caps? Those at RS we have here cost like hundreds of $, a decent cap would be like whole coil budget for me..
I ended buying 2 biggest (1000uF 400V) solder terminal caps and they appear to suck.
And the large anodised heatsink, it's so beautiful, where can one get new heatsinks like that today?
Re: Mjolnir DRSSTC reloaded.
Steve Conner, Wed Aug 22 2007, 05:03PM
Hi Marko
I found, with my driver circuit at least, that the tighter the coupling between primary and secondary, the better everything works. I get longer sparks for the same primary current as I increase coupling. Also, if the coupling reaches the critical value, it may confuse the PLL as it can't make up its mind whether the system has two resonant frequencies or one. I had trouble tuning with the looser coupling, and that may have been why.
I never thought of trying to cut primary supports with a hot wire, though now I know someone with a CO2 laser cutter, I'd probably get them done on that. (And resist the urge to embed lots of blue LEDs into it...)
The stubby Rifa capacitors were part of a big surplus load from a bankrupt company that manufactured high frequency spot welders. The large Rifa caps are expensive from RS, and they don't seem to sell the stubby kind at all.
The big heatsink was bought from RS, it's part # 658-025.
My next project is fibre optic isolation for the remote controller...
Steve Conner, Wed Aug 22 2007, 05:03PM
Hi Marko
I found, with my driver circuit at least, that the tighter the coupling between primary and secondary, the better everything works. I get longer sparks for the same primary current as I increase coupling. Also, if the coupling reaches the critical value, it may confuse the PLL as it can't make up its mind whether the system has two resonant frequencies or one. I had trouble tuning with the looser coupling, and that may have been why.
I never thought of trying to cut primary supports with a hot wire, though now I know someone with a CO2 laser cutter, I'd probably get them done on that. (And resist the urge to embed lots of blue LEDs into it...)
The stubby Rifa capacitors were part of a big surplus load from a bankrupt company that manufactured high frequency spot welders. The large Rifa caps are expensive from RS, and they don't seem to sell the stubby kind at all.
The big heatsink was bought from RS, it's part # 658-025.
My next project is fibre optic isolation for the remote controller...
Re: Mjolnir DRSSTC reloaded.
Marko, Wed Aug 22 2007, 05:35PM
I usually bend the supports into L and simply screwing to case.. I love that method.
Blue LED's are in order as well.. ;)
$98.30 OMG
Marko, Wed Aug 22 2007, 05:35PM
never thought of trying to cut primary supports with a hot wire, though now I know someone with a CO2 laser cutter, I'd probably get them done on that. (And resist the urge to embed lots of blue LEDs into it...)
I usually bend the supports into L and simply screwing to case.. I love that method.
Blue LED's are in order as well.. ;)
The big heatsink was bought from RS, it's part # 658-025.
$98.30 OMG
Re: Mjolnir DRSSTC reloaded.
Bennem, Thu Aug 23 2007, 08:30AM
Marko......Rifa caps are sometimes available on
ebay
i got these two for £10 ..and used one of them in my last
DRSSTC....they're good caps!
steve....what spark length are you getting now with your new MMC?
Bennem, Thu Aug 23 2007, 08:30AM
Marko......Rifa caps are sometimes available on
ebay
i got these two for £10 ..and used one of them in my last
DRSSTC....they're good caps!
steve....what spark length are you getting now with your new MMC?
Print this page