Old-school? BJT SSTC
brtaman, Tue Jul 14 2009, 11:02AMHey,
I have to start off by saying that since joining the forum I never found SSTCs to be all to interesting, I was planning for my next project (big this time) to be a VTTC or a MOT SGTC, but I am a changed man. :)
I had some spare time on Sunday and decided I would build my first SSTC (small and simple), with either a royer or armstrong oscillator. I ran into my main hurdle when I realized I did not have any suitable FETs for the job, so I used a pair of "ultra-fast" BJT transistors from an old switching PSU, I really wasn't expecting much given the initial transistor handicap, but I was quite surprised after some tuning (Tuning a Royer for SSTC is a Royal PIA) when I got a nice ~1cm brush from my breakout point.
Video:
Specs:
Primary: 3+3 Feedback 2
Secondary: d:2.6cm h:6cm (tiny)
Driver: Royer Oscillator using 2x E13009L (Ton=1.1ns Tstg = 3ns Tf = 0.7ns..ultra fast :I)
Tank cap is around 14nF
(Input is 12v -.- draws around 3-4 amps)
(Schematic, is a buck standard Royer oscillator like the one on HV wiki, only I am using 235ohm as the resistor)
My question to you SSTC experts is what could I do to this SSTC to improve performance besides uping input voltage? The transistors get quite hot, much hotter than on my ZVS driver, but still within operating specs, anything I can do to help this? Some Zeners and diodes? I think I can squeeze a bit more out of the system by playing with the tank cap, but not much...I really need to get myself a decent +24v PSU, any suggestions? On a side note I think I am really going to have to invest in an o-scope, if I want to continue in the SSTC direction :(.
Thanks,
brtaman
Re: Old-school? BJT SSTC
Sulaiman, Tue Jul 14 2009, 06:55PM
Try 2 + 2 turns as the primary with about 32 nF
and have the wires to the capacitor(s) as short as possible
it's possible that the feedback winding is drawing enormous power from the primary
and dumping it in the transistors as massive base current,
try adding some resistance in series with the bases (e.g. 10 Ohms)
and/or reducing the feedback to one turn.
If you run the primary without the secondary
the current/power drawn should be very low,
if not then modify the primary circuit.
Check the temperature of transistors, capacitors and primary,
all should run warm only with no secondary.
Sulaiman, Tue Jul 14 2009, 06:55PM
Try 2 + 2 turns as the primary with about 32 nF
and have the wires to the capacitor(s) as short as possible
it's possible that the feedback winding is drawing enormous power from the primary
and dumping it in the transistors as massive base current,
try adding some resistance in series with the bases (e.g. 10 Ohms)
and/or reducing the feedback to one turn.
If you run the primary without the secondary
the current/power drawn should be very low,
if not then modify the primary circuit.
Check the temperature of transistors, capacitors and primary,
all should run warm only with no secondary.
Re: Old-school? BJT SSTC
brtaman, Wed Jul 15 2009, 10:07AM
Thanks for the tips Sulaiman. Will try them as soon as I can find time, 2+2 should definitely give some improvement. I was hesitant to go below 3+3. Otherwise, the feedback does suck up an enormous amount of extra power for every extra 1/8 of a turn, I am right on the balance of reliable start-up/power ratio atm, though it should change as I go to 2+2.
Thanks
brtaman
brtaman, Wed Jul 15 2009, 10:07AM
Thanks for the tips Sulaiman. Will try them as soon as I can find time, 2+2 should definitely give some improvement. I was hesitant to go below 3+3. Otherwise, the feedback does suck up an enormous amount of extra power for every extra 1/8 of a turn, I am right on the balance of reliable start-up/power ratio atm, though it should change as I go to 2+2.
Thanks
brtaman
Re: Old-school? BJT SSTC
Steve Conner, Wed Jul 15 2009, 12:18PM
I think you'll find that the times you quoted are microseconds, not nanoseconds, unless you got your transistors from a dumpster in Area 51 or something. And they'll be MJE13009 from Motorola/ONSemi.
Steve Conner, Wed Jul 15 2009, 12:18PM
I think you'll find that the times you quoted are microseconds, not nanoseconds, unless you got your transistors from a dumpster in Area 51 or something. And they'll be MJE13009 from Motorola/ONSemi.
Re: Old-school? BJT SSTC
brtaman, Wed Jul 15 2009, 01:58PM
Whoops :D
Yes, Steve, you are correct they are in microseconds I guess it was just wishful thinking on my part, though you bring up an interesting idea...dumpster diving at Area 51, anyone up for it? Though the assumption that they actually throw anything out before incineration is perhaps even more out there than nano-second BJTs. ;D
It is the KSE13009 transistor though by Fairchild, don't know if thats better or worse, but it is so damn hot today that I will wait until night to compare datasheets. Do you have any suggestions on a really fast BJT transistor for the coil? I am really going to see how far I can get without going FET/IGBT (that's reserved for the larger SSTC, which will most likely be based of your excellent schematic).
Thanks
brtaman
brtaman, Wed Jul 15 2009, 01:58PM
Whoops :D
Yes, Steve, you are correct they are in microseconds I guess it was just wishful thinking on my part, though you bring up an interesting idea...dumpster diving at Area 51, anyone up for it? Though the assumption that they actually throw anything out before incineration is perhaps even more out there than nano-second BJTs. ;D
It is the KSE13009 transistor though by Fairchild, don't know if thats better or worse, but it is so damn hot today that I will wait until night to compare datasheets. Do you have any suggestions on a really fast BJT transistor for the coil? I am really going to see how far I can get without going FET/IGBT (that's reserved for the larger SSTC, which will most likely be based of your excellent schematic).
Thanks
brtaman
Re: Old-school? BJT SSTC
Firefox, Wed Jul 15 2009, 07:14PM
I've found that varying the resistance from the feedback winding to the +12V can increase or decrease the current drawn by the Royer circuit. You may want to try varying this to see what happens.
Firefox, Wed Jul 15 2009, 07:14PM
I've found that varying the resistance from the feedback winding to the +12V can increase or decrease the current drawn by the Royer circuit. You may want to try varying this to see what happens.
Re: Old-school? BJT SSTC
brtaman, Sat Jul 25 2009, 04:03PM
A small/big update on this little proof of concept coil. (Firefox thanks for the suggestion, tried a bunch of settings seems like my original 235ohm works the best for this coil?)
Well this is officially one of the weirdest SSTC coils I have seen looking through the
I will start with a list of what has changed since my opening post:
-Input voltage raised to 24v (from 12v)
-Capacitance of LC was raised to ~19nF
-Much larger heatsink was employed to keep heating down (still gets hot but within reason)
And the kicker:
-While looking for a good RF ground plate I came across my old secondary (beaten to h*ll and back) and the experimenter inside me shrieked connect its ground to the active secondaries ground. After a bit of tuning the results were quite surprising (quite good performance), I have read a bit about magnifier coils (not much, so I am in the dark atm), is it working like a magnifier coil to some extent?
A picture:
The video (some very hot RF arcs :P):
What I am really looking for in this coil, which is basically an intro into SSTC (I like to start small and work my way up learning on the way). What I am really looking to achieve is small "bolts" as opposed to brushed output. Could I achieve this with a simple 555 interrupter? If so where would be the best place to insert it?
Help with the interrupter would be greatly appreciated, I have seen many schematics for some really great interrupters however this simple little coil does not require such an extravagant interrupter IMO? I was thinking something along the lines of a simple 555 or TL494 (single chip) based interrupter, would something like that do the trick? I am trying to learn as much as I can on my own, but advice from the more experienced members would really help in this situation.
The arcs it throws off are rather hot (yellow, doesn't show in video), so I think some nice sparks can be achieved. (Would probably burn my finger to the bone -.-) I am still not finished tuning but could use a bit of input into the exact mechanics of the current set-up with insight for the interrupter.
I am going to keep this TC BJT powered, but these transistors aren't exactly top of the line, does anyone have a suggestion for some ULTRA fast transistors with similair specs? Perhaps with a voltage rating above 1000v so I could run it off-line? (220v 50hz here in Slovenia) I am in the process of making a huge order for parts and I would just throw them in to save on shipping..
Thanks
brtaman
brtaman, Sat Jul 25 2009, 04:03PM
A small/big update on this little proof of concept coil. (Firefox thanks for the suggestion, tried a bunch of settings seems like my original 235ohm works the best for this coil?)
Well this is officially one of the weirdest SSTC coils I have seen looking through the
I will start with a list of what has changed since my opening post:
-Input voltage raised to 24v (from 12v)
-Capacitance of LC was raised to ~19nF
-Much larger heatsink was employed to keep heating down (still gets hot but within reason)
And the kicker:
-While looking for a good RF ground plate I came across my old secondary (beaten to h*ll and back) and the experimenter inside me shrieked connect its ground to the active secondaries ground. After a bit of tuning the results were quite surprising (quite good performance), I have read a bit about magnifier coils (not much, so I am in the dark atm), is it working like a magnifier coil to some extent?
A picture:
The video (some very hot RF arcs :P):
What I am really looking for in this coil, which is basically an intro into SSTC (I like to start small and work my way up learning on the way). What I am really looking to achieve is small "bolts" as opposed to brushed output. Could I achieve this with a simple 555 interrupter? If so where would be the best place to insert it?
Help with the interrupter would be greatly appreciated, I have seen many schematics for some really great interrupters however this simple little coil does not require such an extravagant interrupter IMO? I was thinking something along the lines of a simple 555 or TL494 (single chip) based interrupter, would something like that do the trick? I am trying to learn as much as I can on my own, but advice from the more experienced members would really help in this situation.
The arcs it throws off are rather hot (yellow, doesn't show in video), so I think some nice sparks can be achieved. (Would probably burn my finger to the bone -.-) I am still not finished tuning but could use a bit of input into the exact mechanics of the current set-up with insight for the interrupter.
I am going to keep this TC BJT powered, but these transistors aren't exactly top of the line, does anyone have a suggestion for some ULTRA fast transistors with similair specs? Perhaps with a voltage rating above 1000v so I could run it off-line? (220v 50hz here in Slovenia) I am in the process of making a huge order for parts and I would just throw them in to save on shipping..
Thanks
brtaman
Print this page