The hunt for 10 MHz / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

The hunt for 10 MHz

previous 1 2 3 4

Move Thread

LAN_403

Marko

Sat Mar 18 2006, 12:59AM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Only way to drive this coil at 10Mhz is E class, and I can maybe use another E class amplifier for drive.
And it is very hard to get coil working stable at 10Mhz, even royer died instantly.
Mosfets had transistent times over 100us, and I wanted to run them at 10mhz, lol...

Signal source isnt big problem, it will probably be some kind of oscillator or HC4046 IF I get some ever...

ragnar

Sat Mar 18 2006, 05:49AM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

Firkragg wrote ...

Only way to drive this coil at 10Mhz is E class,

Firkragg, with all due respect mate, I don't believe you. =-)

Somehow I don't think it would be impossible for a pair of UCCs (let alone a discrete driver) to swing 1nF to +/- 12V at 10MHz. Sure though, things aren't as simple as they first seem... but give me some time...

EEYORE

Sat Mar 18 2006, 07:04AM

Registered Member #99 Joined: Thu Feb 09 2006, 06:10PM
Location: florida, usa
Posts: 637

Has anyone given thought to just using a vacuum tube for this?Seems like it would be alot easier to get to 10mhz than with transistors(considering the application here),no?Also, perhaps winding the secondary on a ceramic form would solves the PVC melting problem.

ragnar

Sat Mar 18 2006, 07:32AM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

Sure, vacuum tubes would be a great shortcut, but for my applications thats a sacrifice of principles

I'm trying to keep it all solid state.

Plasmaniac

Sat Mar 18 2006, 08:44AM

Registered Member #206 Joined: Sat Feb 18 2006, 03:17PM
Location:
Posts: 72

I can try a tinier coil. I think the MOSFETs and their driver might make some problems. Also, I'd have to use a smaller capacitor for the PLL (I'm using 40pF right now). Maybe it'd be better if I use just one PLL and no driver, let me check the datasheet.... low/high level output current: 0.88mA... I can't. I'd have to build my own driver. BC5xx transistors work great in my little FM transmitter, so they should do the trick. I hope to find a complementary pair. I have some IRFP460 MOSFETs (I use the 2 IRFP250 MOSFETs I#ve left for the audimod. coil). I don't think they can switch 10MHz properly.

The reason, for that I use a CDs and not a HC is, that they have nearly the same specs, but the CD-version works at 3-18V. I have more voltage to drive the MOSFET drivers with if I use CDs.

EDIT: Ok, the CDs have a lower maximum frequency. I'll have to use my SMD 74LV4046. They can generate frequencys <=38 MHz (yes, 38MHz)

Marko

Sat Mar 18 2006, 10:39AM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Firkragg, with all due respect mate, I don't believe you. =-)

Somehow I don't think it would be impossible for a pair of UCCs (let alone a discrete driver) to swing 1nF to +/- 12V at 10MHz. Sure though, things aren't as simple as they first seem... but give me some time...

I meant driving coil, not gate-driving.
I cant find UCCs anywhere so I must statisfy with 7667s, but they may also show slowness and degenerated waveform at 10Mhz.
When I get my scope Il see that...

Making half/full bridge for 10Mhz is a silly idea without special, expensive mosfets and drivers.
It is 'impossible' even to find relatively cheap mosfets that have total (turnon/turnoff) transistent time lesser than 100us, and if there is no soft switching (class E) they can't do nothing.

Simple E class driver is best idea, if driver ICs show to be too slow (could be at 10mhz) they can be replaced by another tuned E class amplifier as richie brunnet did.

I don't have resources to do anything with this coil now, parts are arriving in a maybe week so it is put on hold.

Plasmaniac

Sat Mar 18 2006, 10:53AM

Registered Member #206 Joined: Sat Feb 18 2006, 03:17PM
Location:
Posts: 72

UCC wont work at frequencys around 10 MHz. I've read, that 7667 are even better for higher frequencys. I scoped the driver's output signal and it is just horrible. But fortunately, the secondary's output signal is just great

Using a 2 MOSFET pseudo fullbridge like I am using right now would erase that fullbridge issue and also saves much money. Even if both MOSFETs are on for a long time because of a driver failure they wont be blown up. A little disadvantage is the fact, that you have to use 2 primaries.

Marko

Sat Mar 18 2006, 03:10PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

HAlfbridge with voltage doubler can do the same.

I recomennd trying E class, witho one mosfet, you will reduce losses greatly and maybe get better preformance...

10Mhz seems far bigger problem than 2, 3, 5Mhz, its edge where most of our mosfets can switch
(ones with low ON resistance and transistent times < 100ns are very rare).

So vacuum tubes may win again

but as BP said it has no point then...

vasil

Sat Mar 18 2006, 04:44PM

Registered Member #229 Joined: Tue Feb 21 2006, 07:33PM
Location: Romania
Posts: 506

Did you read the "Smallest SSTC in the world" thread from the old forum?

http://old.4hv.org/index.php?board=7;action=display;threadid=1129

Jim Mitchell won the contest with a small SSTC built with a FDH15N50 transistor and an IXDD414CI driver, driven up to 8.89 MHz.

http://forum.4hv.org/attachments/DSC01618.JPG

So, the 10 MHz dream is doable....

previous 1 2 3 4

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob