cold-cathode TC / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

cold-cathode TC

1 2 next

Move Thread

LAN_403

HVPaul

Tue Jan 26 2010, 06:52PM

Registered Member #2321 Joined: Fri Aug 28 2009, 05:13PM
Location: Boston, MA
Posts: 34

Hi all,

I'm looking into possibility of building a desktop TC based on a Cold-cathode power supply unit.
Here's the plan i have so far:

Judging by the unit i have, the output is spilit into two 600 volt connectors giving it a potential of 1.2kV if wired serially.

The 1.2kV output will be fed into a 6 stage multiplier for output to the TC. Multiplier will be made out of IN4007 diodes.

I've yet to make calculations for the MMC and i will most likely be using a stationary spark-gap.
The secondary will be about 6 inches tall by 1 inches wide wound with 36 - 40 AWG wire.

This is inspired by similar projects involving flash transformers but I had trouble finding anything using the CFL power supplies. It seems intuitive to use off-shelf parts for this scale of things, and certainly makes mistakes less lethal as this is my first SGTC attempt :)

I'd like to hear your input on approaching this project.

Arcstarter

Tue Jan 26 2010, 07:13PM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Yay, a typical current fed royer supply :).

You will have to bypass those two blue caps on the output side. They are used to resonate to kick the voltage up to start a lamp up, and then they limit the current. Also, due to the high freq, you cannot use 1n4007 diodes for the multiplier... I would go with UF4007 or some other ultrafast diode. Also, i would suggest only use maybe 2 stages. At this low of current, corona loss at higher voltages could be significant.

This project should be a cool one :). This power supply can push more power than a camera's power supply, and it'd be more efficient, too. Especially since you are starting out with 1.2kv instead of 330 or so volts.

HVPaul

Tue Jan 26 2010, 07:44PM

Registered Member #2321 Joined: Fri Aug 28 2009, 05:13PM
Location: Boston, MA
Posts: 34

Do I need to do any current-limiting considering the size?
I certainly have a lesser margin for error here.

The specs of the supply state 12V at 350mA ~4.2 watts.
Would special considerations need to be taken for the primary winding size and the MMC bank?

radiotech

Tue Jan 26 2010, 07:45PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

I have a few questions. What is the make of that balast in the photo
and do you think a small sealed TR tube, that used be part of old radars would do for the arc gap? They are abut 2 inches long with a pin at either end and a gap in the middle. Somewhere I have a whole bunch of them.

As for a 6 inch coil, in a tester here (aerial tester) there is a 8 inch air core flyback, but the secondary is made up of wound pi's about 8 of them spaced half inch. The primary is heavywound at the bottom.the bobbin is 1.5 inch bakelite.

It uses a tube, 6Y6 or 6K6 and the thing drives 3 -1B3s whose heater are powered by single loops in the field at the bottom. The tripler uses doorknobs and puts out about 19000 volts DC. The kit comes with a telescoping probe, about 16 feet long, with a hook on the end. The idea is to hook it onto an aerial, probably on a ship,
and if it arc over, a sensitive relay in the oscillator circuit closed and rings a bell inside the case. The secondary seems to be cotton or silk covered litz wire, the primary is solid.

For a short TC this thing belts out some volts and I seem to remember a bit of corona. To wind one of those, the pis are woundcrisscross a small winder with a very tight cam about 1/8 inch
would be needed. those pi winds were common in old radio sets.

It also has a dynomotor to run on 28 volts DC.

this the coil used. scanned from Boyce W. and Roche J. Video Handbook. HW SAMS & Co. Indianapolis, 1954

radiotech

Tue Jan 26 2010, 07:50PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

sorry about 2 images-the first belonged to a prev thread about static groundings in wall warts for tvs.
never could find it until- today-i hate computers!

HVPaul

Tue Jan 26 2010, 08:07PM

Registered Member #2321 Joined: Fri Aug 28 2009, 05:13PM
Location: Boston, MA
Posts: 34

I don't know the make of the transformer.
It seems like a common design. Like the ones Here

dmg

Tue Jan 26 2010, 09:18PM

Registered Member #2628 Joined: Fri Jan 15 2010, 12:23AM
Location:
Posts: 627

considering that its high frequency and that you plan on using a voltage multiplier, i think that you should build a fullwave multiplier, and due to the high frequency, your tank capacitor might proove rather troublesome, with my experience with high frequecy TC drivers with a spark gap, usually the output is rather poor if and I had some capacitors fail on me while exposing them to this kind of abuse, do not use ceramic disk capacitors or tantalum capacitors, they explode like cherry bombs, If you can get a hold of polypropylene film/foil capacitors those will work the best, and considering that the output of a multiplier is DC, you might want to utilize some type of charging inductor before you tank capacitor and maybe a blocking diode. (DC resonant circuit)

dont take this as a discouragment, but if you want a small tesla coil, I think you will have an easier time by using a small 60hz transformer (bug zapper XMFR, OBIT, etc.) rather then the cold cathode supply.

also, from personal experience (or disaster more like it), last time I tried to use those things as an HV supply of any kind, I had an arc flash, kicked breaker, a bunch of smoke and a fire engulfed circuit board, and little bits of components and silicon permanently lodged into my wall, It was not a pretty site.

none the less, if you do intend on using a multiplier, like arcstarter said, there will be corona, coating the finished multiplier with epoxy or mineral oil submersion in something like a pill bottle or other plastic case might be a nice idea.

Myke

Tue Jan 26 2010, 11:13PM

Registered Member #540 Joined: Mon Feb 19 2007, 07:49PM
Location: MIT
Posts: 969

I think it's fine to build a small TC out of these CCFL drivers. The current fed royer is a very common design for these drivers since it's cheap to make, simple, and can supply enough power without dissipating much heat.

The multiplier should be rather easy to build since it's a HF supply so the caps don't have to be all that big. You don't need all that high of a voltage to make a small coil. To reduce coronal losses, try to avoid sharp bends and clip/file all things that form a point. Since this is such a low voltage, I don't think there will be much of a problem.

Remember that these supplies aren't designed to supply a lot of current so you may want to put a current limiting resistor or capacitor on the input in case the output of the multiplier shorts. The input of the multiplier already has a cap so I'm not sure if you even need some current limiting.

I think all the rest of the design should be similar to a SGTC tesla coil otherwise.

gatedbreakdown wrote ...

do not use ceramic disk capacitors or tantalum capacitors, they explode like cherry bombs

I was not aware that they made non-polar high voltage tantalum caps.

dmg

Wed Jan 27 2010, 02:45AM

Registered Member #2628 Joined: Fri Jan 15 2010, 12:23AM
Location:
Posts: 627

Myke wrote ...

I was not aware that they made non-polar high voltage tantalum caps.

I made a string out of a bunch of them and I was using DC resonant charging, so I assumed that they will charge, however they had very low capacitance, considering I only used one string and didn't parralel them. and I may have overestimated thier voltage capabilities, (I didn't even check, I had about 50 of them in several sizes and just soldered them all together for the hell of it). since my supply was a flyback(black with diode stick potted inside) , I used an AC flyback as a charging choke and a string of 5 Microwave oven diodes in series for the blocking diode. (the diode string to my suprise survived)

so at the time I didn't really know those caps dont like high voltage, and as you can immagine, they went boom.
(didn't matter anyway, at 50 pieces in a string the capacitance was probably non-existant)
as for the ceramic disk, I made a bank of 10 2kv disks (forgot the capacitance but they had a blue collor.) and while the ratings were higher then the input voltage, they ran for a few seconds, and then the bank had random capacitors fail. (probably because of the high frequency)

the only capacitors I ever had that worked for TC duty were either home made polly plate capacitors or polypropylene capacitors.
so thats why I never use the other types of capacitors and high frequency power supplies for TCs, flybacks work okay, but sometimes either they or the driver mysteriosly dies. (killed a pair of 2N3055 power transistors and a DC flyback this way.)

HVPaul

Wed Jan 27 2010, 07:02AM

Registered Member #2321 Joined: Fri Aug 28 2009, 05:13PM
Location: Boston, MA
Posts: 34

I measure the frequency of this CCFL power supply and it's between 32 and 38kHz which is way below what some members run their resonant SSTCs. Would ZVS generate much lesser rate?

gatedbreakdown wrote ...

considering that its high frequency and that you plan on using a voltage multiplier, i think that you should build a fullwave multiplier, and due to the high frequency, your tank capacitor might proove rather troublesome, with my experience with high frequecy TC drivers with a spark gap, usually the output is rather poor if and I had some capacitors fail on me while exposing them to this kind of abuse, do not use ceramic disk capacitors or tantalum capacitors, they explode like cherry bombs, If you can get a hold of polypropylene film/foil capacitors those will work the best, and considering that the output of a multiplier is DC, you might want to utilize some type of charging inductor before you tank capacitor and maybe a blocking diode. (DC resonant circuit)

dont take this as a discouragment, but if you want a small tesla coil, I think you will have an easier time by using a small 60hz transformer (bug zapper XMFR, OBIT, etc.) rather then the cold cathode supply.

also, from personal experience (or disaster more like it), last time I tried to use those things as an HV supply of any kind, I had an arc flash, kicked breaker, a bunch of smoke and a fire engulfed circuit board, and little bits of components and silicon permanently lodged into my wall, It was not a pretty site.

none the less, if you do intend on using a multiplier, like arcstarter said, there will be corona, coating the finished multiplier with epoxy or mineral oil submersion in something like a pill bottle or other plastic case might be a nice idea.

1 2 next

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