Seek guidance to step-down 1kV pulse to 12volts / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Seek guidance to step-down 1kV pulse to 12volts

1 2 next

Move Thread

LAN_403

GammaRay

Fri Jun 15 2012, 08:11PM

Registered Member #5323 Joined: Fri Jun 15 2012, 02:14PM
Location:
Posts: 104

I've inherited a Zamboni Dry-pile electrostatic battery. It outputs about 1kV in the microamp range. No telling how old this thing is, but it still works. A google of the Zamboni battery leads one to believe it can operate for decades with a very light load. I connected a 1kV spark gap (a gas spark arrestor) to the Zamboni and it causes a spark to jump the arrestor at about 1Hz, slower if there is low humidity. I am interested in an efficient method to step down the 1kV spark to around 12 volts. That is my reason for making this post on this Forum. From there, the 12 volts will be used to pulse charge a small cap that will trigger an SCR when X voltage is reached and dump the cap's accumulated charge into a small load, such as LEDs or audio buzzer. I am aware this will only provide occassional-intermittent operation of the load, if it is even possible at all. Why am I doing this? For fun, but mainly to see if it can be done.

Sulaiman

Sat Jun 16 2012, 05:52AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

I would start with a low-leakage capacitor
(e.g. a plastic film capacitor, ideally polypropylene, 1nF to 100nF)
with a voltage rating higher than the battery voltage (e.g. 1500V)
and put it across the battery.
This will slowly accumulate a usable amount of energy.

You could periodically dump the energy
- directly into a high voltage load such as a xenon flash tube
- or via a pulse transformer into a low voltage load

If you are lucky you may find a small xenon flash tube that fires due to over-voltage below the battery voltage, in which case you would only need the capacitor with the xenon tube directly across it.

A 100nF capacitor charged to 1kV holds 50 mJ, not lethal - but painful!

GammaRay

Sat Jun 16 2012, 02:44PM

Registered Member #5323 Joined: Fri Jun 15 2012, 02:14PM
Location:
Posts: 104

Thank you for your suggestion, I like it. I have already placed an order for the 100nF polypropylene cap.

I reviewed the large selection of pulse transformers at Digikey, I must admit that selecting the correct pulse transformer appears a bit challenging. Inductance and E.T. requirements for your recommendation are unknowns to me. I am not confident on which pulse transformer to purchase. I fully understand this forum is not to be used as a crutch, but rather for legitmate assistance and idea sharing. I have put forth a reasonable effort researching the pulse transformer offerings, prior to asking the below.

Rather than spend good money and wasted time on a wild guess, I kindly ask for a recommendation on which pulse transformer part # to purchase. Then I'll move forward with this project.

Respectfully.

Sulaiman

Sat Jun 16 2012, 05:21PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

I would be surprised if you could choose an appropriate transformer - not easy !

I misled you a bit
... the transformer has to handle pulses but it would not be called a 'pulse transformer'
they are usually for triggering or isolation, not power.

I would think that an smps transformer would be your best bet,
(other than a carefully made diy one, e.g. on a ferrite antenna rod, or similar)
look in disused smps for tv/monitor/pc/etc.......Bigger = more V.A.s = better
Your first attempt may result in internal flashover
in which case a second transformer will be required, maybe under oil?
(cooking oil will do)

If you are going to rectify the low voltage output
(the output will probably be a large pulse of one polarity, followed by quite a bit of ac 'ringing')
be sure to use 'fast' rectifiers.

------------------------------------- -------------------------------------------------- ------------------
OR
you could (just for novelty) make an 'inverse Tesla coil'
Use the 'secondary' as a primary to the spark gap and battery
rectify the output from the 'primary'
OR
if you want really bizzare - use the 1kV battery as a power source for an actual spark gap Tesla coil, very low power with occasional (unexpected!) large sparks!
(for a small sgtc the spark length is determined by energy per discharge,
the power mainly affects the repetition rate)
THESE IDEAS WOULD NEED SOME THOUGHT TO PREVENT RF IN THE BATTERY
------------------------------------------ -------------------------------------------------- -------------

GammaRay

Sun Jun 17 2012, 02:21AM

Registered Member #5323 Joined: Fri Jun 15 2012, 02:14PM
Location:
Posts: 104

Checking various vendors, not much luck finding SMPS transformers with large enough turns ratio to convert 1kV to 12v, but I will keep looking. What do you think about my trying the ballast of a fluorescent light, in reverse?

Sulaiman

Sun Jun 17 2012, 06:00AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

The transformer in a typical CCFL inverter would be excellent.

The small capacitor on the output of the CCFL transformer would also match well,
the 100 nF that you have ordered may saturate the core :(

GammaRay

Sun Jun 17 2012, 12:17PM

Registered Member #5323 Joined: Fri Jun 15 2012, 02:14PM
Location:
Posts: 104

Found a CCFL inverter (12v to 1kV) used for a UV lamp at:

Do you think this is a good candidate for reducing 1kV pulse to 12v pulse when used in reverse?

Antonio

Sun Jun 17 2012, 01:16PM

Registered Member #834 Joined: Tue Jun 12 2007, 10:57PM
Location: Brazil
Posts: 644

You can try something as in the schematic. Connect a high-voltage capacitor (~100 pF) in parallel with the battery. The spark gap acts as a switch in a buck DC/DC converter, transferring ideally all the energy in C1 to the load capacitor C2 (~1 uF) at each firing. The inductor L shall have low inductance (~100 uH) to keep the energy transfer time short, while the spark is active. An air-core inductor would have enough insulation easily.
1000 V over C1 will charge C2 by 1000*sqrt(C1/C2) = ~10 V at each firing. D1 is a low-leakage high-voltage diode. D2 is not critical, but must be fast. The load Z uses the charge in C2 when a certain (low) voltage is reached, and shall consume very little current from C2 before this.

GammaRay

Mon Jun 18 2012, 08:40AM

Registered Member #5323 Joined: Fri Jun 15 2012, 02:14PM
Location:
Posts: 104

Thanks. Will order the parts. For the inductor material, would ferrite core be more efficient than air coil, or doesn't matter? For D2, a standard Schottky should do? I understand the buck converter circuit except for D2, what purpose does it serve?

Antonio

Mon Jun 18 2012, 04:31PM

Registered Member #834 Joined: Tue Jun 12 2007, 10:57PM
Location: Brazil
Posts: 644

A ferrite-core inductor would be smaller, but take care with the insulation. D2 impedes the return of the charge in C2 to C1. Really, its presence makes little difference at this large capacitance ratio.

1 2 next

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