Freds world of science / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Freds world of science

Move Thread

LAN_403

paris

Wed Feb 27 2013, 09:31PM

Registered Member #3042 Joined: Wed Jul 28 2010, 12:36AM
Location:
Posts: 121

I found this site and sent fred an email, the spark gap he uses = ZR7516. I thought the 4HV folk might find it interesting . theres stills of the apple more or less 'walking into the light ' for the last time , amazing what can happen with energy compression.

GammaRay

Sat Mar 02 2013, 12:55PM

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

I have witnessed this water-explosion phenomenon when experimenting with a carefully spaced spark gap submerged in water. My theory is that a small qty of hydrogen/oxygen is created by the spark energy via electrolysis. The spark exists just long enough to ignite the newly formed gases and â€¦ pow! Whatever the real reason, this water-explosion phenomenon is interesting and surprisingly loud.

paris

Sun Mar 03 2013, 02:58AM

Registered Member #3042 Joined: Wed Jul 28 2010, 12:36AM
Location:
Posts: 121

hard to say, freds transformers look pretty big and that spark gap is a few inches long , apparently they use them in pershing missiles.(spk gap)

the cap discharge would have to be DC so maybe room for polarisation of molecules but at 1 micro sec or so I think there isnt much time for separation, certainly the current tho.

that apple must surley be instantly melting ,he says it smells sweet in the room.
for that amount of energy to pass thru the water with out turning to large amounts of hydrogen must indicate how strong the molecular bond is , even tho dissociation potential is only 1.23V DC.

Sandia use water caps with MV across them and they do not break down into gas , the cycle is only micro sec any way.
Ive read on net of guys using misters or garden foggers and sparks , maybe I should try that some day.

Pinky's Brain

Sun Mar 03 2013, 10:07AM

Registered Member #2901 Joined: Thu Jun 03 2010, 01:25PM
Location:
Posts: 837

paris wrote ...

Sandia use water caps with MV across them and they do not break down into gas , the cycle is only micro sec any way.

My guess is that though the discharge time is short in the explosion experiments, the time the voltage is across the water is actually large (ie. there is a significant delay between trigger and the current/explosion starting). Whereas Sandia cuts off the voltage before that can happen.

Steve Conner

Sun Mar 03 2013, 07:49PM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

The electrical properties of water vary depending on the timescale. For nanosecond pulses, it looks like a capacitor with a dielectric constant of 80. (Huge compared to most materials normally used as capacitor dielectrics.) Sandia Labs fill their pulse forming lines with carefully purified water. The high dielectric constant lowers the impedance of the line, allowing it to store more energy and discharge higher currents into the load.

For longer pulses, the water molecules start to come apart under the electric stress and it becomes conductive. The sparks you see in the Z-machine photos are probably something to do with this, but they're driven by leftover energy after the main discharge has passed.

Pure water is quite a good insulator even at long timescales, but it's also a good solvent that loves to pick up impurities. Any water that gets into electrical equipment in the field will be dirty enough to conduct quite well.

GammaRay

Sun Mar 03 2013, 08:22PM

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

I found that proper gap spacing of the underwater electrodes was all important in order to produce the water "pop" experience. Too large, or too little gap, and only bubbles were produced, no pop. While in bubble producing mode, I placed an upside down glass in the water directly over the spark gap. It took a while to collect very much gas. When the collected gas was lit by a match, it ignited. I can't speak 1st hand about Fred's experiment, but for my experiment, I have little doubt the bubbles are Hydrogen/oxygen mix. As to how the gas becomes ignited underwater to produce the pop and splash, my guess is the spark itself provides ignition.

paris

Sun Mar 03 2013, 08:55PM

Registered Member #3042 Joined: Wed Jul 28 2010, 12:36AM
Location:
Posts: 121

^^^worth looking into that more.

i had a look at sandia a few years ago, sites changed. but wiki Z machine page is the same,
impressive pic . they say the discharge is 100nsec. some where on the original sandia site it had the circuit for the water caps and the lines, measurements and all.

They use MARX gen to get the HV and they use smaller water switches which give way 1st ,that discharge the larger water caps = clever

I still think what freds doing is cool, they say intrinsic properties of water =permittivity / conductivity , also a conductor has infinite permittivity , that sample of water he has ,is limited by its dimension , it cant be a cap because of the electrode shapes are pretty much points, and it couldnt be dielectric breakdown , the apples made of water too.

he mentions the switch as having negligible R so all power is dissipated in the load R.

(reakon gamma ray gotta keep us up on what hes doing too)

Mattski

Mon Mar 04 2013, 06:36AM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

Steve Conner wrote ...

The electrical properties of water vary depending on the timescale. For nanosecond pulses, it looks like a capacitor with a dielectric constant of 80. (Huge compared to most materials normally used as capacitor dielectrics.) Sandia Labs fill their pulse forming lines with carefully purified water. The high dielectric constant lowers the impedance of the line, allowing it to store more energy and discharge higher currents into the load.

...

I thought that even pure water was a fairly lossy dielectric at nanosecond timescales (

, is that not true or does it work out that it is loss but is still a good choice for pulsed power systems for the high dielectric constant?

Steve Conner

Tue Mar 05 2013, 10:35AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

At 20'C and 1GHz that calculator shows a loss tangent of 5%. 0.5% at 100MHz. I'd say that was worth it for the dielectric constant of 80.

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