On Frequencies and Fireballs

[giantkiller]

In the paper I posted a link to that discussed how electrons begin to drift in a conductor, a traveling wave of dispacement current initially travels along the outside of the wire and pulls electrons with it at it's velocity (C - give or take)

Repeated reflections of this displacement current cause more and more electrons to drift as the energy diffuses into the conductor.

The link is:

http://www.google.com/url?url=http://espace.library.uq.edu.au/eserv/UQ:9792/saha-edwards-aup.pdf&rct=j&sa=U&ei=vXi_SvL_JNLr_AaK74WCBQ&ct=res&cd=1&sig2=9ivonCUCUzWCr0BQKFcNew&q=electrons+current+diffuses+displacement+conductor+edwards&usg=AFQjCNECf3P_pak1lkQY7cvzDF3KwKKJDw

Sticking with the initial impulse, we have a traveling wave hauling ass along the wire and pulling electrons like a venturi effect - perhaps causing Chef's puncture from an arc through aluminum.

So what change does the "system" undergo during the rise of the initial pulse?  Immediately on closing of the circuit, there is a very sudden change in the polarization of the space around the conductor - a polarization current - radial around the conductor.  When this occurs and it's rate of propagation / change (velocity?) exceedes the speed of light in the system, which can be much lower than C, then the shockwave similar to braking radiation occurs. 

This is not a particulate energy.  I would guess that it is photon-based, but only for lack of a better explanation.

Does the fall of the pulse "reset" the system?   I suspect that it does and a slow fall slows the reset, and falling below the baseline, should reverse the effect.

I understand this first part.
I split these into two by basis...

Have you found that the repetition rate is related to the impedance of the item being excited?  Yes. The 15" loop takes longer to compile or buildup

Have you ever tried exciting Litz or simlar multiconductor?  It is a guess that  it is detrimental, and I have never heard of anyone trying this. Yes before. I used the EZflow audio cable. Litz wire strands coated with silver. Have in inventory and can pull it into buid.

You mentioned MOSFET's before (840's), are you using one to start the pulse and another to stop it?  Yes. Dual pulse protocol using Jason's hardware, controller and SSR boards.

Have you tried them in avalanche mode?  Not yet. Didn't pursue that and am not sure how to get that done.

How high of a voltage / current are you using? 5v fet gate. +200v source to drain coil bias on one bifialr run and pulse other bifilar run.
<100ma current.

The cross-talk between the MOSFE's may be some sort of standing wave effect (it looked like a clean sine wave) between the switches.  One side of the switch looks like a brick wall, so it reflects between them.   I see that.

How long did this reflection occur before it smoked the MOSFET's or dissipated?  This is very significant even if no one else sees it as such. The event hits and no hardware damaged. Between 1 and 2 minutes. Controller just shuts off because it's supply is a lm7805, an inherent firewall! The fets left on cause the current rise thereby triggering thermal protection in the v reg.

[Grumpy]

Sorry GK, I postd in the wrong thread - thanks for the reply though.

I was referring to this:
http://www.overunity.com/index.php?topic=7926.msg202336#msg202336

The MOSFET's kept oscillating when the sig gen was turned off.

===============================================

We will be departing Kansas momentarily, en rout to Norfolk.  Thank you for choosing Tesla Air Lines.

[darkspeed]

http://www.metglas.com/products/page5_1_3.htm

SATURABLE REACTORS
Saturable reactors utilize the large change between unsaturated and saturated permeabilities of their cores to delay current for a preset period of time. Similarly, once saturated in the forward direction, they act as a diode temporarily blocking current in the reverse direction. (see figure, below)

MAGNETIC SWITCH PROTECTION (MSP)
MSP devices made with Metglas®cores have dramatically enhanced the reliability and overall lifetime of pulse power systems. By using a saturable reactor in series with either a semiconductor or thyratron switch, the circuit designer can reduce losses in the switch and extend its life. The saturable reactor is designed to hold-off current until the switch becomes fully conductive (see figure, right). This delay reduces the overlap between current and voltage in the switch, thereby reducing power absorber in the switch. MSP also offers other advantages. Higher di/dt's to the load are safely achieved by waiting for full conductivity in semiconductor switches. The diode-like characteristic of a saturated reactor provides time for switch recovery.

MAGNETIC PULSE COMPRESSION (MPC)
Magnetic pulse compression utilizes reactors in conjunction with capacitors to shape input pulses into narrow output pulses of much higher current . MPC, therefore, allows the designer to use less expensive input switches with lower current ratings. MPC can also extend the lifetime of the input switch. Advanced MPC devices - capable of generating power levels of multi-terawatts in tens of nanoseconds - have been realized utilizing Metglas® cores.

http://www.metglas.com/images/ppc-chart.gif

And for those who see the potential > http://www.metglas.com/design/magamp.asp

[Grumpy]

what is a realistic repetion rate and rise time with saturable reactors?

[darkspeed]

what is a realistic repetion rate and rise time with saturable reactors?

Depends on the material, size, etc..

I think the repetition rate is not as important as the rise/peak/fall time
I have heard mention of using a second control winding wired opposite of the main dc winding to snap desaturate the core. needs more research.

I have seen reported 5ns transition times.

Im may see if these guys will put together a HF developer kit for me with a variety of cores to play with.

I had hoped to outline a test setup today related to my mosfets but its been a busy day.. maybe tomorrow..