magnetic pulse compression / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

magnetic pulse compression

first previous 2 3 4 5 next last

Move Thread

LAN_403

Proud Mary

Fri Oct 01 2010, 04:19PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

This paper has practical constructional details of a ferrite coaxial line pulse sharpener using readily available toroids:

FERRITE LINE TO DECREASE RISE TIME OF HIGH-VOLTAGE
NANOSECOND PULSES
V. Korchuganov, Yu. Matveevâ€ , D. Shvedov

cedric

Mon Oct 04 2010, 02:07PM

Registered Member #2941 Joined: Fri Jun 25 2010, 08:08AM
Location:
Posts: 143

2Spoons wrote ...

You really want a square loop ferrite such as 3R1 for this sort of thing. They have a very abrupt change into saturation, and they stay saturated until a reverse field is applied. In other words, the remanent field is enough to maintain saturation as the inductor current drops to zero - like a switch in fact.

would it mean that the ferrite become a magnet? if the core remane saturate ,it mean that I have to desaturated it before the next pulse by applying reverse current ,not very practicle,for what's related to the abrupt change in saturation ,it is appealing.
what is the typical use of this type of ferrite (3R1)?

2Spoons

Mon Oct 04 2010, 11:29PM

Registered Member #2939 Joined: Fri Jun 25 2010, 04:25AM
Location:
Posts: 615

Yes, it will retain a small magnetic bias, as do most, if not all, soft magnetic materials. The remanent field (often notated as Br) is the reason the B-H loop is a loop, rather than a line, and results in hysteresis loss in magnetic devices.
Don't worry about it - it will only take a small reverse current to remove the bias.
3R1 is actually intended for use in mag pulse compression, magnetic amplifiers and similar devices.

2Spoons

Tue Oct 05 2010, 04:46AM

Registered Member #2939 Joined: Fri Jun 25 2010, 04:25AM
Location:
Posts: 615

look about half way down the page for a couple of orange toroids : two types of Square Orthonol toroid!

cedric

Wed Oct 06 2010, 09:47AM

Registered Member #2941 Joined: Fri Jun 25 2010, 08:08AM
Location:
Posts: 143

thank you...

cedric

Wed Oct 06 2010, 10:48AM

Registered Member #2941 Joined: Fri Jun 25 2010, 08:08AM
Location:
Posts: 143

I order some ferrite today ,I will post my result when I get some...

cedric

Fri Oct 29 2010, 10:54AM

Registered Member #2941 Joined: Fri Jun 25 2010, 08:08AM
Location:
Posts: 143

as promise ,I post some of the result I get .
the first step was to build a solid state pulser,this one is rated at 4 kv as it is just a proof of concept
Pulsar

the picture is pretty self explanatory I suppose ,those solid state switch are igbt ,rated for 600 volt and 250 amp pick ,important note ,the small diode running through there leg are tvs diode ,in between the base and emitter so I can provide enough current to the gate regard less of the voltage spike and in between collector and emitter so that if they don't fire exactly in the same time the igbt stay protected.the gate are trigger with a gate drive transformer to provide dielectric isolation ,the primer is the red cable passing through them all(20kv rated)off course the magnetic coupling is not very good so the gate signal have to be monitor to assure proper triggering of the mosfet.

the rising time in this configuration was about 80 ns for a pulse with of 180 ns ,the input voltage was about 4 kv and the output around 3 (problem with impedance matching.) the measurement was taken from a shunt on a 75 ohm cable (one volt on the scope is 3600 volt in the cable.)

the magnetic pulse compression board is made of epoxy ,the ferrite are of the type specified before on this thread, (3R1)
Mpc

the value of the capacitor and inductor was found empirically,but it must be noted that the capacitor should be from the same value if it is just to compress the pulse current.
one important think about those ferrite ,during my previous experiment ,the system didn't work because I was passing only one conductor in the middle of the ferrite,the magnetic coupling was very bad ,much better result can be reach by winding the cable few turn around the ferrite.
Tek00000

the result is very explicit ,the gray trace is the pulse with out magnetic pulse compression while the yellow is with it ,
the voltage is now almost reaching the input voltage,the rise time is about 15 nanosecond.
that was just a prototype.
the system still need to be optimize, I am now trying to build a 10 kv pulser with magnetic pulse compression.