Reverse Voltage in Flashlamp PFN / General Science and Electronics / Forums

Forums

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

« Previous topic | Next topic »

Reverse Voltage in Flashlamp PFN

1 2 next

Move Thread

LAN_403

MikeT1982

Wed May 23 2007, 03:23AM

Registered Member #621 Joined: Sun Apr 01 2007, 12:37AM
Location:
Posts: 119

I've been checking out as many flashlamp design circuits as I can while designing mine. I am running 16 seriesed electrolytic caps (244uF at around 5500 volts total), and will be flashing a 40 something inch long helical tube. I have noticed that some employ a Reverese Polarity prevention diode across the flashlamp circuit near the tube. How can a diode handle that type of current?! and why do you need one? I had a hard enough time finding the big diodes to rectify my step up transformer I can't imagine a thousand something amp diode...

ragnar

Wed May 23 2007, 03:39AM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

I suspect the diode is not for the protection of the tube, but to stop the capacitors charging in reverse.

...

Wed May 23 2007, 05:30AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

As Matt pointed out, the diode is probably to protect the caps from reverse charging, which would not be very good for them. It could also to spikes formed when the tube stops conducting and there is a sudden change of current across the inductor.

In any case, you would be suprised at the peak current rating of normal silicon diodes, they can be had with Ipk well over a thousand amps. Although a string of enough to stand off 5kv would probably be pretty expensive...

MikeT1982

Wed May 23 2007, 06:08AM

Registered Member #621 Joined: Sun Apr 01 2007, 12:37AM
Location:
Posts: 119

Thanks!

Dr. Shark

Wed May 23 2007, 07:59PM

Registered Member #75 Joined: Thu Feb 09 2006, 09:30AM
Location: Montana, USA
Posts: 711

Correct, the continous current rating is determined by how fast heat can be conducted from the diode to the outside world (thermal resistance), whereas peak current is basically a function of die area which is related to the thermal capacitance of the diode. Basically you are looking at the energy that can be dissipated in the die in an adiabatic process (no time for the heat to leave) before it reaches its maximum temperature.
EVR knows a lot about this, since he is trying to calculate the allowable peak currents in his commercial DRSSTC designs. If you are really interested I could try digging out one of his old threads on this.

Steve Conner

Wed May 23 2007, 09:11PM

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

As far as determining peak currents goes, I always used the "Crank it till it explodes then back it off a quarter turn" method. :P

Bert

Thu May 24 2007, 02:09PM

Registered Member #118 Joined: Fri Feb 10 2006, 05:35AM
Location: Woodridge, Illinois, USA
Posts: 72

If you are directly discharging your capacitor bank through the flash tube, you should not encounter ringing. Initially, the flashtube exhibits a negative resistance characteristics during initial spark breakdown and arc formation. However, once the arc channel is fully formed, the plasma is constrained from further expansion by the cool walls of the glass tube. Because the arc channel is restricted from further radial expansion, the tube exhibits a strong positive resistance characteristic at higher currents instead of the negative resistance characteristic associated with freely expanding arcs. This behavior is true for all radially-constrained arc discharges. As a result, high current flashlamp discharges are virtually always nonoscillatory unless you have purposely inserted significant inductance into the discharge loop (i.e., a series trigger transformer winding).

BTW, you may still wish to add individual diodes across EACH capacitor in the chain, since capacitors with lower capacitance may otherwise see reversed voltage, even during nonoscillatory discharges...

Tesladownunder

Thu May 24 2007, 03:07PM

Registered Member #10 Joined: Thu Feb 02 2006, 09:45AM
Location: Bunbury, Australia
Posts: 1424

Helical flashlamp tubes have inductance with more turns than a can crusher although smaller currents are involved. I don't have hard data but suspect that this may well be one cause of catastrophic failure in helical flashlamps with the radial expansion and axial compression. These are the same forces that explode the work coil that Bert knows so well.
If you are blowing diodes then you are catching a voltage reversal. You need huge diodes. I was using 4 hockey puck SCR's wired as diodes and blew them at 13kA reversal 3kV. And that was only warming up to the estimated 40kA reversal current.

TDU

MikeT1982

Thu May 24 2007, 09:35PM

Registered Member #621 Joined: Sun Apr 01 2007, 12:37AM
Location:
Posts: 119

Is "ringing" the reverse current itself or the magnetic feild and electrical feild's interaction causing the reverese voltage?

Tesladownunder

Fri May 25 2007, 12:54AM

Registered Member #10 Joined: Thu Feb 02 2006, 09:45AM
Location: Bunbury, Australia
Posts: 1424

Ringing is the shiftng of energy from inductance to capacitance.
Similarly a childs swing shifts energy from potential energy to kinetic energy.
How much ringing depends on how much energy you remove from the system such as in resistence causing heating.
In can crushing about 60% of the energy returns after the first cycle. Enough to blow very big diodes.

TDU

1 2 next

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