Xenon flashlamp as high energy switch? / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Xenon flashlamp as high energy switch?

Move Thread

LAN_403

Signification

Wed Dec 30 2015, 03:00PM

Registered Member #54278 Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 367

Has anyone tried using a small xenon flashlamp, such as those from disposable cameras, as high power switch. I am playing around with it now, but would like to know of any previous attempts. My main objectives is the ease of remote isolated triggering with the HV trigger wire (like a standard laser flashlamp). I would like to go from small to large scale with minimum loss in the Xenon tube (compared to load energy). To test the the tiny lamp from a disposable camera, I rigged it to fire continuously with the original electronics and a 1.5V adapter--it's been flashing over 24 hrs at about 4s intervals (the firing rate slowed slightly). I have also tested one separately at a much higher energy level -than intended- using a larger capacitor and transistor. However, to increase the firing rate to several Hz I had to use a 6V rig. rather than the single 1.5V AA power or some of the original circuit. So far, I failed to blow it up with seven (7) paralleled caps from these cameras. Also when I got the firing rate fast 5-6/sec, the tube started to glow orange and bend near the contacts after a while, but kept going--the first fail component was the transistor (a Chinese in a TO-220 case).

Soon as I get time in about another week, I will do more conclusive work, but would love feedback from anyone who may have used the cheap glass tube (which is incredibility robust) as a remote triggered high energy switch for SINGLE PULSE applications. So, I assume the main question is: How much energy is lost in a --quickly-- ionized Xenon tube compared to a fitting load?

hen918

Wed Dec 30 2015, 03:35PM

Registered Member #11591 Joined: Wed Mar 20 2013, 08:20PM
Location: UK
Posts: 556

Flash lamp swiches are/were quite popular in (very) small coil guns and, although I've never used one, they are apparently very inefficient, which you would expect given that the entire energy from a photoflash capacitor is designed to be dissapated in one.

Signification

Wed Dec 30 2015, 06:23PM

Registered Member #54278 Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 367

Maybe I could adjust flashlamp timing parameters (mainly via Ko) such that the flashlamp very rapidly becomes ionized (a wire?) much sooner than the cap is discharged to an insufficient energy level.

Sulaiman

Thu Dec 31 2015, 07:27AM

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

Roughly, the voltage across a conducting xenon flash tube is a fixed voltage + K.sqrt(I)
which does not really help much, the important thing that we know is that it is less than 300 V (say 200 V)
because that is the initial discharge capacitor voltage.

So if you have a resistive load in series with the flash tube
the peak load voltage would be 300 - 200 = 100 V ... usable but inefficient.

If you use a higher voltage, e.g. three flash inverters in series, 3x 300V = 900 V,
then the load voltage would be 900 - 200 = 700 V ... much more efficient.

From memory, the small xenon flash tubes can easily hold off >1000 V

hen918

Thu Dec 31 2015, 02:51PM

Registered Member #11591 Joined: Wed Mar 20 2013, 08:20PM
Location: UK
Posts: 556

According to page 15 of

;
The resistance of the flash-lamp, R(t) = (p*l)/a
where p = plasma resistivity, which depends on pulse width, and is supposed to be somewhere between 0.015 and 0.025
l = arc length in cm
a = area of the flash lamp in cm^2

Signification

Tue Jan 05 2016, 11:53AM

Registered Member #54278 Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 367

Sulaiman wrote ...

Roughly, the voltage across a conducting xenon flash tube is a fixed voltage + K.sqrt(I)
which does not really help much, the important thing that we know is that it is less than 300 V (say 200 V)
because that is the initial discharge capacitor voltage.

So if you have a resistive load in series with the flash tube
the peak load voltage would be 300 - 200 = 100 V ... usable but inefficient.

If you use a higher voltage, e.g. three flash inverters in series, 3x 300V = 900 V,
then the load voltage would be 900 - 200 = 700 V ... much more efficient.

From memory, the small xenon flash tubes can easily hold off >1000 V

I really like this: looking at voltage v(t) from this point of view (instantaneous) seems very useful--that is: flashlamp voltage v(t) equals some constant voltage (V) plus Ko*sqrt[i]. Or:

(1): v(t) = +/- Ko * sqrt[i]; expressing INSTANTANEOUS voltage: v(t) as an INSTANTANEOUS current: i(t)
where PEAK current (Io) is given as:

(2): Io = Vo / (Zo + R); where Vo, the initial capacitor voltage v(t=o); Zo=total impedance [sqrt(L/C)]; and R is that somewhat elusive (to me) flashlamp resistance (a constant depending on lamp parameters--these can be found in a datasheet if you are lucky enough to have one).

NOTE: All capital letters represent peak values, and lowercase variable functions are variables of t: such as, v(t) and i(t) are instantaneous WRT t.

The main goal of all of this is to get critical damping (CD). In tuning, I think I would shoot for as close as I could get to CD, then use the least critical component at the time (usually V, maybe L: see the golden rule) to tweak to CD with a DSO.

Anyway, for those interested, there is a golden rule associated with all this--it goes:

================================================ =============================
For any given flashlamp {Ko in units of [Ohm * sqrt(i)] }, pulse-width (in seconds) and energy (in Joules): there is one-and-only-one value of Vo [initial capacitor voltage V(t=o)], C (the constant capacitance), and L (the constant inductance) that can yield critical damping.
========================================= =====================================
...fasinatin g!

EDIT: @ hen918.
Your reference to the parameter "peak resistance=(R)" clears up all confusion I mentioned above concerning "R", thanks.

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