The problem with high speed photography is getting a standard camera to
react fast enough to capture a short lived moment. There are several
ways to work around the problem, one of them being high speed flash
units. The idea here is that instead of using an exceptionally fast
camera with amazing light sensitivity, one instead uses an exceptionally
fast blitz along with garden variety camera. The camera is set up to
take a long exposure with large aperture, in complete darkness.
Sometime during the long exposure, the event you wish to capture takes
place, and simultaneously triggers the blitz via a sensor. The blitz
exposes the scene to an intense, but very short flash of light, which
exposes the image. Since the flash duration is ultra short, only events
which took place when the blitz was triggered are visible in the final
image.
The
difficulty is creating a flash which is both bright enough to expose
the image, and short enough to not cause blurring. A solution to this
problem is to discharge a high voltage capacitor in a relatively high
pressure (ie. 1 atm) gas. This creates a short-lived, but bright flash of light
suitable for high speed photography. For anyone who's experimented with
flyback transformers, the final setup will be easy to create.
The blitz here was triggered using the microphone trigger seen at the bottom right.
The
air flash unit consists of a circuit for charging a high voltage
capacitor, a circuit for triggering a discharge on demand, a high
voltage capacitor and the air flash tube itself. All of these
components can be made using scrap materials. The way the unit works is
by first charging the HV capacitor to full potential. The flash tube
contains two wires which are separated just enough to not spark over
at max potential. There exists a third wire in the flash tube, which
is isolated from the other two. This wire is connected to a
trigger/pulse transformer, and when the blitz unit is activated, this
pulse transformer ionizes the gap between the two capacitor leads. This
causes the gap to breakdown and a spark to form, thereby creating a
flash of light. If my setup is comparable to the one presented in
Scientific American, the duration of peak intensity is only 300ns, which
equates an exposure time of 1/3000000 of a second.
I constructed my flash
tube using an olive jar and a glass test tube for vanilla sticks. The
glass test tube is the most essential part, as the spark must travel
along it's surface. By allowing the spark to surface track, it
quenches much faster than an open air spark, shortening the blitz
duration. At the same time a third wire must be brought in through
the test tube, to ionize the gap between the capacitor leads on the
outside. I soldered a piece of copper pipe to the trigger lead, so it
had a more constant distance from the walls. The spacing between the
capacitor leads is about 30mm. The greater the distance the greater
the voltage needed to break the gap down, and the brighter the flash
will be. 1mm of distance does not correspond to 1kV here, since the
spark is tracking along a surface. I found at just 20kV, my capacitor
would breakdown the 30mm gap when triggered. The gap would breakdown
without external influence at less than 25kV. The voltage/distance
you choose here will depend on the HV capacitor you use, and the
capabilities of the flyback transformer you have. Remember that most
flybacks are only designed to handle 30kV at the most, so stay below
this if you want to get the maximum lifetime from your setup. When
positioning the leads on the test tube, I found special care had to
be taken in order to achieve high reliability. The HV electrode
should be wrapped around the tube in a circular fashion, this reduces
corona jets which would others pre-ionize the gap. The ground
electrode and trigger electrode should be spaced apart, this is so
the trigger pulse ionizes the glass between them. If the trigger
electrode is placed directly under the ground electrode, it will be
difficult to trigger a spark. The spacing between the trigger and HV
electrode aren't critical. In the images above the ground and HV
electrodes were swapped, otherwise they can be used as a guide.
The
high voltage capacitor is a critical part of the air flash unit. In
addition to a relatively large capacity at such voltages, it also needs
to be pulse rated so the spark forms and decays quickly. In an attempt
to save myself 80USD, I built my own high voltage capacitor. It uses
seven rolled capacitors, of roughly 2nF each. They are built using
nothing but duct-tape, tin foil and overhead transparencies. To keep
them isolated from their surroundings, I placed them in the PP-pipe
seen above. Four sheets of transparency were used between layers,
giving it an approximately 40kV voltage rating. Instructions on constructing the capacitors can be found here.
The
high voltage charging circuit is a current-mode controller which limits
the primary current to about 5A peak. By adjusting the "High Voltage
Adjust" potentiometer, you can vary the primary current from zero to
max. In effect, this varies the output voltage. The supply voltage for
the capacitor charger must be 18V if using a UC3842. This is because
the chip has built in UVLO, and the IRFP450 can't have more than 20V on
it's gate. If you use the UC3845, you can get away with 12V supply
voltage for the capacitor charger. The flyback transformer is of the
standard type found inside of old CRT monitors or TVs. It must be
rectified to charge the HV capacitor. Since the power requirements are
so low, I found it sufficient to use the internal primary. If using a
low supply voltage, such as 18V, an external primary may be required.
See my flyback driver pages for more details. The HV capacitor needs to
be at least 15nF and 35kV, more capacitance gives more spark
energy, and thus a brighter flash. This capacitor will be the most
expensive component if purchased.
The trigger circuit works by
discharging a small capacitor through a pulse/trigger transformer. This
creates a tiny spark, similar to touching a door handle charged with
static electricity. By placing a conductive object inside of the flash
tube, and triggering the pulse transformer, the outside of the tube is
ionized. This breaks down the air and causes the HV capacitor to
discharge. The trigger transformer I used was a CD45 purchased from Information Unlimited.
The supply voltage of 125V was acquired using a voltage multiplier
on the 40V AC used to power the capacitor charging circuit. 40VAC was
acquired using mains transformer found in an old stereo. To trigger the
pulse transformer, simply short the two trigger leads together. If
using a transistor to do this, keep in mind that the voltage on the
collector will be about 25V when open circuit.
Disclaimer:
I do not take responsibility for any injury, death, hurt ego, or other
forms of personal damage which may result from recreating these
experiments. Projects are merely presented as a source of inspiration,
and should only be conducted by responsible individuals, or under the
supervision of responsible individuals. It is your own life, so proceed
at your own risk! All projects are for noncommercial use only.
This work is licensed under a
Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
