Gas pipes
4.2 Experimental set up
4.2.1 Cylindrical shocks
A very good cylindrical shock can be obtained by exploding a piece of thin wire, as described in section 3.1.2 on page 73. The idea is to store energy into a big capacitor and then discharge it quickly through the wire and ensuing plasma using a triggerable spark gap (or a Thyratron, if available). The circuit is simpler than in the case of the CSL, because it does not require any coil (see figure 3.9 on page 88 and 3.1 on page 72). In fact, the voltage is carried to the spark gap by the wire itself. The absence of the coil makes the experiment less noisy from an electromagnetic point of view.
One critical factor for this experiment is the resistance of the wire. If the resistance is too small, the plasma forms very quickly, when there is still plenty of energy in the capacitor. In this case, once the plasma has formed the resistance drops to almost zero and the circuit oscillates radiating away a large amount of energy. The optimal resistance is the one that allows the circuit to dissipate most of the energy into the wire during the first cycle of the oscillation. The resistance of the wire depends on the material and on the thickness. If the wire is very thin, the resistance can be very high and
also the amount of debris is reduced.
In our experiments we used a copper wire 150 ~m thick. For the eight- shocks experiment described in section 4.3 on page 104, the wire was 50 cm long, with a total resistance of 8 D.
Screen Pin Hole Shocks
Figure 4.1: The diagnostic setup for the pipe experiment. From the left, the two lenses used for the telescope have a focal length of 5 cm and 50 cm. The lens used for the "spatial filter" has a focal length of 10 cm.
For the diagnostic of the shock waves we used a nitrogen laser, described in chapter 6 on page 169, of the square electrode type. The laser beam is sent through the pipe and produces shadowgrams of the shock waves. The refractive index of the shocks is so high that they just refract the light out of the collecting optics. Since the exposed film is a "negative," the regions corresponding to the shocks therefore appear "white," against an undisturbed
"black" background.
The set-up of the experiment is shown in figure 4.1. The small telescope on the right is necessary to compensate for the divergence of the laser beam.
In this experiment, in particular, the beam has clearly to be parallel and aligned with the pipe. The "spatial filter" on the left was used to eliminate the light of the explosion, which is very bright. In addition to that, when
Figure 4.2: A SEM image of a piece of the wire used during the experiment.
It can be noted that the wire has a thickness of about 130 \-lm. In order to be sure that the thickness of the wire was uniform, we took several images of different pieces of wire.
the photographic paper was very close to the device, a UV filter was used to let through only the light of the laser. The single lens detector also offers the possibility of changing the magnification of the image only by moving the detector, i.e., the photographic paper. This is true only if the detector is in the near field, where the effects due to the lensing of the pipe can be neglected. The magnification itself can be measured afterwards by imaging an object of known size.
Our first unsuccessful attempt to create a pipe was made by exploding a coil of wire. Unfortunately, the wire is so thin that it needs to be supported.
In our case, we used a cylinder of very thin paper, on which the wire was wound. The idea was that the paper would have been blown away by the explosion without affecting the shock waves, but the first experiments showed only a very strong shock coming from one side of the experiment, from the bottom in figure 4.3 on the next page. What happened was that the discharge jumped across the coils, instead of flowing through the whole wire, probably helped by some skin conduction effect due to the paper.
The second attempt was made using six short pieces of wire connected
III parallel. The wires were held between two metal disks, to which the high voltage was applied. With this approach, we could get all the wires to explode, but not at the same time. In this case, the formation of the pipe does not take place because the shock waves arrive at different times. This system may probably work, but one has to be sure that the wires are strictly identical in both length and thickness. In our case, we did not have any
Figure 4.3: The single strong shock coming from the coil, due probably to the jumping of the arc across the coils. The shock is travelling toward the top of the picture. Turbulence left behind by the shocks is clearly visible.
This is a clear sign of the energy being too high. The white circle is the cylinder of paper used to hold the wire, that is usually blown away by the explosion. It should be noted that all our pictures have a negative grayscale, so the highest light intensity corresponds to the blackest parts of the image.
means to tighten the wires, which had therefore a slightly different length and even small kinks.
The third way is the only one which allowed us to obtain some good results. Instead of using separate wires, we used only one long piece, going up and down between two insulating holders. In this case, the explosion has to occur everywhere at the same time. With this technique we obtained some good quality pipes as long as 2.5 cm. In principle, given enough energy, there is no limit to the length of the pipe. The main problem, in this case, was that our particular holder was not built with enough precision, leading to some differences in the arrival time of the shocks, which travel at speeds approaching 1.5 times the speed of sound.