Rutgers University Experimental Nuclear Physics

Straw Chambers

Contents

• Credit
• Introduction
• Straw Geometry
• More on gas gain
• Straw References

The straw chamber design described here is based on a design originally created for the EVA cylindrical straw chamber, experiment E850 at Brookhaven National Laboratory. The best writeup on this detector of which we are aware is Brookhaven Informal Report EP&S 91-4, by M. Kmit, M. Montag, A. S. Carroll, F. J. Barbosa, and S. H. Baker. We are particularly indebted to F. J. Barbosa and M. Montag for many helpful discussions.

A straw chamber is a set of cylindrical tubes, or straws, with a thin wire running along the central axis of each tube. Our chambers have the wire at positive high voltage relative to the straw. When a charged particle passes thorugh the straw, it ionizes gas atoms inside the straw. The high voltage causes electrons to drift towards the central wire. The mean free path of electrons in a gas can be estimated from the size of the gas atoms and the gas density; lambda = 1/sigma*rho is of order a micron (10^-6 m). The effect of ``continually'' colliding with gas atoms is that the electrons drift with aboout constant velocity towards the wire, for a large range of electric fields. When the electrons get near the central wire, the large field gradients of the 1/r electric field accelerate the electrons enough between collisions with gas atoms to ionize additional gas atoms, leading to an avalanche and a gain, an increase in the number of electron ion pairs. The electrons are collected on the anode wire, and the ionized gas atoms and molecules drift to the outer cathode, leading to a signal on the wire. The signal is read out into a computer for analysis, to determine the position of the particle passing through the chamber.

The gas mixes that we have used, Ar-ethane (C₂H₆), Ar-isobutane (C₄H₁₀), and Ar-CO₂, have an electron drift velocity of about 50 microns per ns. This leads to a maximum drift time of about 100 ns for our straw radius. The Ar and other gas ions drifting to the cathode are heavier than electrons by a factor of order 10⁵, and thus require a few tens of microseconds to reach the cathode.

RUNP uses straws that are about 0.418 inch / 0.522 cm inner diameter. The straws are built by wrapping a 10 micron thick aluminum foil and two two-mil thick mylar layers plus heat setting glue around a mandril. A brass ferrule (doughnut shape) is inserted into each end of the straw to allow ground connections to be made. A delrin insulator / feedthrough is positioned in the ferrule. The delrin has a center hole to position a pin with a slot, and three gas holes around this center hole. The brass pin is sawed with a 4 mil wide 30 mil deep slot so that the central thin wire is positioned and soldered in place. The wire is 1 mil diameter LUMA wire, gold plated tungsten with a few percent rhenium added. The mechanical tolerances add to a precision of about 3 mils (sigma) in centering the wire in the straw.

We have measured gain as a function of applied high voltage for two gas mixes, 50/50 Argon/ethane and 80/20 Argon/CO₂. We used an ⁵⁵Fe source, which emits 5.9 keV X rays, to provide a known energy deposition in the chamber. (One can see from the ~ 1 micron mean free path of electrons that the energy is deposited - roughly speaking - at a point.) From Sauli's report, CERN 77-09, we expect about 70 primary ionizations and 220 total ionizations for this energy deposition in these gas mixes. We measured the pulse shape and size at the input to our readout boards in a fast analog oscilloscope, and integrated to get the total observed charge. Dividing by the expected 70 primary ionizations gives an approximate gain. The measurement is good to perhaps 20%.

Our results for the gain of the Ar-CO₂ and Ar-ethane were:

In small test chambers, for which wire deflection is not a problem, we have found that increasing the HV for Ar-ethane to 2200 V leads to a sharper rise in the drift time spectrum and to improved resolution. For the long CEBAF FPP straws, with deflections ~ 100 microns and proportional to V², this might not be the case.

There are numerous references to straws in both refereed literature and in preprint form. These include the following:

The list above is not comprehensive; it includes almost exclusively circular tubes, with the exception of the LHC/SSC prototype hexagonal tubes. The large amount of work done with rectangular drift tubes has been omitted.

Created May 2, 1996 Norma Lucero