Progress Report of Region3


1) VDC progress

Talked again with Stesalit about segmenting the frame into 4 pieces like Roger and Greg suggested it. The material cost for a 1 piece frame

is ~1680 Euro , a 4 piece frame will cost ~1500 Euros + ~300 Euros for machining chamfers for the overlapping region.

The cutting precision is only +- 2mm in length and +-0.1mm in thickness. Anyway the 4 pieces then have to be machined

before and after gluing. Not sure if we could glue then at W&M at all ( insufficient lasertable flatness, jigs to be constructed)

Beginning December I will order the raw frames from Stesalit (lead time: ~6 weeks) include some (2?) spare frames. The spare frames will be send to

Atlas Fiber and Schmidt Company for test runs.

2) Wire Scanner

The stepper motor controller works fine but it does not support the readout of the linear encoder like we thought. The linear encoder (LE) is basically a

precision edged glass scaler with LED’s and photodiodes counting the gratings while moving. There is a reference mark in the middle of the LE, so you have to

count the gratings to the left or right relative to this “home” position . The intrinsic resolution of the LE is about 4um combining 3 LED/Photodiode

readout positions while one stepper motor step is about 25um. At present we are checking out if we should buy a better controller or a special PC scaler card .

Both solutions are supporting this type of LE and cost ~$800. The scaler card can be read out by LabView and allows an direct access of the LE while a controller would handle

the counting of the gratings internally while moving).

3) Garfield Simulation

The author of Garfield has released some weeks ago a new version of Garfield (v.9). Main difference to the older versions is the incorporation of the newest Magboltz

Version (v.7 instead of v.3) . Magboltz computes for electrons, the drift velocity, the longitudinal and transverse diffusion coefficients as well as the Townsend and

attachment coefficients needed for the electron avalance close to the wire. Major Magboltz change included now the most recent (in)elastic cross section of the gases

which were not available to earlier Garfield/Magboltz versions

We recalculated the gas file for Argon-C2H6 50:50 and rerun the signal and drift line garfield scripts. Below are the results for a typical track

through one wire plane (left/right: HV plane).

#1 : Drift line plot of track (loer left corner to upper right corner). The green dots are the clusters, the yellow lines are the drift electrons

#2-5 : Induced current on the wire (yellow), induced current on the HV plane (magenta)

#4 : central wire

In addition Garfield allows to calculate the number of clusters (mean: ~165 with mean: ~1.9 electrons per cluster).

The total energy loss of the track for one (out of 4) wire plane is ~8keV

4) Frame quality control project

For measuring the flatness of the frames as a part of a quality control, we will adapt a system used by ATLAS. The location/position of ATLAS straw tubes is

measured along ~2m with an precision of ~<10um (http://dilbert.physics.wm.edu/elog/Construction/168) . Key element is a “flat” beam laser and a CCD chip

measuing the location of the blocked light. Depending on the analysis of the shadow boundaries by simple threshold crossing or interference pattern, the position

resolution is ~10um or even ~0.1um.

David Armstrong suggested a cheaper version by using a simple laser and photodiode. But first back of the envelope calculation indicate that a regular elliptic/round

laser spot with a typical 10% power fluctuation will make it impossible to resolve 50um and better at all.

The advantage of the laser system offered by the german company Schaefter + Kirchhoff (http://www.SuKHamburg.de) is that it can be used to measure the absolute thickness

variation (measuring perp. to frame) besides the surface flatness variation (moving a gage block) along the frame.

That's all folks !