Design Aspects and Fabrication Techniques

1
Design Aspects and Fabrication Techniques

• Aluminum versus stainless steel
• Stainless will avoid oxidation patch breakdown at
  high rate?gain
• No affordable surface treatment
• We should procure a hot Ruthenium source to test
  aluminum at high rate.
• Still concerned about Alum in light of Errede
  Lyons experience
• Could contemplate thinner walls to save on weight
  and material costs
• Thin walled stainless much tougher to retro
  machine.
• Choice of wire diameter
• 90 ? dia wire was choice of SDC since easier to
  string (more visible)
• Because of tension-area scaling, dia irrelevant
  to gravity sag and EM stability
• Appears that no wire support will be necessary
• Higher diameter means higher voltage (Harder to
  distribute and avoid corona)
• Larger diameter means a steeper log(gain) versus
  V/VT curve
• More critical operation according to Sauli
• Suggest 25 ? - 50 ? dia. Gold plated tungsten?
  Thin may be more sensitive to mechanical
  tolerance , shipping vibrations etc, and harder
  to crimp.

2
Multi-tube extrusions versus single tube arrays

• It is not obvious to us which is easier to
  implement given apparent problems with extrusion
  venders making thin walled rectangular tubes.
• As walls thicken, maintaining constant wire pitch
  between butted planks gets increasingly
  difficult.
• The option of 2? webs and 1? outer walls may be
  difficult owing to differential cooling of
  extrusion.
• This leaves us with the differential gas volume
  option.
• An open, Ioracci comb design would be very easy
  to implement if we could get it to work and
  survive the cracks necessary to maintain a gas
  seal.
• There are mechanical advantages to a round tube
  close pack if we can survive the packing
  fraction dead space and inefficiencies near the
  cell walls since we have only space for two
  layers per view.
• Mounting round tubes should not be a problem with
  CNC endplates.
• Gluing a set of rectangular tubes together may
  also be a viable option.
• A closed, rectangular extrusion still seems an
  open issue to us, but this is the concept we have
  been working on.

3
Thoughts on gas flow
We believe that a parallel gas flow arrangement
between two reservoirs may be the most effective
and simplest distribution method. A series
method may be possible as well but requires more
fittings and we dont see advantage to it.
4
Thoughts on End Cap design

• The end cap which holds the wires also acts as
  the gas manifold.
• Manifold is in a channel below pins.
• End cap injection molded plastic with several
  pulls to create cavities
• Actual brass crimp pin is held in by a delrin
  insert to avoid charge buildup
• Rounded edges avoid corona
• Free space above the pins to mount electronics
• Placement of gas inlet is a concern.
• Try to avoid arcing by getting too close to wire
  or brass pin.

• If the end cap must serve a double layer of
  tubes, price will increase significantly, if
  possible at all. We also lose ability to stagger
  units in along wire direction for gas inlets.

5
Stringing techniques (Closed Extrusions)

• End caps glued in but two pin holders not
  inserted.
• Thread wire through spool end pin. Attach wire
  to clip. Drop slug taped to wire through dowel
  hole.
• Pull to pulley with magnet and exit dowel hole.
• Thread through brass crimping pin on pulley end
  and press fit both dowels into end caps.
• Run full length of clean wire through both brass
  pins to avoid dirt and kinks.
• Crimp at spool end
• Tension from pulley using a weight.
• Crimp pulley end.
• Slide pulley, spool to adjacent cell.
• Repeat as needed. (67,000 times!)

Simultaneous stringing?? Probably not worth
effort of special jigs etc.
6
Beam test prototype?

• Full injection molded end cap seems too ambitious
  for early summer 99 run.
• May involve significant lead times as well.
• Machined prototype with gas channels too
  difficult for more than 10 planks
• Propose using single tube plugs with gas fittings
  for serial gas flow
• Labor intensive due to individual gas fitting.
• A good solution for 100 tubes prototype but not
  67,000 production run
• Could be jobbed out to CNC machine shop for less
  than 2 per unit.

Down side is we wouldnt be testing actual
detector.