To: E614 collaboration
From: Robert Henderson
Date: 11 Jan 2000
Re: Design of Cradle and Track

                       Design of Cradle and Track
                       ==========================

  The design of the Cradle/Sled/Track had undergone considerable change. In
particular, we dropped the sled entirely and adopted a Cradle/Track design
that used curved sections. Currently the track covers about 120 degrees and
the cradle about 140 degrees, with two thick longitudional ribs.

  The plan has been that the design process would pass to the TRIUMF design
office for refinement and detailing. However, Guy Stanford has looked at the
design with a `fresh' viewpoint and proposes a far simpler and lighter
designs for the Cradle and track. I will not discuss the schemes for aligning
and locking the track to the magnet at this time, the concepts for the cradle
and track inside the magnet are of greatest interest at the moment. We should
keep in mind that these are only concepts at the moment.


                              New Concept
                              ===========

  Guy was NOT impressed with the thick curved memebers of the cradle and
track, since they are very heavy and have poor moment of inertia. Remember,
that the moment of inertia of a beam is proportional to W (where W is it's
width) and proportional to L**3 (where L is it's vertical extent). My posted
figures (LAM-50-7.EPS and LAM-50-8.EPS) do not quite have Guy's specified
beams, his were infact 10" high and actually hit the detector edge. Also the
flat bottom of his cradle interferred with the gas inlets. I shortened the
cradle and track beams one inch (in the vertical direction) and lowered the
cradle bottom, but the concept is unchanged.

  The two track beams are connected together beyond the ends of the solenoid.
They are attached to the solenoid via alignment and locking apparatus that
are not detailed yet. Again, such things are common to any scheme.

Cradle
------

  With the above in mind, Guys cradle concept is based on two aluminium angle
beams, 9 inches high, 4 inches wide and 1 inch thick. These beams have their
inside vertical edge about 305 mm from the center line, which puts them
outside the detector `feet' at +/- 45 degrees. The horizontal lips of these
beams continue towards the outside edge (see LAM-50-7.EPS and LAM-50-8.EPS).

  The cradle has four cam rollers (or groups) at the corners, from these it
is simply supported on the track beams. The trach beams have indents for
these rollers. How the cradle finally comes in for an aligned `landing' on
the track  has not been specified. Such issues are common to any cradle/track
scheme, so are not discussed in this posting.




Question 1) It is obvious the horizontal lips could infact extend another
----------- 2.75" in that direction!? Why not do so? 

Answer  Guy's idea is to mill these beams from bars of aluminium. Making it
------  wider means machining off far more material. Also, the strength 
        difference between 4" wide and 6.75" wide is only 0.37% stiffer,
        but is 23% heavier. In LAM-50-7.EPS and LAM-50-8.EPS, I have added a
        light-weight 2.75" wide bolt-on extension to the lips.


Question 2) The corner of the cradle beams comes close to the edge of the
----------- detectors! How do the gas lines pass by that point.

Answer  Good question! The gap there is about 6 mm in my drawing. The edge of
------  the cradle beam is champhered. We can certainly increase this gap to
        say 10 mm, enough for a 6.4 mm diameter hose. We also have the option
        of bringing these polyflo lines into the gap between detectors, i.e.
        attach to 1/4" G10 plate of each detector with cable ties. The short 
        UV-module has the citals extend 12.6 mm outside the G10 face, so the
        hoses have quite a lot of room. Once past that corner, the hoses come
        out again and enter the cable trays.


Question 3) How does this new cradle design effect the supports for each
----------- detector?  

Answer  In LAM-50-8.EPS, the detector feet are `floating', i.e. not track or
------  spring bumpers are shown. But clearly there is as much (or more) room
        available for them as in the existing scheme. With the new cradle
        design, the support system would be bolted to the inside edge of the
        9 " beams. Guy and I have been discussing detector support/alignment
        schemes, but they are not ready to report yet.


  The present cradle is about 40 mm thick. The 9" beam is 5.72 times as high,
so is 186 times as stiff for the same width. Of course, the 140 degree curve
of the cradle adds some stiffness to, but the stiffness advantage of the 9"
beam is obvious. The weight saving of the cradle would be very large. The two
beams weigh only about 31% of the existing curved cradle and longitudional
ribs. This ignores the cradle endcaps and any gussets at the ends. Since the
detector stack weighs only a fraction of the present cradle weight, the
reduced cradle weight further reduces the cradle deflection.

  The flat bottom of the cradle does NOT need to be solid. Several angled
struts would lock the two sides beams together, and of course the cradle
endcaps also lock it together. A thin sheet of aluminium across the bottom
would provide the helium gas seal. The helium covers over the top of the
detectors would need to seal to the lips of the side beams, just as they
connect to the present cradle longitudional ribs.

Track
-----

  In this new concept the track is also greatly simplified. Instead of being
a 50 mm thick arc of aluminium covering 120 degrees, it becomes two
rectangular beams 5" high and 2" wide. The new design weighs only 25% of the
present design! With the cradle weighing a lot less, track is also not loaded
so much. If we wanted more stiffness from these track beams, there is space
to `angle' the bottom of this beam. That is, to increase them on the bottom
inside edge. The bottom inside edge could be 7.5" high, reducing to 5" at the
outside edge. This would double their moment of inertia and increase their
weight only 25%. We should remember that track deflection is of little
importance during cradle insertion/removal. When installed the cradle would
be supported only at the ends, where the track is strongly supported.


                                My Conclusions
                                ==============

  I believe this new cradle/track concept is a definate improvement over our
present design. I suggest that Jan do an ANSYS analysis to confirm it's
perfomance. In addition to being a lot lighter, the new designs would
undoubtedly be far easier and cheaper to fabricate.

  Ziggy is at present finishing the magnet design. He should be available to
work on the cradle/track design in about three weeks. I am introducing the
raw concepts now, in the hope that we cane look for problems etc and perhaps
be in agreement when Ziggy is available.

Regards Robert Henderson

Filename: lam-50-7.eps

Filename: lam-50-8.eps

Created for the The Center for Subatomic Research E614 Project Projects Page.

Created by The CoCoBoard.