HATHAWAY                                                                                                                     1080 19th Sideroad CONSULTING                                                                                    King City, Ontario

SERVICES                                                                                                                   Canada L7B 1K5

 

Lab: (905) 795-9382

Lab FAX: (905) 795-8951

Home: (416) 923-8586

 

 

Notes to accompany TAJMAR MkI sketch                                                               Jan 9, 2008

G. Hathaway

 

            This is a representative sketch (see below) which tries to incorporate several possible improvements to Tajmar’s original. The sketch only shows the side view. A top view would indicate that the main drive shaft is eccentric to the axis of the main outer dewar to allow extra radial length for sensors. The scale is mainly to show relative dimensions and is not meant to be final. For instance, it would probably be desirable to increase the lengths of the dewars to hold more cryogen. Although this design incorporates what I consider design improvements, some may be far too costly or intricate to actually build. It is hoped that the ideas presented will assist in the final design.

 

Key

 

A.                Air Motor. We have two here at the lab which are about the size shown. They are older units made by Gast and ours are rated at 3000 rpm max at 61 in-lb. Others available in the Gast series are 6000 rpm @ 6.0 in-lb (Gast Model #1UP), for example.

 

B.                 Upper Support Plate. This is a custom-made thick, heavy steel structure typically resting on 3 massive legs bolted or not to the lab floor. It is designed to be rigidly fixed to the Dewar Top Plate (L) via thick rods (E) (which pass through but not touching Sensor Support Plate (F)) which in turn supports the Vacuum Can (O) by means of Connecting Rod (M). Air Motor (A) is mounted on the Upper Support Plate (B).

 

C.                 Upper Coupler. This coupler connects the shaft of (A) to Steel Drive Shaft (H) to allow for slight misalignment of the shafts.

 

D.                Upper Bearing. This bearing is typically ˝” or 5/8” dia and supports a collar to take the weight of SS Drive Shaft (H). On the top of the Dewar Top Plate (L) is another similar bearing which has a small o-ring to seal in the He gas. This lower bearing & o-ring may need a small heater as the Dewar Top Plate (L) will likely become quite cold, probably ~250 K I would guess.

 

E.                 Dewar Support Rods. These (threaded) steel rods keep the Upper Support Plate (B) and Dewar Top Plate (L) rigidly connected.

 

F.                  Sensor Support Plate. This is a triangular (likely), thick, very heavy, probably welded steel anti-vibration table which is supported on 3 anti-vibration Pneumatic Legs (G). The Sensor Support Plate (F) supports Sensor Platform (P) only via Sensor Support Tubes (I). Thus the Sensor Platform (P) is mechanically decoupled from everything else (including the Vacuum Can (O)). If we could get a suitably sized triangular 4” thick optical table, this would be ideal.

 

G.                Pneumatic Legs. These are standard optical table legs of which 3 are required. They rest on the lab floor and kill vibration to the Sensor Platform (P).

 

H.                Stainless Steel Drive Shaft. This shaft connects Air Motor (A) to the main Fiberglass Shaft (R) via a second coupler at its lower end which connects to a steel shaft inserted into (R). This design minimizes the length of the Fiberglass Shaft (R) and also minimizes heat flow down from the top.

 

I.                   Sensor Support Tubes. These rigid thin-walled stainless steel tubes (or can be rods) connect the Sensor Support Plate (F) to the Sensor Platform (P). They are at or near RT and they are in vacuum.

 

J.                   Inner Bellows. These 3 (of which 2 are shown) welded SS vacuum bellows allow the vacuum to be maintained in the Vacuum Can (O) while mechanically de-coupling it from Sensor Support Plate (F).

 

K.                Outer Bellows. These 3 plastic/silicon bellows keep the He gas from escaping from Outer Dewar (Z) into the atmosphere while allowing mechanical de-coupling of it (and Vacuum Can (O)) from Sensor Support Plate (F).

 

L.                 Dewar Top Plate. This is a thick Al plate which supports the Vacuum Can (O) and is rigidly connected to Upper Support Plate (B). It forms the lid to the Outer Dewar (Z) and is sealed with an o-ring.

 

M.               Connecting Rod. Rigid rod supports Vacuum Can (O) under Dewar Top Plate (L).

 

N.                Upper Heated Bearing. This small bearing will be below RT but likely above 200K so a small heater may keep it in reasonable shape. It is supported by a thin steel tube extending down from Dewar Top Plate (L).

 

O.                Vacuum Can. This thin-walled SS container is evacuated to keep the sensors from freezing. Its top lid is removable for placement of sensors within.

 

P.                  Sensor Platform. This is a rigid platform on which the various sensors are placed. It is only connected (rigidly) to Sensor Support Plate (F), thereby minimizing vibrations.

 

Q.                Upper Labyrinth. This is a thermal labyrinth of thin Al disks which prevents radiative losses down along Fiberglass Shaft (R). Half of the disks are attached to the spinning shaft (R) and the other half are rigidly fixed to a can which is supported at the top by the Vacuum Can (O). This also may prevent some cold He gas from creeping up the shaft and further cooling the bearing (N).

 

R.                 Fiberglass Shaft. This 1” (nom) fiberglass, possibly hollow, shaft supports the Drive Wheel (U) and prevents heat loss by conduction from the cold area. It has heated bearings at both ends via short SS rods inserted into its ends.

 

S.                  Superconductor Ring. Here is where the SC sits.

 

T.                  Sample Holder. This castellated-bottom, copper sample holder is designed to closely mesh with, but not contact, a similarly-shaped copper Cold Finger (V).This allows efficient radiative, as well as some conduction cooling of the ring (S) while not actually contacting/stirring up the LHe. It is designed to shrink fit around Drive Wheel (U) when cooled while remaining concentric.

 

U.                Drive Wheel. This stainless steel wheel is comprised of a central hub, which fits on shaft (R), and spokes which support a SS rim onto which Sample Holder (T) shrinks. The spokes reduce heat conduction to the sample (S) from shaft (R) and lighten the wheel.

 

V.                Cold Finger. This is a castellated copper disk which meshes with Sample Holder (T) without touching. It has several copper “fingers” extending from its bottom surface down into the LHe bath so that as the LHe level drops, the top of the Finger remains substantially at 4-5K.

 

W.               LHe Bath. Liquid Helium is stored in LHe Dewar (X).

 

X.                LHe Dewar. This custom-made LHe dewar may be made from “NASA” PU foam if we’re lucky, or using conventional design. It is in the shape of a stepped annulus as shown in the sketch. The step accommodates the lower shaft bearing and labyrinth. However, it may be possible to simplify the design by eliminating the step with the disadvantage of lengthening shaft (R).

 

Y.                Thermal Baffles. These thin stainless plates minimize radiative losses from the lower part of the apparatus and also conduct cold He gas away from the underside of the Vacuum Can (O).

 

Z.                 Outer Dewar. A conventional wide-mouthed liquid nitrogen (LN) dewar to reduce overall costs. However, it may need to be LHe capable – this needs more work. Also for additional cooling, an eccentric LN bath may be included in the space between the LHe Dewar (X) and the inner wall of (Z).

 

Some advantages of this design are:

 

1.                  Maintains essential elements of Tajmar experiment.

2.                  Bearings top & bottom of main shaft.

3.                  Bearings at >>4.2K

4.                  Bearings in He gas (not LHe).

5.                  LHe level can drop but cooling continues.

6.                  More efficient cooling of sample while rotating due to castellations.

7.                  LHe surface not whipped up by nearby rotating plate as in Tajmar’s design.

8.                  Generally improved thermal design plus possibility of additional LN cooling minimizes LHe consumption.

9.                  Eccentric design (I’ll say!) allows more in-plane radial measurements.

10.              Sensors attached only to anti-vibration platform.

11.              No need to fix assembly to ceiling or to have sandbox.

12.              Shortest length between sensor and anti-vibration table.

 

Some disadvantages of this design are:

 

  1. General complexity and parts count.
  2. Requirement for custom fabrication – eg. Upper Support Plate (B), Sensor Support Plate (F), LHe Dewar (X), Sample Holder (T), etc.
  3. Difficulty in assembly & disassembly.
  4. Difficulty in filling LHe Dewar (X) and general cool-down procedure.
  5. Weight (and thus moment of inertia) of Sample Holder (T).

 

 

It may be possible simplify the design by getting rid of the entire LHe Dewar (X), Cold Finger (V) and complex bottom of Sample Holder (T) and replacing with a simple, larger bath of LHe like Martin did. The sample cooling would probably not be as effective. This would necessitate removing the lower bearing and lower shaft coupling so that shaft (R) is “hung” from upper bearings only. However, shaft run-out may be a big problem at high speeds.