These pictures show the final design of the new Booster collimators and shielding as of 3/12/03. The stand design is almost finished, and most of the details of the shielding and vacuum liner are worked out. The main things missing from these pictures are the LVDT readback elements and the limit switches. Pictures of these will be added as the design is finished.

The basic design concept is that by eliminating a large air gap between the collimator and shielding present in the previous design, less shielding is required. Air activation is lower and the total amount of shielding steel required is lower.

In this design, because of the tight integration of the collimator and shielding steel, both the collimator and the surrounding shielding move. The actuators have to be sized to moved the 11.6 ton block, a typical weight for remotely operated magnet stands. Many of the components in this design were first developed and tested for E760 at FNAL and the OOPS experiment at Bates.

Click on any of the thumbnails to get an enlarged view. You are welcome to download any of the images. If they are used for other than private viewing, credit to Bartoszek Engineering would be appreciated.

Overall views of the collimators in Long 6

These views show a section of the Booster tunnel at Long 6 in solid model. The view looking upstream along the beam pipe (right picture) shows that some of the unistrut supports for pipe may have to be shortened to keep them from interfering with the whole range of motion of the collimator. When the collimators are centered on the beam they have a 3" square aperture making them look like a length of beam pipe with no aperture restriction inside that area.

The bellows have been designed to allow simultaneous horizontal and vertical translations of 1.5 inches each, a total lateral offset for the bellows of 2.12 inches. The middle section of beam pipe between the collimators will have to be supported from either the floor or wall.

Close-up of one collimator and its stand

The collimator stands have four degrees of freedom. They can translate vertically and horizontally by +/- 1.50 inches, and they have a yaw rotation and a pitch rotation. For safety, the yaw and pitch axes are limited to a total rotation of +/- 10 mradians. Making the movements all plus and minus allows us to build all three of the collimators identical, minimizing the possibility of improperly installing different devices.

The assembly scenario envisioned is that the 3 ton stands would be lowered down the Booster access hatch in the cross gallery. The lower block and vacuum liner, 6.3 tons, is then lowered onto the stand, and finally the upper block, 5.2 tons, is lowered onto the lower block. The whole structure will be mounted on Hilman Rollers so that it can be rolled down the Booster tunnel and installed in the beam line. The Hilmans are temporary just to make the move to the beamline.

Each of the axes will have a "manual" or local mode of operation so that initial alignment of the collimator to the beam-line can be done.

The shielding blocks and their assembly

The thickest plate of plain carbon steel commonly available is 8 inches thick. The upper and lower blocks are welded together plates of 8 and 4 inch thick slabs. The blocks will be blanchard ground flat to minimize cracks, and some small notches have to be milled to make room for the larger block on the upstream end of the vacuum liner.

Not shown are any fasteners between the upper and lower blocks, between the lower block and the stand, or any lifting points for handling the blocks. All of these will be added soon. The thermal calculations have shown that differential heating of the vacuum liner caused by the beam hitting it primarily on one side produces very small bowing deflection and forces that are easily handled by welds and bolts.

Also not shown are thermocouple probes that will be attached to the liner at the upstream end. These will monitor the temperature of the liner in the case of a beam mis-steer.

Details of the vacuum liner

The 3 inch by 3 inch square opening over the downstream 40 inches of the collimator (darker blue color) is fabricated from two 4 x 4 x .38" stainless 304 angles welded together at two opposite corners. To minimize a hot spot at the upstream end of the collimator, and to move the radiation further into the shielding volume, the upstream 8 inches of the collimator (light blue in the first picture and transparent gray in the second) is tapered larger by 2 centimeters vertically and horizontally. The second picture shows the taper of the block. This piece is currently imagined to be fabricated from a 4" x 4" x 8" bar of stainless steel cored out and cut to the required taper.

Details of the Stand Starting with the Vertical Drive System

The vertical drive consists of four Simplex J10 ten-ton jacks with the traveling nut configuration driven by a single motor/reducer. All of the screw jacks on this stand are equipped with the optional stop nut to prevent them from being driven too far in the event of an electronics or software failure. All of the motors on the stand are radiation-hardened NEMA 42 frame stepper motors from Empire Magnetics. The vertical drive motors have to be driven with 170V drivers to provide enough power to drive this axis.

The reducers are all Hub City single reduction worm gear reducers. In the case of the vertical drive, the reducer is a Hub City Model 261 with 40:1 reduction. It drives two Hub City Model 150 bevel gear boxes that deliver the torque to the input shafts of the screw jacks. Every coupling is keyed and most have set screws as well. The vertical speed of this drive train is .63 inches per minute.

Each of the four jacks is coupled to the splitter bevel gear boxes by roller chain couplings. The advantage of these couplings is that they do not require the bevel gear box to be moved to take out any screw jack. Also, with the chains removed, each screw jack can be individually adjusted to take its share of the load. This will be part of the manual alignment of the stand once it is moved into the beam-line. The 10 ton capacity of the jacks means that two diagonal jacks can take the entire load of the collimator without damage during the initial setup.

Pictures of the position readback LVDTs and limit switches will be added as these features are designed into the stand.

Details of the Horizontal Base Plate Assembly

The horizontal base plate assembly translates up and down as the vertical drive system screw jacks are actuated. It also contains the motor, reducer and screw jack that push the next layer up on the stand horizontally perpendicular to the beam-line. This jack is a Simplex J5 jack, traveling nut style, with 5 ton capacity. Details of this drive can be seen two sections below in the underside view.

The square thin plates near the corners are aluminum plates that have a Dicronite (tungsten disulfide) surface coating. This coating has a published coefficient of friction of .03 and has been used in radiation environments at Fermilab before.

The bronze colored oblong shapes are Boston-Bronz keys to guide the horizontal motion. They are fitted to keyways in the plate above.

There are four blue flanged bushings shown assembled to this plate. These actually engage slots in the plate above to prevent those two layers from separating vertically. They are loose fits in the horizontal directions to not interfere with the guidance of the keys. This should become clearer in the next few pictures.

The Horizontal Base Plate Assembled to the Vertical Drive

This picture shows how the horizontal base plate attaches to the vertical drive system. In these pictures, the J10 traveling nuts are shown bolted to the horizontal base plate assemblies. This is only one possible way to assemble these sections. A more likely approach would be to leave the nuts on the J10 jacks and lower the horizontal plate down to them.

As you will see in the progression of these pictures, the load path from the steel shielding to the floor is only through the upstream and downstream ends of the stand. Every effort was made to minimize bending in the plates and transmit the loads in pure compression through the thickness of the 2" plates in each layer. An exception had to be made in the gusseted assemblies that connect the vertical and horizontal drive systems because the vertical screw jacks could not fit under the stands and had to be placed outboard.

The Yaw Base Plate mounted to the Horizontal Base Plate

The brown plate shown in these pictures translates horizontally with respect to the green plate. The left view shows what the assembly looks like once the yaw base plate is attached. The assembly on the right is a view looking up from below at the horizontal drive mechanism with the vertical drive not shown for clarity. The horizontal drive motor is driven by a 48V driver, and the reducer is a Hub City model 131 15:1 worm gear reducer. All of the motors on this stand are driven at 900 RPM. The horizontal drive speed is a maximum of 1.7 inches per minute.

The milled pocket on the right side of the green plate in the right picture is a relief because the vertical drive reducer and motor is too tall to allow the full range of vertical motion without having this area relieved.

One can see the same Dicronite pads that allow the yaw rotation to happen in the corners of the brown plate. The pin that restrains the yaw rotation is also shown at the left side of the brown plate in the left picture.

The Pitch-Yaw Plate mounted to the Yaw Base Plate

The pitch-yaw plate is quite busy because two different motor drives are mounted on this plate. The drive on the left is the yaw drive. Note that the yaw jack (a J5 5-ton jack) has a clevis at both ends. This arrangement allows it to push the dark blue plate in a flat rotation around the yaw pin without experiencing any side loads. Because the jack rotates with respect to the blue plate, the motor and reducer must be mounted to the jack, and not to the plate. The right view shows a close-up of the yaw drive train.

The yaw drive is also powered by a 48V driver. The reducer is a Hub City model 131, 15:1 worm gear reducer. Speed for this drive is the same as for the horizontal drive, 1.7 inches per minute. The maximum stroke of this drive is +/- .171 inches.

The pitch mechanism is a J10 jack driven by a Hub City model 211, 15:1 worm gear reducer. The pitch motor must also be driven by a 170V system to provide adequate power. The stroke of this jack is +/- .353 inches. It was a little tricky fitting in the J10 jack in this space, so a non-standard attachment to the lead screw was called for. The lifting block attached to the J10 ACME lead screw fits up into a pocket in the stand top plate which acts as a keyway. This keyway prevents the pitch lead screw from rotating, something that must be true or the jack will not raise and lower. Since the pitch jack is mounted securely to the base, there is a horizontal load on the pitch jack as it raises and lowers. To counter this, the pusher block will also be Dicronited.

The Stand Top Plate Assembled to the Pitch-Yaw Plate -- The Stand is Complete

These pictures show the assembly of the stand top plate to the yaw-pitch plate. The first row right picture makes the top plate transparent so you can see the arrangement of components underneath it more clearly. The left picture in the second row shows how the pusher block on the pitch J10 jack is keyed into the top plate.

The right picture in the second row shows the yaw base plate transparent to make the lock down bolt spacers visible. Note that these flanged spacers do not pull down on the plate. There must be a gap between the flange of the bolt spacer and the slot in the yaw base plate. Their only function is to keep the two plates from separating if the stand were lifted by the yaw base plate. Similar features that hold the pitch-yaw plate to the yaw base plate have not been modeled yet and it is a point of discussion about whether these features that hold the layers of the stand together are necessary.

Diagram of the Stand Rotational Degrees of Freedom

With the yaw and pitch drives shown in these pictures it should become clear that the vertical axis of rotation of the yaw mechanism intersects with the horizontal axis of rotation of the pitch drive, and the intersection point is directly below the beam axis. Ideally, one would want this intersection point to happen at beam height. That way, any rotation of the pitch mechanism would not cause a translation of the collimator along the beam axis.

It was impractical to make this intersection point at beam height because that would have required the pitch bearings to be on either side of the shielding steel at beam height and they would have interfered with the plumbing on the wall of the Booster. The arrangement shown was felt to be a reasonable compromise from the ideal. There is one point in space below the collimator (the intersection point of the yaw and pitch axes) which does not translate during a rotation of either axis. The amount of translation at beam height is easy to calculate from the position readback system. Also, the collimator is completely insensitive to the small translation caused along the beam-line and the control system will be programmed to correct the vertical height of the center of the collimator after a pitch rotation.

Back to the Booster Collimator Shielding Main Menu

Back to the Bartoszek Engineering Home Page