Support Tube Reinstallation Lessons learned S J Metcalfe
Support Tube Reinstallation. Lessons learned. S. J. Metcalfe, Revision 9, 10/20/03 Introduction. • On 16 th October 2002 the PEPii Support Tube was successfully reinserted into Ba. Bar having been removed on 31 st July to add cooling to the Vertex Chamber bellows. Removal had caused serious but reparable damage to the inside of the Drift Chamber so the reinsertion procedures were closely scrutinized and amended to ensure that no further damage would be caused. There follow a number of proposals resulting from that reinsertion with the goal of minimizing the risks and reducing the time associated with future removal and reinsertion. Contents: a) b) c) d) e) f) g) h) i) What worked well? What didn’t work well? Other issues Proposed Forward rail system Benefits of adding a Forward rail system Backward rail system Proposed amendments to procedure Fastener and Bridge loadings Conclusions 12/21/2021 Support Tube Reinstallation. Lessons learned. 1
a) What worked well? 1. We were able to find a ‘sweet spot’ trajectory for the Insertion Beam which tracked down the center of the Drift Chamber to within +2 mm without periodic realignment. The position of the new tooling ball holders were subsequently recorded for future use. 2. Sensors on the Support Tube (ST) backward end worked very well and inspired confidence during the installation. 3. ST temporary supports: The backward supports worked well. The new split feature of the forward support worked well, enabling easy insertion and removal of the support. 4. All the modifications that we carried out to eliminate protrusions beyond the outside diameter of the ST were successful. This included replacing the hex head bolts (which caused the Drift Chamber damage) with low profile button head bolts. 12/21/2021 Support Tube Reinstallation. Lessons learned. 2
b) What didn’t work well? 1. ST Installation took far too long. Target time for future removal/installation should be about 4 hrs, so the schedule for reinstallation would look like this: • Transport ST to IR 2 outer hall the day before installation. • AM, prepare ST and setup on beam; hookup and dress SVT cables around midday. • PM, inspection, installation. 2. Link between end of Insertion Beam and Support Tube was too rigid, causing the ST to be steered in ‘x’ whenever the beam was steered in ‘x’. The link has been redesigned using a large spherical bearing to constrain ‘x’ and ‘y’ but not x-angle (yaw), nor y-angle (pitch). ~$14 K. (Action: SJM) 12/21/2021 Support Tube Reinstallation. Lessons learned. 3
Present design of link between end of Insertion Beam and Support Tube Sensors P 919002 a Support Tube Hook Beam Hook on end of the beam engages and supports bar attached to the end of the Support Tube. 12/21/2021 Support Tube Reinstallation. Lessons learned. 4
Support Tube Position sensor housings New link design Yoke with ‘x’ adjustment Spherical bearing Locking cotter pin Hitch pin Insertion Beam Backward 5 Rod end bearing in bracket on end of beam engages with yoke attached to end of Support Tube. Load is carried by 1. 5” diam. hitch pin in double shear. 12/21/2021 Support Tube Reinstallation. Lessons learned. 5
View of new link showing beam approaching yoke with pin about to be inserted Support Tube Insertion Beam explode 3 Hitch pin 12/21/2021 Support Tube Reinstallation. Lessons learned. 6
3. Manual readout of gaps at the forward end was time consuming so we need to add sensors near the Drift Chamber entrance. The inductive sensors (as used at the back of the ST) are not appropriate because of the ST Carbon fiber, but a simple wheel-onarm running along the ST with a potentiometer readout would do the job. (Action: SJM, S. Ecklund – sensors purchased, need to design mounting). 4. A ‘z’ position sensor would also be required to plot the path; i. e. x(z), y(z). The ASM, WS 3 -7500 uses a cable on a reel. Cost $1 K for sensor plus $1 K for mounting hardware. (Action: SJM) Pos sensor 12/21/2021 Support Tube Reinstallation. Lessons learned. 7
5. Display all gap readouts from sensors on one screen, with a duplicate screen at the forward end a local printer to provide hardcopy but we do need backup system in case of computer crash. (Action: S. Ecklund) 6. The ST forward temporary support has a crossmember which fouled the 90 deg elbow at the 6: 00 position on the ST. The elbow had to be removed but this caused concerns about the possibility of water spillage near the Calorimeter. The crossmember has been modified to clear the elbow. (Action: SJM) 7. Communications equipment was inadequate because of interference, limited battery lifetime and crosstalk from other channels. An upgraded voice communications system (wired or wireless) for use in IR-2 needs to be investigated. Perhaps Ba. Bar is interested in pursuing this, as it may help them in many other operations. 12/21/2021 Support Tube Reinstallation. Lessons learned. 8
8. Fix the crane. On Ba. Bar centerline, the crane cannot be ‘inched’ in ‘x’ so we had to use fast speed when adjusting the crane to maintain the hook on the Ba. Bar centerline. This made Raft and Support Tube installation difficult, relying heavily on crane operator skill for success. (Action: Jim Krebs). Specialists have inspected the crane rails and quoted $4. 8 K for reshimming. The crane downtime will be two days. 9. In order to make an ‘x’ adjustment on the backward pier roller plates to steer the insertion beam, it was necessary to turn four bolts (two in one direction and two in the other direction). This was potentially confusing and on one occasion the beam was moved in the wrong direction. It is proposed that the upper ‘x’ adjuster screws are backed off by about 3 mm so that only the lower rollers constrain the beam in ‘x’ – we can do this because the beam no longer controls the roll of the ST. The sense of operation of the two pusher screws will be clearly marked. A further enhancement might be to redesign the lower roller blocks to incorporate a single ‘x’ adjuster screw. (Action: SJM) 12/21/2021 Support Tube Reinstallation. Lessons learned. 9
10. Having to reconfigure the come-along during the reinsertion was both inconvenient and dangerous. A continuous chain in a closed loop would automatically constrain and move the Insertion Beam in both directions. (Action: SJM). Brake lever Insertion Beam Drive arm Continuous chain Chain 2 Idler sprockets here Sprocket with drive attachment. Return sprocket with tensioning device. Bracket will be mounted on one of the beamline floor supports. 12/21/2021 Support Tube Reinstallation. Lessons learned. 10
11. Whenever we stop the beam we need to ensure that the brakes are applied to prevent unexpected ‘Z’ movement (as happened on one occasion). The braking bolts that were used for this purpose worked OK but they took too long to lock and unlock. We need to add a lever type toggle clamp which would be applied whenever the beam is not being moved. The braking bolts would still be applied during any extended stoppages. ~$400. (Action: SJM). 12. Reinforce the brass slides on the roller plates with steel channel section to prevent bowing. ~$2000. (Action: SJM). P 7310096 a 12/21/2021 Support Tube Reinstallation. Lessons learned. 11
13. The D. M. D. gauge did not prove useful, mainly because of the confidence inspired by the sensors. The gauge was only used twice but it was difficult to maneuver and it hung up on a number of items on the beam, causing unnecessary concern. 14. Consider adding a counterbalance weight to the back end of the insertion beam in order to keep the weight acting on the lower rollers in the fully inserted position thus avoiding the load transfer from lower rollers to upper rollers. (Action: SJM). This would require a box of lead weighing 3000 lbs, measuring ~20 x 20”. Alternatively we could modify the procedure to P 7310096 b close up the gap on the upper rollers just before the beam rocks over to the upper support. After passing this point, the gap would be increased by adjusting the lower rollers. 12/21/2021 Support Tube Reinstallation. Lessons learned. 12
c) Other issues 1. Provide an adequate supply of flashlights and batteries. In particular, we need two lights with bright, narrow beams to aid visibility between the ST and DCH. 2. Need contingency plan in case of power failure in IR 2 during ST removal/insertion (fortunately, on this occasion, IR 2 was not affected by the power failure). Persons responsible for power to be notified in advance that power to IR 2 is critical. 3. Fully test the revised systems in advance. (Action: SJM) 4. Rewrite the detailed removal/insertion procedures to incorporate these modifications. See section ‘g’ for proposed procedure. (Action: SJM) 12/21/2021 Support Tube Reinstallation. Lessons learned. 13
d) Forward Rail System. • A forward rail system is proposed which would consist of two parallel rails attached to a bridge structure which would sit on the two existing rear raft supports and the S. T. temporary support. The rails would be adjustable in ‘X’ and ‘Y’ and optically aligned parallel to the DCH centerline. ~$44 K. (Action: SJM. ) 12/21/2021 Support Tube Reinstallation. Lessons learned. 14
Proposed Forward rail system Bridge Pair of round linear bearings End of LER beamline ‘x’ adjusters ‘z’ pos’n sensor Forward Calorimeter S. T. extension arms 10 ton Duff Norton jack Turnbuckles ‘x’ adjuster Carriage with four self aligning, recirculating ball bearings ‘x’ adj Forward raft supports ‘y’ supports S. T. temp support Pier Barrel Calorimeter Side 7 12/21/2021 Support Tube Reinstallation. Lessons learned. 15
Iso. view of rail system with ST removed Forward Calorimeter Rails HER beampipe S. T. extension arms Carriage LER beampipe Pier ISO 6 Out 12/21/2021 Support Tube Reinstallation. Lessons learned. Extension table for Forward Calorimeter extraction system 16
e) Benefits of adding a forward rail system: 1. The crane is not required until the ST is fully outside of the DCH. This was one of the recommendations of the ST accident review committee and will remove the possibility of crane operator error. 2. This system would remove the concerns raised on page 9 para. 8, regarding the lack of slow speed in ‘X’ during ST removal/insertion (but still a big problem for raft insertion and removal). f) Backward rail system The review committee’s main concern about the existing backward rail system was that the curved beam might collide with the DCH when inserted. We have shown that this is not the case (see page 1 para. 1). The beam tracks well if it is positioned correctly in the first place and we now have the alignment data necessary to do this, therefore we do not propose to significantly change this system. 12/21/2021 Support Tube Reinstallation. Lessons learned. 17
g) Proposed Amendments to procedure Backward Prep. 1. Optically align insertion beam to the ‘sweet spot’ which was identified during the 2002 shutdown. 2. Roll beam towards ST. Adjust spherical bearing on end of beam in ‘x’ and ‘y’ by pitching beam to adjust bearing position in ‘y’ and slide yoke sideways to adjust in ‘x’. Do not adjust beam in ‘x’ at this time. 3. Insert hitch pin, sleeve and locking cotter pin. Forward Prep. 1. Attach extension arms and crossbar to ST. 2. Install bridge supports on pier and pedestal. 3. Crane in rail bridge, place on supports and rough align. 4. Using alignment tool, position center of jack under ST pickup point on end of extension arms using ‘x’ adjusters on ST temp support frame. 5. Optically align rails parallel to DCH ID using only the ‘x’ adjusters on the pier plus all the ‘y’ adjusters, as required. Recheck alignment tool. 6. Raise the jack and engage hitch pin, sleeve and locking cotter pin. 7. Place a hydraulic jack under the bridge near the ‘z’ motion sensor. 12/21/2021 Support Tube Reinstallation. Lessons learned. 18
Rollout. 1. Continue raising Duff Norton jack to transfer ST load from temporary support frame to rails. Stop raising jack when adjusting screws on support frame are just free to turn i. e. jack is taking all the load. 2. Pitch Insertion Beam upwards to take the ST load. Check that ‘Y’ adjusting screws are free to rotate then remove mounting clamps. Place a spirit level on ST extension arms and attach turnbuckles which will be used to control ST roll if required. The turnbuckles will also be set to eliminate any lateral motion (‘X’ and ‘Z’) in the jack screw. 3. As rollout begins, the two beam trajectories (2002 and present) will be compared and adjustments made to the beam if required. 4. Initial rollout movements will be small, say 4 -6” until we gain confidence that the sensors (particularly the new front sensors) are performing correctly and the beam and rails are tracking as required. 5. If forward sensors indicate that the slope or angle of the rails requires adjusting then make the ‘x’ adjustments at the pier support only – this ensures that, upon reinsertion, the ST returns to the same position from which it started. 6. Whenever the ST is not rolling, the toggle clamp brake is to be applied. 7. On completion of ST rollout, reattach backward mounting clamps, tilt insertion beam down to set back end of ST down on supports in temp support frame. Continue to tilt beam until hitch pin is free to rotate then remove locking cotter pin, pull out hitch pin and roll back insertion beam. The ST can now be either removed with the spreader bar (if time permits) or strapped down and removed the following day. 12/21/2021 Support Tube Reinstallation. Lessons learned. 19
h) Fastener loadings 4 x 3/4” Hex Hd Gr 8 bolts Allowable= 150 Kpsi Actual= 1. 34 Kpsi (C 3) Fs=112 Forward Support Tube Hitch Pin 1. 5”diam double shear Allowable= 70 Kpsi Actual= 665 psi (C 5) Fs=105 Backward Support Tube 12 x ½” Hex Hd Gr 8 bolts 12 x ½” Button Hd Gr 8 bolts Allowable= 75 Kpsi Actual= 9 Kpsi (C 4) Actual= 2. 5 Kpsi (C 1) Fs=8. 25 Fs=30 Note: Bolts won’t actually see these shear forces because brackets will be a good fit in the S. T. 12/21/2021 Support Tube Reinstallation. Lessons learned. 20
h) Bridge loading 2088 lb (176/198 x 2350) = 1044 lb per rail 2350 lb Weight of ST = 4700 lb Defl = 0. 018” w/o prop Defl = 0. 016” Defl = 0. 004” 12/21/2021 Support Tube Reinstallation. Lessons learned. 21
i) Conclusions ØAll of the proposals that have been outlined above will contribute in some way towards the stated goals of minimizing the risks to Ba. Bar and reducing the time associated with future removal and reinsertion. ØSome are more desirable than others, for example we must not consider removing the ST again without using the new link proposed on page 3. This link has been designed and the drawings are ready for manufacturing. ØThe forward rail system is desirable for a number of reasons, the primary one being the elimination of the crane which does not follow the same ‘Z’ axis as Ba. Bar. ØWill the combined effects of these proposals enable us to rollout the ST in 4 hours whilst still having absolute confidence in the security of the system? We won’t know until it actually happens but I believe I’ve considered all the issues that delayed us in 2002. 12/21/2021 Support Tube Reinstallation. Lessons learned. 22
- Slides: 22