From wagner@fnal.gov Sun Nov 4 20:06:11 2001 Date: Wed, 31 Oct 2001 16:18:11 -0600 From: Bob Wagner To: mukherjee@fnal.gov, Morris Binkley , JC Yun , ambrose@fnal.gov, Ken Schultz Subject: Procedure, equipment, time for working on SL4-227 (whenever). Comments welcome. General plan to diagnose and fix SL4-227. This is likely going to be a sufficiently complicated procedure to warrant writing down a specific plan, partiularly if it comes to using the stringing engine. Time estimates would be useful also. This plan would likely also be useful in the future, if the opportunity does not present itself during the October shutdown. The first section goes over what we know about the SL4-227 problems. Section 2 is the procedure, with itemized time estimates and equipment needed. The total time estimated is 14 - 19 hours, depending on whether or not the stringing engines are used. Section 3 lists the necessary equipment. I. What we know about the problem in SL4-227. ----------------------------------------------------------------- Sense wires 0, 2, 3, 4, and 5 have all been observed to discharge. Sense wires 3 and 4 most often discharged simultaneously. They were the first observed discharges, occuring before the Commissioning Run at B0. We observed many of these S3 and S4 discharges, and one could often see current sink to P4 and P5. Twice we saw a clear voltage increase on P4. Once S2 was observed to discharge with S3 and S4, and later S0 was observed to discharge with S2. Sense wires 0 and 2 were disabled after the Commissioning Run, but before the start of Run 2. Sense wire 5 started tripping after the start of Run 2 and was observed in the data as a "Megablast". Examining the ASDQ DB afterward, the 25 ohm series resistor for S5 had "blown" and read open with a DVM. Resistors on the other channels were ok. I don't believe we have a good guess as to what's going on. Connections through the chamber have always measured good with the capmeter until the ASDQ series resistor blew for S5. The "simultaneous discharging" makes one think of some electrostatic instability suddenly occuring, like a loose wire or field plane "piece" suddenly moving or the wires themselves suddenly moving, possibly due to being loose from the pultrusion. The fact that neighboring potential wires to S3 and S4 were sometimes seen to sink current (and in two cases increase in voltage) indicates that the problems may be on one of the wire plane endboards. Even if the problem is on an endboard, there is a good chance that we may have to pull it clear of the endplate slot in order to fix it. In my opinion, this is most safely done using the stringing engine. If one is going to the considerable trouble of clearing the area and cracking open the gas seal (likely on both sides), I believe we should at least have the capability of hooking up the engine and pulling the planes clear of the slot. 2.Procedure to address the SL4-227 problem. ---------------------------------------------------------------------- -->Fill with "enough" P5 to check for leaks after the fixes. Alternatively use Ar/Et, but flow little enough to not allow a flammable mixture inside the chamber (as we did before). Equipment: P5 bottles (should always have a few on hand). -->Carefully measure the chamber leak rate, so we know what to get back to. One can use the mass flowmeters to measure the minimum flow for bubbling at our present exhaust bubbler oil level. One should also record pressures at both this minimum setting and 20 SCFH. -->Remove HDDB's and/or ASDQ DB's to clear a reasonable area around the motherboards. This involves likely removing the HVDBs for 225, 226, 227 and 228; and the ASDQ DBs for 224/225, 226/227 and 228/229. It should only take about 1 hour total. Equipment: Usual tools for removing and installing these boards -- all in silver case. -->Crack open the HVMB over 226/227. Hopefully one can work on the 227 wire plane endboard without also removing the MB for 228/229, but this remains to be seen. This should take about 1 hour. One has to be careful not to spread the "crack" in the epoxy to neighboring cells. Equipment: Soldering iron with blade (in cabinet), Chisels (ask Ken), Attachment for vacuum (cork and small tube). -->Tie strings to the wire plane endboard as a safeguard. Remove the G10 piece blocking the wedge and verify that the ledge is still intact and epoxied well to the endplate. Remove the wedge and examine it: Is the Kapton flap in tact and not bent back? Are there signs of discharge on the flap or endboard? Is there a loose conductor on the endboard? Look inside the chamber as far as one can. This work should take about 1/2 hour. Equipment: Usual string (in a cabinet). Chisel. Magnifying glass. Fiberscope. -->Crack open the ASDQ MB over 226/227. Hopefully one can work on the 227 wire plane endboard without also removing the MB for 228/229, but this remains to be seen. This should take about 1 hour. One has to be careful not to spread the "crack" in the epoxy to neighboring cells. -->Tie strings to the wire plane endboard as a safeguard. Remove the G10 piece blocking the wedge and verify that the ledge is still intact and epoxied well to the endplate. Remove the wedge and examine it: Is the Kapton flap in tact and not bent back? Are there signs of discharge on the flap or endboard? Is there a loose conductor on the endboard? Look inside the chamber as far as one can, this time illuminating from the other end (much more effective). This work should take about 1 hour. ***IF WE NEED TO CONTINUE DIAGNOSTICS AND FIXES, SET UP THE STRINGING ENGINE*** -->Shine a laser through the slots to mark the points where the stringing engine should be mounted on the endplugs. Without the wedge on an axial layer, this should be straightforward. Allow 1/2 hour. Equipment: Laser pointer. -->Stud weld the studs on the Endplug and mount the stringing engines (without the plane replacement spool). Allow 2 hours. Equipment: Stud welder, studs, stringing engines. -->Hook up the engine to the 227 wire plane and apply tension. Crack loose the remaining epoxy and pull the wire planes free in the slot (need handle). Insert the mylar funnel and pull one endboard clear of the slot. See if the wires are loose on the pultrusion. by looking through the slot. If need be, pull the other endboard clear of the slot and examine it. Fix problems as needed. Allow 2 hours Equipment: Plane handle / puller, Chisel, mylar funnel, soldering iron (possible), magnifiers, walkie-talkies, phones, flashlights -->If one determines that one needs to replace a wire or field plane, stop and replace the wire plane for work at another time. ***BUTTON UP*** -->Replace the wire planes and latch them. CAREFULLY seat them against the notch. Reglue the plane to the endplate, including the separate G10 peice. Check that one has continuity through the plane with an ohmeter. Allow 1 hour, including time for epoxy to set Equipment: DP110 translucent epoxy, ohmeter, Q-tips, Kim-wipes -->Replace the HVMB and ASDQ MB. Check for continuity with the capmeter before epoxying. Epoxy with DP110 using the mylar strip between boards. Allow 2 hours, including drying time. Equipment: Capmeter, mylar strips -->Resume flow with P5 to check for leaks. Fix as necessary and reproduce the original leak rate measurement. Allow 2 hours for leak check and "repairs" Equipment: Red flammable gas leak detector. -->Reinstall HVDBs and ASDQ DBs and check for continuity with the capmeter. Allow 2 hours -->Reinstall Microcoax boards and Faraday cages -- Check readout for pulses and noise. Allow 2 hours ***Time*** -->Open both sides, but don't use stringing engine: Open + work: 4.5 hours -->Use stringing engines to pull endboard from endplate (assume engines on hand and working): Mount, allign, work: 4.5 hours (likely optimistic) -->Button up: Endboards, MBs, continuity test, leak test, DBs, Electronics tests, F-cage: 10 hours. -->Total: 14 to 19 hours with access to both sides. It would need to spread over two days. 3.Tools ------------ -Mylar funnels -Mylar strips for epoxy joint between motherboards (1/4"?) -P5 bottle(s) for leak checking (I'll check with Del). -Soldering iron with blade tip. -Chisels -"Cork 'n' Tube" vacuum attachment -String (tie to wire planes) -plane handle / puller (missing, so Kourosh making) -Magnifying glass -flashlights -fiberscope -Laser pointer -Stud welder (? from Mike Lindgren, Rob?) -Studs (? Do we have the right size on hand) -normal soldering iron (possibly needed) -DP110 translucent epoxy -ohmeter -capmeter -Q-tips -Kim-wipes -Red flammable gas leak detector (Ken or gas techs) -Stringing engine, controls, string -Wires to attach to wire plane (Kourosh has made some). -Long control cables for stringing engine (run around detector) -- **TEST** -Walkie Talkies, phones with headphones