TiC#7 failed after two weeks of operations - measurements in the electrical room indicated that the negative side of the CHT was broken (35kOhms to the rest of the target). This was likely due to a large number of proton trips during this target run. At every trip I could see the CHT voltage getting higher and more unstable. This cahnge in voltage was reflected in the unstable anode current and fluctuating RIB beam intensity, as a direct result of the fluctuating power on the CHT. The break was confirmed in PIE, loosening the screws caused the target container to come off without the cathode. Otherwise nothing unusual was seen. The anode wire connections seem very robust, dragging the target around with the multipin disconnected, flipping it up and down, etc and the anode wire stayed attached. Though we saw lots of outgassing on this target, things looked the normal level of dirty. There was no buildup between PE and EE.

Nb#10 never worked online - repeller had no effect on the beam and one of the RF wires was touching HVC. Max 60pA was seen on the source cup. See notes here ITW-TM2-Nb#10/#11-LP-IGLIS  (Web view)

Target was handled roughly after use - kicked around in NHC for a long time, then loaded into a pail (multipin had to be dettached for this) then moved to SHC, unloaded from pail, then plates covering the RF electrodes had to be pried off. Lots of knocking and shaking.

Theory was that the repeller wire was disconnected. But found it not obviously broken, had good continuity from the ring tongue to the end of the wire pulled out of the multipin, even after all the abuse. Same for the RF (from the end of the electrode to the end of the wire pulled out of the multipin). Did find some of the multipin wires were flaky - if moved around they showed intermittent shorts to HVC. But not clear if this is from the rough handling afterwards or not. The problem was not in the module since NB#11 ran immediately afterwards with no issues. Most likely culprit seems to be the insulation beads on the wires. Should look for more robust insulation solutions. On the other hand, the connections of the wires to the electrodes seems very good.

Still not sure why no effect of the repeller was seen. Could be a intermittent break/short, though this seems unlikely. Maybe the connection in the multipin was not good? I did not get a chance to inspect that.

Over the period of July 20-21st TM1 had several repairs completed and tested. The easiest of the repairs was the installation of a gasket and a blank off. The next repair was to replace the 2 oring on the side panels that were removed for inspection and repair. The instrument panel oring was replaced again and the previous one measured. Both were found to have the correct diameter of 0.131 inches for the seal groove. When attempting to cut the thermocouple line back to an acceptable distance it detached from the stainless tube running down the high voltage chase. It was discovered that it was only being held in place with heat shrink and that the bend consisted of a hand made copper tube that was flared slightly, a truly poor electrical and mechanical connection. Also discovered was the thermocouple wire was left in the tube, this too was removed. The stainless steel tube was then secured to the tube heater (C) lines with a PEEK tie wrap. The port where the thermocouple line was then blanked off. The extraction electrode and mounting plate feed through stem repair went as practiced and no unexpected events occurred. One thing of note is that because the lines are twisted and some of the tube was removed the quick connects are no longer facing the same direction they were before. During these repairs the Ion gauge screen contacted a filament and caused a fault, this was repaired by edi who rotated the gauge. Roughing was started at ~10 am on July 21st and the turbo pumps turned on at 10:45.  At 13:20 all of the lines and external seals were tested and no leaks were found. The module was at a pressure of 2.9 x 10^-5 Torr (IMG) and 1.4 x 10^-5 Torr (ION). The leak test cart was at a pressure of ~2.7 x 10^-2 Torr and a leak rate of 1.1 x 10^-2 atm cc/sec for the external leak test and the repaired lines. A small response was found in the Tube heater (D) where the leak increased to 2.1 x 10^-8 atm cc/sec. The leak test cart would then no longer drop below 1.9 x 10^-8 atm cc/sec and there was no response measured for lines C, B, and A. Over all very positive results! After 6 hours of pumping down the pressures were 2.0 X 10^-5 Torr (IMG) and 9.7 x 10^-6 Torr (Ion).  

Bevan Moss wrote:

Over the period of July 20-21st TM1 had several repairs completed and tested. The easiest of the repairs was the installation of a gasket and a blank off. The next repair was to replace the 2 oring on the side panels that were removed for inspection and repair. The instrument panel oring was replaced again and the previous one measured. Both were found to have the correct diameter of 0.131 inches for the seal groove. When attempting to cut the thermocouple line back to an acceptable distance it detached from the stainless tube running down the high voltage chase. It was discovered that it was only being held in place with heat shrink and that the bend consisted of a hand made copper tube that was flared slightly, a truly poor electrical and mechanical connection. Also discovered was the thermocouple wire was left in the tube, this too was removed. The stainless steel tube was then secured to the tube heater (C) lines with a PEEK tie wrap. The port where the thermocouple line was then blanked off. The extraction electrode and mounting plate feed through stem repair went as practiced and no unexpected events occurred. One thing of note is that because the lines are twisted and some of the tube was removed the quick connects are no longer facing the same direction they were before. During these repairs the Ion gauge screen contacted a filament and caused a fault, this was repaired by edi who rotated the gauge. Roughing was started at ~10 am on July 21st and the turbo pumps turned on at 10:45.  At 13:20 all of the lines and external seals were tested and no leaks were found. The module was at a pressure of 2.9 x 10^-5 Torr (IMG) and 1.4 x 10^-5 Torr (ION). The leak test cart was at a pressure of ~2.7 x 10^-2 Torr and a leak rate of 1.1 x 10^-2 atm cc/sec for the external leak test and the repaired lines. A small response was found in the Tube heater (D) where the leak increased to 2.1 x 10^-8 atm cc/sec. The leak test cart would then no longer drop below 1.9 x 10^-8 atm cc/sec and there was no response measured for lines C, B, and A. Over all very positive results! After 6 hours of pumping down the pressures were 2.0 X 10^-5 Torr (IMG) and 9.7 x 10^-6 Torr (Ion).

 I again checked the TM this morning (July 22) at 10:00 am (24 hours of pumping), There is no external leaks carts pressure was 0.0x10^-4 Torr and leak rate was 0.0X10^-9 atm cc/sec. The IMG gauge was reading 6.8 x 10^-6 Torr and ION gauge was reading 3.1 x 10^-6 Torr. The lowest we have yet to record on the IMG gauge was 6.6 x10^-6 Torr and that was when the ION gauge was reading 1.2 x 10^-6 Torr and there were several leaks. This is a very good sign. I will continue to check the vacuum throughout the day. 

One more thing of note is that the copper feed thru stems have a design flaw, the holes for the locating pins are too large. As such a custom step pin had to be made. It work functionally the same but is not as intended.

 TM1 with Ta #37 target was moved from ITW to the south hot cell. The rotation of TM1 may have been off slightly during the final removal from ITW but it was minimal. Entry into the south hot cell went with out incident. This was my first time moving the module without the presence of Travis Cave. 

 TM1 with UC#2 had the target installed, the heat shield line was pumped on and achieved a vacuum of 0.0 x 10^-4 Torr and a  leak rate of 0.0 x 10^-9 atm cc / sec with no response when sprayed with helium. An electrical check was completed by David and myself all values were withing tolerance.  The module was then moved from the south hot cell to the conditioning station where it will start pump down. The move was uneventful every operated as it should and there were no crashes minor or large. It was attempted to fill the containment box with argon prior to closing it but assuming that didn't work I estimate the target was exposed to air for a total of ~4 hours during installation and move. 

 Pump down in the conditioning station started at ~2:40 for rough down. The turbo pumps were turned on at ~3:30 and at 3:52 the pressures were reading as follows:

IMG in the containment box 9.2 x 10^-5 T

IMG in the secondary 2.0 x 10^-3 T

ION in the service cap 4.6 x 10^-5 T

Note that the cross has been installed on the vacuum vessel which now includes a turbo pump and the IMG 100.

Leak check will occur tomorrow.

 A leak check was performed on the module when it was in the conditioning station today. There were no external leaks found. When the coolant lines were check a small leak was found in the heat shield circuit. Base leak rate was 6.7 x 10^-9 atm cc /sec and rose to 4.1 x 10^-9 atm cc /sec quickly (~3 seconds) and remained there for the 2 minute test. The remaining lines had small responses as usual but the highest response was in the low 10^-8 atm cc /sec range. The pressures were recorded as 1.2X10^-6 T, 2.7 x 10^-6 T and 1.1 10^-5 T in the service cap, containment box, and secondary respectively. The module was then vented with argon which took ~1.5 hours, vessel was pressurized to ~ 1 psi. A panel was then removed to leak check the feed through stem of the heat shield pressure of the line was recorded to be 0.0 x 10^-4 T and a leak rate of 1.3 x 10^-8 atm cc/sec was achieved. The stem was sprayed and no response was found. Swipes of the were taken of the panel (150 counts) and the containment box (3000 counts). The panel was then returned and the vessel roughed out for the weekend, it will be vented and moved to the south hot cell on Monday where the VCR fittings will be checked again and then tightened.  

 On August 22nd TM1 was vented with Argon in the conditioning station. It had been pumping with just the roughing pump over the weekend so the UCx#2 target would not be exposed to oxygen. The move went without incident and took roughly an hour to complete. Once at the hot cell the panel was removed by Tom Lyth and a leak check completed with Bevan, David Wong, Grant Minor, and Tom Lyth present. The connections to source tray were checked starting at the top. Each connection was checked on either side of the VCR and at the center through the hole in the nut. For both connections on the source tray there was no response from the leak cart which was at a pressure of 0.0 x 10^-4 T and had a baseline leak rate of 0.0 x 10^-9 atm cc / sec. The leak check then proceeded to VCR connections on the Target. A leak was found on the upper VCR connection of the target and was repeatable. This leak was detected at the hole of the VCR nut connection by priming the line with helium and placing it next to the hole in the VCR connection. Response time was between 5 and 10 seconds to a maximum of 2.6 x 10^-8 atm cc / sec. The lower connection also appeared to have a small leak rate of 4.4 x 10^-9 atm cc/ sec but could be a result of helium drift. Unfortunately the target is not design to be repaired with manipulators and required the target to be removed from the south hot cell and repaired in the anteroom. The anteroom was prepared for the target by lining much of it with plastic so to prevent the spread of contamination. Once the target was in the anteroom it was leak checked again, the base pressure was 0.0 x 10^-4 and the base leak rate was 0.0 x 10^-9. A response from the upper connection again was detected in the 10^-8 range but no leak was found in the lower connection. It was attempted to tighten the joint to stop the leak but this caused it to increase to the 10^-7 range. This increase in leak rate is as expected from past experience because copper gaskets are used rather than silver plated ones. The heat shield cover was loosened and the gaskets from from both joints were replaced with the silver plated ones. The target was then leak check and passed with no response once again with a full pump down. The target was had an electrical checked with the megger and was within spec. The target was then placed in an argon bag and returned to the hot cell. The bag was damaged went placing it in the hot cell and there were not any acceptable options for leaving the target over night. The target was removed and and placed in 2 bags filled with argon for over night.

On August 23rd the target was removed from the argon bags that were still inflated and returned to the hot cell. The target was re-installed on the target module by Grant Minor, Tom Lyth, an Bevan Moss. One VCR gasket fell off during the first attempt and it appeared that the right target leg may have been slightly bent at some point. Other than that the installation went smoothly but slowly. The heat shield line was then leak checked once again by Grant, Tom, David, and Bevan and no response was detected from the source tray or target VCR connections, the base leak rate and pressure once again achieved the minimum detectable levels. The target module passed the electrical check with the megger, this test was completed by Bevan and Pierre Bricault. The module was then moved from the south hot cell to the conditioning station. Roughing was started at 3:00 and the turbo pumps were turned on at 3:45. The module was left to pump over night.     

Today a leak check on TM1's external seals and the heat shield line was completed by David Wang. After 16 hours of the turbo pumps running on the conditioning station stations pressures were 3.6 x 10^-6 T, 1.2 x 10^-5 T, and 1.8 x 10^-6 T for the containment box, secondary vacuum, and service cap respectively. This is a significantly quicker pump down after the cross with small turbo pump was installed on the vacuum vessel. The base pressure of the leak testing cart was 0.0 x 10^-4 T and the leak rate was 1.2 x 10 x 10^-8 atm cc / sec. There was no response from the external test where the module was sprayed with helium. The heat shield line was pressurized with 80 psi helium which caused the leak rate to climb to a maximum of 2.5 10 ^-8 atm cc / sec but dropped to 1.9 x 10^-8 atm cc / sec and was stable there, the total test time was 2 minutes. This test was repeated another time and had similar results.  These results although showing a slight increase in the detected leak rate are withing acceptable and indicate that the module is leak tight. 

At 9:30 the HV check and conditioning started (~17 hours turbos on), the secondary vacuum, containment box and service cap had the following pressures in Torr respectively, 1.0 x 10^-5, 3.2 x 10^-6, 1.6 x 10^-6.  The voltage was steadily raised with virtually no current up to 28.5 kV which was achieved at 9:50. When raised to 30 kV the current increased significantly to ~300 uA and a large amount of sparking occurred. The voltage was dropped to 0 volts for 2 minutes and then was returned to 30 kV at 10:03, the current draw was around 200 uA and climbing. After a few sparks and time the current had dropped to 120 uA. This indicates that the module is still conditioning and matches what is observed by the control room when the module is in ITW. The module was deemed acceptable for this current run and was then vented with argon this took about 2 hours.

The module was then remotely moved from the conditioning station to ITW. When first attempting to transfer the controls to the remote control room they were not transferred and when testing the up command as per standard procedure the module dropped down slightly. David Wang went to lock out the target again and the controls were successfully transferred. A note of this was made in the log book. Apart from this the move went smoothly.  

Roughing on TM1 started at ~15:00 today by ~16:30 the module had reached a pressure of ~200 mTorr and the turbo pumps were turned on. TP1 failed to achieve normal operation and we running at approximately 4KRPM and 50W with 6.1 A of current draw. TP2 was able to achieve normal operation. The controller for TP1 was then powered off and back on but that failed to resolve the issue. The controllers for TP1 & 2 were then switched at which point TP1 again failed normal operation and TP2 achieved it, thus eliminating the controller as the issue. The turbo pumps have been turned off and the roughing pump left on for overnight in case Vacuum group wishes to do any additional diagnosis. 

A leak check was performed on the module on October 27/10/2011 by David Wang and was reported to Bevan Moss but it never made it into the Elog system. This entry is to summarize the results of that testing. All circuits were pressurized to 80 psi with helium and every line except the tube heater (+) showed no or small responses as experienced before. The tube heater line had a slightly larger response under the conditions summarized below:

Varian 959

Base Pressure                   7.0 x 10^-3 Torr

Base Leak Rate                  7.5 x 10^-9 atm cc/sec

Gauges

IG1                                         2.0 x 10^-6 Torr

PNG5                                    9.6 x 10^-6 Torr

PNG4 (C.S.)                        2.4 x 10^-5 Torr

Convectron (Back line) 20mTorr

 The leak rate quickly climbed to 3.0 x 10^-8 atm cc/sec then constantly climbed to 1.4 x 10^-7 atm cc/sec for 2 minutes. David repeated this test again at the end of the leak test and found the same results. In the past we have called this a leak and attempted to find its location and repair it. David believes that this isn’t a real leak as we have been able to achieve very good vacuum on TM4 all year and that we do not normally touch the feed through stem on this circuit. This result makes me nervous as the total differential is roughly the same as that we found on the UCx#2 run(which was at a higher vacuum).

The result of this test caused the module to move back to the hot cell for further more localized leak checking.

 After Chad and Davids leak check results from the containment box it was decided to perform a leak check on the feed through stem in the service cap. This is a known point of failure on TM1 and the same design flaw is likely in TM4. Once the service cap was open a visual inspection was performed and some oxidization of the copper line on the tube heater (+) circuit was observed (see attached pictures). The leak check was performed with the cart at a base pressure of 0.0 x 10^-4 Torr and base leak rate of 0.0 10 x 10^-9 atm cc/sec. A total of 4 joints were checked and no response was found. The service cap was then flooded with helium and once again no response was found. It was decided to give the module a clean bill of health, make note of the baseline test from the conditioning station, and continue to monitor the circuit.  

 The silicon carbide targets voltage was raised to 42 kV today before continuous current draw was observed. It took roughly 10 minutes to raise the module to this voltage with only the occasional spark. The ION gauge reading was 2.0 x 10^-6 T, The service cap IMG 3.6 x 10^-6 T, and the secondary vacuum was 4.5 x 10^-6 T. Setup check and conditioning was performed by Pierre.

 TM1 was high voltage tested without a target on November 16, 2011 with Don Jackson, Pierre Bricault, and Bevan Moss. The module was able to hold up to 25 kV but between that and 32 kV there was lots of discharge which could be measured in the vacuum. See the attached plot for the voltage current and sparks.

 High Voltage (HV) testing was completed on TM1 today. Vacuum was very good 1.9 x 10^-7 T in the service cap, 3.2 x 10^-7 in the containment box, and 2.0 x 10^-6 in the secondary. The voltage was successfully brought up to 27kV with a current draw of 0.12mA and o sparking. At the next step, 30 kV, the current draw was above 0.3mA and sparking was beyond 20 sparks per minute. The module should not be taken above 27kV. There is the possibility this voltage can be increased with conditioning but it is very unlikely. See attached for a graph of the testing. These results are comparable to the results without the source tray (see previous elog)

 A high voltage test was completed on a target free TM1 in the conditioning station. Only the turbo pumps on the target module were operational due to phase 2 construction. Testing started at 2:51PM with the following vacuum gauge readings, 1.6 x 10^-6 Diagnostics Box, 7.4 x 10^-7 Service Cap, 2.0 x 10^-6 Containment Box. Voltage was increased ~3kV at a time every minute until 21 kV was reached with a current of 0 uA. Voltage was then increased in ~1.5kV steps every 2 minutes until 30 kV was reach at a current of 60uA. This was held for 7 minutes until a spark occurred and the current spiked to 300uA and the voltage dropped. The voltage was dropped back to zero. The vacuum gauges had increased and read the following  1.6 x 10^-6 Diagnostics Box, 7.9 x 10^-7 Service Cap, 2.1 x 10^-6 Containment Box. Testing was restarted 2 minutes later (the vacuum gauges were still recovering for ~ another 2 minutes) and the voltage was quickly increased to 24kV reaching a current of 100uA. at 3:22 The current then fluxuated but trended up for the next 12 minutes peaking at a total current of 240uA. After this point the current started deceasing, at 3:41 the testing was concluded and the average current was ~170uA. The vacuum gauge readings were 1.6 x 10^-6 Diagnostics Box, 7.4 x 10^-7 Service Cap, 2.0 x 10^-6 Containment Box. It was deemed that this is "not ideal but good enough" to put the target on.  The maximum known required voltage will be ~22kV for Berillium which TM1 should be capable of providing.

 A HV test was complete on TM1 with Friedhelm, David, and Myself. The module behaved as it has in other tests. The vacuum achieved was 1.6 x 10^-6 in the diagnostics box, 2.3 x 10^-6 in the containment box, and 8.6 x 10^-7 in the service cap. Maximum voltage without continuous sparking achieved was 27kV. Current was unstable varrying from 40 to 300 uA. This voltage was held for 10 minutes. 

Ta#38 has been placed into the south hot cell. Swipes of the tool port were taken there was a count of 600 on the floor and 150 on the walls and top. 

 TM1 has been moved from ITW to the SHC. The move did encounter some issues, one of the railings for preventing people from falling into the MAA was hit with the containment box. This caused the module to rotate. The module rotated ~90 degrees before the problem was noticed. The module was backed off from the post once noticed but had already freely rotated. The box was inspected with the cameras prior to being inserted into the SHC. No damage was detected. The error came as a result of over confidence. It did appear it was going to be hit but as the module moved closer I believed that there would be clearance. The angle of the camera has caused the module to appear it was going to hit thing before in the past but it actually (once the module was closer to the object) had several feet of clearance. This is not a mistake I will make again.

Chad Fisher, Grant Minor, and Don Jackson, have been alerted about this. Chad will complete another visual inspection in the hot cell. David will take swipes of the pole to ensure there is no contamination.