Instrument Issues History

Agilent ICP-OES

• 19 Jul 2019 Installed a cover for the autosampler. Vented cover into the ceiling vent.
  
• 06 Jun 2019 Wall mounted gas regulator was blocked/stuck. A spare regulator was found and installed.
• 16 Apr 2019 Sent a request to shore to requisition covers for the SPS 4 autosampler to prevent dust and debris from contaminating open vials during an analysis. 30 Apr 2019 The cover will be sent for the next expedition.
• 15 Apr 2019 ICP Expert was having display issues due to the MCS's updating the laboratory monitors several days ago. Mike fixed the display by changing the resolution specifically to 1600 x 1200 and restarting the computer.
• 13 Apr 2019 Updated the interstitial waters ICP Expert template with correct concentrations in µM and mM units instead of the previous ppm units.

Agilent Cary 100 Spectrophotometer

• 17 May 2019 The following picture indicates the effect of not shaking the vials after each addition of reagent when determining ammonium:
  
• 16 Apr 2019 Sent a request to shore to requisition a cover for the SPS 3 autosampler. 30 Apr 2019 The cover will be sent for the next expedition.

ThermoFisher FlashEA Elemental Analyzer

• 07 Nov 2019 The software crashed when we used  and measured BBOT as a check standard. Given the high content of sulfur (7.44%) and carbon (72.53%) in this standard, we intend to weigh it as to be below 5 mg. It worked well at the beginning with a 5-6 mg, but it started to give the error window during the run, as shown in the screen shot below. 
  
• 11 May 2019 Loss of ship power lead to some erroneous check standard and sample results. On power loss the instrument shuts off gas flow, on instrument startup the initial gas pulse occasionally may push the top section of quartz wool to rise outside the furnace hot zone (the same thing happens sometimes during leak checks). Subsequent samples did not fully combust, and doublet peaks or peak tailing appeared in the chromatograms (see pic below). We removed the autosampler and pushed the quartz wool/combusted tins back into the hot zone using the stainless steel column packer, and resumed the sequence.
  
• 16 Apr 2019
  • CHNS TCD detector baseline show sporadic fluctuations  on the order of ~8 uV, whereas before the random noise was on the order of ~1-2 uV. In contact with vendor. Changed the sensitivity of the software's auto-integration feature to ignore signal changes below ~10 uV to temporarily solve the issue.
    
  • Etienne exchanged the cooling fan.
  • Opened the chassis and cleared the interior of dust with compressed air.
  • Replaced the MAS piston, which had dirt and grime buildup around the O-rings. A software leak test performed afterwards showed carrier leak of 0 ml/min and reference flow leak of 1 ml/min (both below the threshold of 3 ml/min).
    
    

UIC Inc. CM5015 Coulometer

• 07 Sep 2019 When running samples for the JR Academy the coulometer cell current would not revert to 0.0 after a measurement was "finished", which meant the µg C titrated would increase perpetually. The cell current ranged from 0.1-0.9. The ET's had to check several voltage setpoints on one of the interior circuitboards to verify lamp voltage was within the required limit of 2.45 V. It was reading ~2 V, so they adjusted the onboard potentiometer RP1 to within setpoint. The onboard potentiometer RP3 was reading ~9.3 V, from a required 9.60 V DC max, so it was not adjusted. The UIC technical bulletin which explains how to check the circuit boards (TB #70) is listed on this page with the other Coulometer references.

Agilent 7890A GC NGA1

• 22 Jul 2020 The current method to use is 378T_NGA1.m
• 04 Dec 2019 Changed out the molesieve column as per the Lab Officer's recommendation. The old column was placed in the new one's box and clearly labeled "used but good". It is stored in the column cabinet and will be fine to use as a spare as there did not seem to be anything wrong with it. The behavior of the GC continued the same with the new molesieve column in place. 
• 31 Aug 2019
  
  • The method to use now is based off of NGA_UPDATEDVALVE_STANDARDMETHOD.m. Using a prior method will result in FID baseline disturbances due to erroneous valve and pressure control module (PCM) configurations in the software.
  • We see some carryover after running high standards (I ran STD C last night and did a "No Injection" today and still found peaks for C1-C6 in the FID signal). So we need to troubleshoot that still.

Agilent 7890A GC NGA2

• 22 Jul 2020 The current method to use is 378T_NGA2.m
• 09 Apr 2020 The good FID EPC makeup He pressure transducer that had been “borrowed” from the TCD module in December was returned to the TCD module. The rest of the FID EPC was removed (kept for spare parts) and replaced with a new FID EPC module. The module has an electrical connector in its bottom left corner. There is enough length in the cables to slide the module all the way up before it becomes necessary to unplug the connector. 

Once the replacement is complete, on power-up the instrument gives an error message "Front detector fault pneumatic module". You have to first unlock the configuration locks if they are locked:

Next, unconfigure and then install the FID. Follow the instructions below:

At steps 4 and 6 you have to press Mode/Type twice before it takes effect. The GC was not rebooted after the FID module removal but only after the installation. Power up was successful and the method was sent to the instrument without any issues. Both detectors are now fully functional.

• 24 Dec 2019 FID makeup He was not able to reach the method flow rate 25 ml/min. It fluctuated around 23 ml/min first, dropping steadily over the course of a day to 9 ml/min. This caused for the flame not being able to light unless the FID makeup He was turned off. The highest makeup He flow rate the instrument was able to maintain with the flame staying on was 5 ml/min. The faulty H2 pressure transducer (see note below from 31 Aug 2019) had been installed on the FID makeup He line in August. This transducer placement worked for a short time but the transducer now seems to have deteriorated further to the point that it cannot even support the makeup He flow. As another temporary solution to the problem the faulty transducer was removed and swapped with the makeup He transducer from the TCD side. We currently cannot operate the TCD on NGA2 until we get a replacement pressure transducer. Two manifolds have been ordered and should arrive in Panama. The method was modified so that neither the TCD nor its makeup or reference flow get turned on when the method is loaded. This method is called 378T_NGA2_TCDOFF.M. IMPORTANT: When the replacement FID EPC manifold arrives, it is crucial that we first remove the good TCD makeup He pressure transducer from the FID makeup He line and return the component to its original location in the NGA2 TCD EPC manifold. See photos below from 31 Aug 2019. The component in question is pictured in the second image from the left "H2 gas weldment". 
  
• 10 Dec 2019 The GC3 section was removed from the lab notebook. It had no entries in it.
• 05 Dec 2019 GC3 was replaced with NGA2. The NGA computer now runs both NGA1 and NGA2. We have two monitors for running two instances of the software. Both NGA1 and NGA2, online and offline versions of ChemStation, can be run from the OpenLab  Control Panel. See the screen capture below. 

• 31 Aug 2019 Installed the recently shipped shore-based NGA2 with the updated valve configuration. On installation we found high noise in the FID signal. The following is a list of steps we took to isolate the issue:
  
  • FID baseline fluctuates on the order of ~0.1 pA, whereas the old instrument was ~0.01 pA. This caused the software to auto-integrate the noise and identify them as analytes.
  • The FID peaks do elute at the correct times but the
  • Found a gas leak at the inlet of the HP-Molesieve column, replaced the ferrule, and re-tightened the connection after heating the oven to 200C.
  • At 35C the FID signal is 35+/-0.5 pA. To check if it was a column or detector issue we disconnected the column from the FID inlet and replaced the connection with a sealed Valco nut (some of these in the GC mini drawers on the shelf). There was no change in the baseline, so it was not a column issue.
  • 
  • Performed a current leakage test. Shut off the FID using the GC front control panel. The value came back 0+/-0 pA, which says that the electrical is working fine. A side note here, I removed the FID castle to inspect the tip of the electrometer. There is a metal spring coiled around the end which can slightly deform when putting the castle back into place. If there is current leakage, as a first step correction remove and re-situate the FID castle before troubleshooting the electronics.
  • 
  • Changed the FID column jet and cleaned the FID, there was no change to the FID noise level. The FID baseline is up to 44+/-0.5 pA.
  • Turned off the FID, plugged the top of the FID with the Agilent FID flow measurement tool (which itself was plugged). Turned off the FID gas flows using the front panel, individually varied the flow rates of Air, H2, and He, and monitored the reported flow rates of the other two gases. If the column jet is situated properly in the FID then the other gas flow rates should increase because the FID is plugged from the top and the bottom so the one varied input gas creates a back pressure in the gas lines of the two that were shut off. This test came back fine. The FID components are seated properly.
  • Removed the plug of the FID measurement tool and connected a flow meter. Stepped through the FID gases and measurement them against their setpoints. All came back fine.
  • We noticed that the readings for Air (400.00 mL/min) and He (25.00 mL/min) were stable to the hundredth decimal place, whereas the reported H2 flow rate was fluctuating around its setpoint of 40.00 by ~0.1 mL/min-which was too small to pick up using our external flow meter. We increased the H2 generator output by ~2 psi and stepped down the regulated H2 gas input to the GC from 60 psi to 50 psi (regulator on the wall) thinking that there was some pressure buffering effect in the line. This didn't change the H2 flow rate noise.
  • We then removed the gas lines to the FID pneumatics module on the rear of the instrument and inspected the module for problems. We swapped the H2 valve solenoid with the He valve solenoid, but it didn't resolve the issue (FID 35+/-0.1 pA). So we decided to test the pressure sensor. We disassembled and sonicated the H2 pneumatics weldment in methanol, replaced the O-rings and swapped the H2 pressure transducer with the He transducer. When we reassembled and reinstalled the gas lines we noticed that the H2 flowrate was stable at 40.00 mL/min whereas He flow was fluctuating (25.00+/-0.1 mL/min). However, the FID signal was greatly improved at 10.0+/-0.05 pA (Unlike H2, the He contributes minimally to flame strength). There is still a little noise but this will be resolved when we order a new pressure sensor to replace the shoddy one connected to the He line.
  • 

Agilent 6890 GC with Gerstel Autosampler

• 16 Apr 2019 When using the Gerstel autosampler, the labels attached to the vials should be the small ones used in the Microbiology Label Printer. The labels should be attached to the top half of the vial. The larger Chem Lab labels impede the autosampler from inserting the vials into the oven and back into the vial rack. The AS will either have a Z-axis error and abort the sample sequence, or it will believe it hasn't injected a sample (when it may have) and then proceed on to the next sample immediately before the previous one has finished.
• 16 Apr 2019 PFMD tracer standards yield a non-linear calibration curve. When making dilutions, you need to extract STD A very slowly using the 1 µL syringe, and pipette the samples slowly into sealed headspace vials. Since the volume of standard is so low, only a small bead will form on the tip of the syringe needle. It is best to have the tip touch the inner wall of the vial to get the bead to come off, otherwise the hexane solvent starts to evaporate and the PFMD present in the bead adsorbs onto the needle tip and doesn't remain in the vial with the needle is removed.
• 13 Apr 2019 Issue with the a electrode prong on the syringe heater block getting stuck and not making full contact with the autosampler. The syringe block temperature never could reach the 70°C setpoint. Johanna bent the prong back into place.

Metrohm Professional IC

• 24 Mar 2020 The IC was reported to have run out of both anion and cation eluent on 378S. The instrument had been on for several hours aspirating air through the entire anion and cation flow paths. When started up, it seemed to be operating as normal. The anion peaks were where they were supposed to be, but the cation peaks were coming out consistently late. Very reproducible and very late. Instead of the normal 15.5 minutes, the magnesium peak was coming out past 17 minutes. The cation baseline was also approx. 50 µS/cm off the usual value - around 850 µS/cm instead of 900 µS/cm. After purging the high pressure pumps, changing out the inline filter and making several new batches of cation eluent (one of them with freshly opened new container of PDCA, just to rule out old PDCA) and still not seeing any improvement, the cation column was eventually changed out. It had 2200 runs on it already, so the column could have been showing signs of aging. Unfortunately, the new column did not make a difference either. Finally, after countless test runs and calibration attempts - on the 24th of March, using the same cation eluent batch as the previous day, the baseline and the retention times came up normal when the instrument was turned on in the morning. A full calibration was run, and everything came out as it should. At this point it can only be assumed that there still was an air bubble somewhere in the flow path, and it finally dislodged. The IC is currently fully operational and is not exhibiting any further troubleshooting issues.
• 15 Mar 2020 While vortexing IC vials, leakage is occasionally experienced through the lid. When bringing this up to Metrohm instructor Ashley Wittig, she stated that those lids are not designed to be waterproof. She recommended we vortex with parafilm and only put on the lid after vortexing to avoid potential skewing of the results and/or cross-contamination.
• 31 Dec 2019 Replaced ascarite in CO2 scrubbers with soda lime. This was recommended by Ashley Wittig, an IC class instructor at Metrohm. Soda lime is much less hazardous than ascarite, and it does not cake up. It should work better and last longer than ascarite. Soda lime turns from white to pink when the absorbent is near the point of exhaustion. The absorbent should be changed out when 2/3 of it has changed color. The soda lime container is stored in the Drierite cabinet.
• 22 Aug 2019 Finished worksheet for creating sample tables in Excel which can be imported into MagIC Net. You can also copy the data exported from the softare (Table Format) into the file and convert the format into something more presentable to the scientists, and convert the data into Spreadsheet Uploader format in one go.
• 20 Aug 2019 Calibrated the IC for ammonium. The peak elutes immediately after sodium and is not present in samples/standards lower than 1 mM (when they are follow the regular 1:100 dilution. The calibration was applied retroactively to the last IC run of JR100, and the samples have a ~20% bias towards higher ammonium values than the SPEC. Will look into why this is.
• 10 May 2019The check standards following a blank show a depressed concentration of calcium and magnesium compared to check standards following samples. Looking at the calibration curves it is apparent that at low concentrations there is a small drawdown of calcium--most likely from adsorption onto surfaces in the flowpath. Further check standards return 100 ±5%. Keep this in mind when interpreting instrument performance. If blanks are run in sequence it can be expected that the calcium concentration of the following sample will be skewed lower. The flowpath must be conditioned.
• 29 Apr 2019 Rerunning a set of samples from U1536 because of some spurious results in Na, Cl, and Br in comparison to ICP results. Changed the in-line filters and anions guard column–the parts removed still looked clean on the inside, however.
  

Metrohm Brinkmann Autotitrator (Alkalinity)

• 27 Mar 2020 Thiourea cleaning procedure was successfully performed on two electrodes that had had so much AgS₂ precipitated in the diaphragms that the diaphragms had turned black all the way through. The electrodes were treated with freshly prepared 7% Thiourea in 0.1 M HCl.
  Procedure:
  Weigh out 0.7 g of Thiourea.
  Dissolve in 0.1 M HCl in a 10 ml volumetric flask. Bring flask to volume with 0.1 M HCl.
  Pour enough solution into a 10 ml beaker to cover the diaphragms. Support the electrodes with a 200 ml beaker.
  A significant change was noticed within the first 15 minutes. The diaphragms started to turn gray, and within an hour they were white on the outside. After the initial rapid change it took four additional days for the rest of the diaphragm to turn white. See the second image indicating how the diaphragms stayed black on the inside until they finally turned white on day 4.
  The electrodes were cleaned with DI water and left conditioning in KCl for two days prior to being tested. Both electrodes are now fully functional. They are stored in DI water in the spare electrode rack in the cabinet above the alkalinity station with labels indicating that they have been "refurbished".

• 30 Jul 2019 Bug in the alkalinity software: When you are at the "Edit Rates" or "STANDARD Manager" menus, and you select a new rate or correction factor, and then hit "Cancel", the most recently selected rate/factor will be written to the ALK parameters file and used for subsequent samples. You need to navigate back to either of those menus and select the correct rate/correction factor to restore the software.
• 19 Apr 2019Replenished the storage solutions of the Metrohm pH probes, Ag probes, and one ion probe. The probes should be stored in DI water long-term, as potassium from the KCl solution will ultimately damage the glass membranes (see Storage of pH Probes). They had been previously stored in KCl and the salt had precipitated around the top rim of the probe's plastic sleeve. The pH probes must be allowed to equilibrate in their respective electrolytes for several hours before use (KCl for pH probes, water for Ag titrodes). This may explain why the first several measurements of IAPSO for the correction factor gives wonky results, at the beginning of the expedition.

Metrohm Autotitrator (Chloride)

• 07 Aug 2019 Replicates of IAPSO were having poor precision. The silver nitrate dispenser tip was cleaned by placing it in a sealed vial with a few mls of ammonium hydroxide (this should be done in a fume hood). Below are before and after pictures of the probe tip and frit. Additionally, the silver electrode was situated a cm or so above the stirrer bar in order for bubbles on the stirrer bar to degas, otherwise they partly envelope the electrode tip for the duration of the titration.
  
• 30 Jul 2019 Issue with the software creating test numbers. For several samples LIMS would generate a test number of "0" when the chloride values were saved. These tests would not display in LORE. To correct this, click "View History" in the titration software, find the samples with test numbers of "0" then re-upload the results using Spreadsheet Uploader.

OI Analytical TOC Analyzer

• 24 Aug 2019 Attempted to get the TOC analyzer up-and-running for Guaymas. The following is a maintenance adventure:
  • The gas-resisting metal frit of EFC Blank is either deformed or partially clogged and the outflow is essentially 0 when it should be 30 mL/min (input gas is 640 mL/min). Tried sonicating it in methanol overnight as per the OI recommendations, to no avail. Neither contact cleaner nor acetic acid helped either. Send a requisition to ALOs to purchase 2 more to be shipped for x385T. The rest of the PM will need to take place in x385 port call once the component is replaced. The pressure reading on the side of the EFC (Port D going to IR chamber) is steady at 20 psi, indicating that gas is still permeating the frit and the instrument is otherwise working properly. Exchanging the EFC Blank takes one minute.
    
  • Changed out the copper and dierite scrubbers on the front of the instrument.
  • Removed the Ascarite scrubber and added fresh Ascarite. The topmost portion of the Ascarite was bonded, so only that portion was exchanged.
  • There was an issue with the Chamber1 (C1) drain not fully removing all solution from the reaction chamber after an uptake. The drain connection on the backside interior of the chassis had rusted over. I used contact cleaner to break the bond and then flushed the outgoing line with compressed air. Then removed the C1 connection to the drain valve (on the front of the instrument), enabled the C1 manual drain using the software, and blew out the path with compressed air once the valve had opened.
  • The carrier gas going to C1 seemed to be hindered by a blockage. I removed all the lines going to the Medusa (the multi-port valve on the front of the instrument, and blew out its ports with compressed air while stepping through the valve positions using the software.
  • Performed an EPC calibration (Electronic Pressure Control). The 25 psi setpoint originally came out at 26 psi and after calibration the 20 psi setpoint came out at ~19.5-20 psi on the gauge, which is appropriate.
  • The syringe connection to the Medusa was rusted over so I removed and cleaned the rust with contact cleaner, then reinstalled the syringe and performed a syringe calibration using the software.
    
  • Removed and inspected the gas connection for the EPC out (coming out of the front panel) and found some rust in the frits. Used contact cleaner, then methanol, water and lab air to clean and purge the frits. When I reconnected the lines, the carrier gas was bubbling more vigorously in C1 and the C1 Drain finally functioned properly. Note: the system applies a back pressure to help push out draining fluids from C1.
    
  • Removed the autosampler needle and needle sheath and exchanged them for new ones. The previous one had rusted over and top flange of the needle had broken off (see pic). Additionally, the plastic housing the needle sits in was rusted over, so I removed it and soaked it in a beaker with 10% HCl overnight to remove the rust.
    
  • Changed the 0.2 micron Acrodisc PFTE inline filter between the EFC Blank and the IR detector, located within the instrument chassis. If the filter is bad the IR signal will start to decrease, indicating that water is getting to the detector.
    

Labconco FreeZone Freeze Dry System

• 30 Jul 2019 Recommendation by sailing chemist Becky Robinson to freeze samples in -86 before freeze drying to quicken the time to dry. Did not end up doing that but took samples out of the sample tubes (leaving tubes in sample bags) and broke them in smaller pieces instead. This works very well for shortening freeze-drying time. Not recommended unless in a hurry. Unnecessarily messy.
  
• 30 Jun 2019  Ship voltage dropped. Adjust the voltages range limit by increase/decrease 2 increments.
• 17 May 2019 Changed oil in the vacuum pump.
• 09 May 2019 Update on the EMF Clean Application Oil Drain Kit. Eric spoke with our supplier. He did not recommend returning the oil collected in the oil mist filter reservoir back into the pump. He said the pump would have to work harder with the oil drain installed. He also recommended just discarding the oil collected in the mist filter reservoir (which we have been doing anyway).
• 28 Apr 2019 Water accumulating in vacuum pump oil (ongoing all X382). There was a request from a scientist early into the expedition to freeze dry this scientist's personal samples, 815 of them, frozen 3cm 25cc QRNDs. These very large volume samples had been frozen in one piece, thus have a very small surface area. A large surface area is important to successful freeze-drying. Only a small number of aforementioned samples could be freeze-dried at one time, maximum ten, and even then it took a very long time, approximately 48 hrs. Excess ice accumulated fast on the collector coil, requiring daily defrosting. A lot of bubbles were observed in and on top of the vacuum pump oil. Also, with the pump working hard 24/7, the oil was quickly collecting in the oil mist filter reservoir. A request was made for an EMF Clean Application Oil Drain Kit. It would continuously return the oil trapped in the oil mist filter to the pump. This would eliminate the time-consuming need to manually return the oil back to the pump.
  It might be a good idea to come up with a "Freeze-drying of Personal Samples" policy, potentially limiting the freeze-dryable sample size to 5 cc to ensure proper and timely freeze-drying process while also avoiding overloading the system.
• 28 Apr 2019 Changed oil in the vacuum pump and replaced the filter element in the oil mist filter.

Hydrogen Generators

• 30 Apr 2019 Generator beeps warning if the water level exceeds the max level indicated on the plastic reservoir. If the ship roll is significant, the alarm will sound due to water sloshing above the max level. Pipette water out to get it to stop.

Weatherford Source Rock Analyzer (SRA)

• 24 Aug 2019 When adjusting SRA flow rates, it is essential to connect the oxidation air and helium line directly to the flow meter. The fitting for this connection is in the flow meter box, marked "SRA ox air & He flow meter connector". The accuracy of these flow rates is crucial for reliable ignition of the FID flame.  
  

Eppendorf Centrifuge 5810

• 22 Nov 2019 Replaced rotor and buckets. The manufacturer recommends replacing the rotor every 7 years.

Barnstead Water Purification System

• 26 Jul 2019 Manually tested the water level indicator. It currently functions as it should. No indication of previously reported issues.
• 26 Jul 2019 The Barnstead Water Purification System quit making water. "Check inlet" was displayed all day long. No water was flowing into the tank, despite several flush cycles being run. There had been no disruptions to incoming water flow either. Nothing made a difference until the CARBON CARTRIDGE was changed out. It was only two months old. Also changed out the PREFILTER, which had turned dark brown even though it also was only two months old.
• 03 Jun 2019 Ongoing issue with RO water reservoir. Possible malfunction of water level detectors. RO water would constantly indicate low level but the reservoir would be full and water would be flowing out of the overflow tubing. Took the top panel off of the reservoir to take a look at things but made no changes. After panel removal and replacement, it seems to work better in that it would stop when the reservoir is full. The lights for reservoir level would show full, but would be flashing. It cycles between that state and the low water level state as RO water is used. So at the moment it's just an indicator for full, or not full and is not indicating actual water level.
• 16 May 2019  Changed both the RO membrane filter cartridges, the prefilter, the MPS cartridge and the carbon cartridge. Replaced the nozzle filters on the dispensers. The instruction manual mentions that there should be less than a 5 psig differential between the prefilter inlet and prefilter outlet pressure gauges. The inlet gauge has a reading of 20 psig, but the prefilter outlet is showing 0 psig, even though the system has good flow. The pressure reading on the carbon outlet is 8 psig–which should be within 5 psig of the prefilter outlet. The filters are connected in series, so it does not make sense that the prefilter outlet would show no pressure unless the gauge itself is faulty. The RO membrane pressure gauge is stable at 58 psig. The carbon and membrane timers were reset. Below are before and after pictures of the prefilter, MPS and carbon cartridges:
  
  

Cahn Balance

• 07 Apr 2020 When it becomes necessary to ship a Cahn Balance, utmost care should be taken to secure the hangdowns "just right". See image.

• 07 Apr 2020 NOTE: When turning on a Cahn balance: If the balance is displaying 8888888, press the TARE button on the balance to activate it. Otherwise, do not press TARE on the balance. Use the tare feature in the software.

• 07 Apr 2020 Cahn balance tag#51470 was experiencing a delay in stabilization after a weight was placed onto the scale. There were no issues if you placed a weight and waited 60 seconds and then clicked "Tare", but if you clicked sooner than 60 seconds, the graph would come out looking similar to the image below. The balance was opened up, and the shaft was very carefully lubricated with CRC ELECTRICAL GRADE 2-26 Plastic Safe Multi-Purpose Precision Lubricant while simultaneously moving the scales by gently blowing onto them. This made the occurrence of the issue less frequent, but as it was still present after the lubrication and several test weighings, a decision was made to send the balance back for calibration and service. A tag on the balance indicates that it was last calibrated in 2005. The balance was temporarily replaced with the tag#52427 unit, which is not exhibiting any of the aforementioned issues.

Carver Presses

• 31 Dec 2019The paintable parts of the presses were sent off to be powder coated in Fiji. They came back black as requested, but it was not a powder coat. The paint has already chipped in a few spots. 
  

• 08 Nov 2019We ran out the available 0.2 um (LS0413). All we have in the UTS are the expired (by 05/2016) ones which apparently is causing problem during the squeezing process. The filter paper inside came off during the squeezing process, it caused the IW to pierce through the filter and spilled out of the syringe. The color also changed as shown in the picture below (left one is the expired syringe filter and the right one is the good one) indicating they are not okay to use anymore. We might need to purchase a new batch.
  

• 31 Jul 2019 There was a leak coming from the seal of the middle carver piston. Etienne mentioned that when the release valve on the front of the piston is turned, hydraulic fluid leaks out from inner metal sleeve (in which the actual piston rod resides) and into a chamber between the sleeve and piston's outer jacket (see 3rd picture). We removed the platen and the top of the piston with the frame still in place using several monkey wrenches from Downhole, then sucked out the extra hydraulic fluid. We then removed the jacket and cleaned the insides with paper towels, replaced with jacket and reconnected the top plate and platen. The AutoPak reservoir was filled with replacement hydraulic fluid. We will need to make a permanent fixture to prevent the release valves from being turned in the future. If the release valve is turned and the piston again fills with fluid, one temporary solution to draining the piston without taking it apart would be to hand pump it (ensuring the front release value is closed), as hand pumping draws fluid from the outer chamber instead of through the AutoPak line. One final note: the outer jacket has an o-ring-like seal at its base but otherwise is only sealed from leaks with a special glue. This chamber is never at high pressure.
  

AMI Oxygen Analyzer

• 03 Dec 2019 The oxygen analyzer was replumbed. The nitrogen now tees off to the coulometer before the oxygen analyzer, and the oxygen analyzer vents into the lab.