Antares Analytical Bulletin for September 2002

Subject:  Vacuum Systems & their Maintenance, Troubleshooting and Basic Repair

Relevant Baird Models: All FSQ and DV Series Spectrometers, All ICP spectrometers (PST, ICP2000, ICP2070) all Foundrymate spectrometers with the ½ meter vacuum option.

Relevant Matrix: N/A

Background Information: Most matrices will be analyzing elements located in the far Ultraviolet wavelength range.

• These elements (P, S, certain B lines, Se, C and others) are using wavelengths located below 2000 Angstrom (200 nanometers) on the optical spectrum.
• These elements will not transmit in atmosphere because of the absorption of these wavelengths by oxygen. An inert gas atmosphere (most commonly argon but nitrogen can also be used) or vacuum is needed to effectively transmit these wavelengths.
• Baird spectrometers use only the vacuum system (except the Baird OneSpark which uses an argon purge system), so this bulletin will cover the vacuum systems only . The general principles will apply to vacuum spectrometers from most manufacturers.
• The normal operating vacuum range for a Baird spectrometer is a vacuum pressure between 1 to 30 microns as indicated on the front mounted vacuum gauge, or indicated on the MC20 “Spectrometer Status” program
• The high voltage supply will be interlocked to the vacuum level,  to prevent damage to the PMT detectors. Please see earlier Tech Bulletins for more information about PMTs and high voltage interlock systems.

Operating Performance:  When the vacuum system on your Baird spectrometer is in good repair and working correctly, you can expect the following performance

• Normal operating vacuum between 0 to 10 microns.
• A pumpdown time from atmosphere to 50 microns in less than 30 minutes. There are exceptions to this pumpdown time especially if the cell has been at atmosphere for more than 1 hour or if the laboratory has humid conditions.
• If you have the earlier DV / ICP model with a vacuum pump motor controller, the cycle time (pump on - off – on) should not be shorter than about 20 minutes. This is assuming the motor controller setpoint (1^st needle on the vacuum gauge) is set for 15 microns and pumping down to 5 microns or less.

Components: The Baird vacuum systems (all models) consist of the following parts, all which must be working correctly to have correct operational vacuum:

• The Optical cell – This is the optical cell which is operated under vacuum
• The Vacuum Pump – Takes the atmosphere out of the cell and keeps the system at operating vacuum.
• Exhaust Filter: This is the filter / housing assembly located on the exhaust side of the vacuum pump. It’s function is to trap any oil vapors, especially during pumpdown. On some Baird DV models this was eliminated and a rubber hose was used instead. Later DV models went back to the filter. All FSQ use the rubber hose in place of the filter.
• The Molecular Sieve – This is also called the “trap” and is located on top of the vacuum pump. This trap is filled with either Zeolite pellets or a Zeolite filter (DV6 / FoundryMate with ½ meter option) which absorb moisture coming from the cell as well as other gases. Another function of the trap is to act as a “barrier” in case of a vacuum pump oil “blowback.”  For a while AlO2 (Aluminum Oxide) pellets were used in place of Zeolite, with poor results. Any spectrometer using AlO2 should switch to Zeolite pellets and this can be done with no modification, and without damage,  to the vacuum system.
• Gate Valve – This is the ¼ turn valve located above the trap on all spectrometers. It’s function is, when closed, to allow the vacuum pump to be serviced and not lose vacuum in the cell. When open it allows free flow of any atmosphere in the cell to the vacuum pump.
• Vacuum Hose – The hose located between the gate valve and the vacuum pump. This is a steel reinforced rubber / neoprene hose which is strong enough not to collapse when the unit is under vacuum.
• Various O rings and Couplers - These are used on the vacuum cell where metal joins metal (cell door for example) and couplers with O rings are used when joining the pump to trap to gate valve. These couplers are held in place with clamps and wing nuts..
• Vacuum Pump Motor Controller - This controller was used on the DV2-4, some early DV5 and some early ICP models. This was used to turn the vacuum pump off when a certain vacuum pressure was reached in the cell and turn the vacuum pump back on when the vacuum level in the cell climbed to a certain level because of air leakage into the cell. These “turn on / turn off” points were controlled by set points on the vacuum gauge.
• Note: This leakage is a common occurrence in any vacuum system. On later Baird models the Vacuum Pump Motor Controller was eliminated and the vacuum pump was left to run continuously.

Maintenance: The vacuum system on the Baird spectrometers was designed to be as low maintenance as possible, but some periodic maintenance is required.

• The vacuum pump oil should be changed once a year, as stated in the manual. Other than the pump oil, the vacuum pump normally won’t need any maintenance, unless there are vacuum problems.
• The manual says the Zeolite pellets / cartridge should be changed once a year. Our practical experience is that, unless you are having vacuum problems (see below) once every two years is sufficient. Never should the Zeolite pellets / cartridge be left in the trap longer than 2 to 21/2 years.
• When Zeolite pellets / cartridge is changed the O rings on the couplings should be removed and checked for wear. When replacing a THIN layer of vacuum grease should be used on the O rings. This layer should not even be visible to the eye. There should be nothing more than a small shine coming from the O ring surface. More IS NOT better and will only cause vacuum problems.

Common Problems and Solutions

Vacuum pressure “creep.”

• This problem happens over a long period of time when the vacuum pressure in the cell slowly rises, month by month, until the pressure is around 30 to 40 microns

• Cause: This is usually caused by the Zeolite becoming saturated with water and oil vapors from the cell. This is similar to a sponge being full of water. The Zeolite, like the sponge, can’t absorb any more moisture so it stays trapped in the cell, affecting vacuum pressure.

• Solution: For the DV6 and Foundrymate using a Zeolite Cartridge, changing the cartridge is the only solution. On the FSQ and DV / ICP models using Zeolite pellets it is possible to “bake” the pellets using the built in trap heater. This must be done with a specific procedure (see “Trap Heating” procedure below) and this procedure must be followed EXACTLY or vacuum problems will happen.

 White Vacuum Pump Oil.

•  Cause: Many times, especially in a humid environment or after the Zeolite has been “baked”, the vacuum pump oil may  become white and look like milk. This does not mean the oil needs to be changed. This is only suspended water vapor and oil mist and can easily be removed by a “filter” type action of the pump.

• Solution: Close the gate valve and leave the pump running. On the right, or front, of the vacuum pump will be a small knob. When this knob is turned clockwise it allows the trapped water vapor to be released to atmosphere. Let this run for 1-2 hours and this should clear most of the vapor from the oil. Close the knob and let the pump run for 5 to 10 minutes before opening the gate valve. NOTE: This will probably need to be done after “baking” the Zeolite.

• Close the gate valve above the pump.
• Turn off the high voltage
• Locate the power plug for the trap heater. On the FSQ this will be a standard 110 or 220 VAC plug and needs to be plugged into the power strip located on the spectrometer
• On the DV and ICP series of spectrometers, this should be a small round plug (male) which mates with the female end hanging down from the spectrometers frame. The other end of the female plug is already plugged in and connecting the male / female ends will start power going to the trap heater.
• WITH THE VACUUM PUMP RUNNING, let the trap heater heat for 5 to 6 hours and no more. The trap will get very hot and this is normal.
• After 5 to 6 hours unplug the trap heater and let the trap cool until it is cool to the touch. Do not open the gate valve until the trap is cool.
• After the trap heater has cooled open the gate valve and let the unit pump down. If the vacuum pump oil is white, filter using the vacuum pump “vent” switch described above.
• If the vacuum pressure doesn’t go below 15 microns, other vacuum problems or maintenance may need to done. Do not “bake” the Zeolite again.
• Turn on the high voltage.

More (problems, solutions etc) on Baird Vacuum Systems will be continued in the October 2002 Antares Technical Bulletin

____________________________________________________________________________

This Technical Bulletin is written by Gerry Smith, Technical Services Manager of Antares Analytical. Having spent 22 years with Baird in various positions, including many years as Final Test / Calibration Manager and Senior Service Engineer, International Operations, he or any of our Technical Support Staff, are well qualified to answer any Technical or Applications problems you may have.