FTS Laboratory User's Guide



Careful consideration should be given to the question of what to ship to Tucson prior to an observing run. Although the National Solar Observatory is primarily an astronomical facility, it is possible for the FTS lab to provide some of the most common laboratory source support equipment and services such as vacuum pumps, certain power supplies, etc. However, experience has shown that set-up time is often greatly reduced when visitors bring or ship the majority of the critical elements for their experiment from home. The final decision regarding what should be shipped should be made after reviewing Section 4 of this manual and discussing the specific needs of the experiment with NSO's support staff.

Shipments should be addressed to:

Equipment shipped from foreign countries should, if at all possible, be sent under an ATA carnet. This considerably simplifies customs procedures both when the shipment arrives and when it is returned. If the shipment under a carnet is appropriately addressed, NSO personnel can receive it and see it through customs without obtaining the services of a customs broker.

Please advise NSO of the details of the shipping arrangements and the approximate arrival date of the equipment so that the Shipping and Receiving Department can be notified in advance. All shipping or customs expenses are the responsibility of the visiting investigator. Any expenses incurred by NSO will have to be back-charged to the visitor's home institution.


Visitors coming to use the McMath FTS are asked to arrive in Tucson on the day prior to the first scheduled date of their run, spending the night, for example, at the Plaza International Hotel. Transportation from the airport to the Plaza can be provided by an airport limousine, the "Arizona Stagecoach." NSO's Tucson offices are then within easy walking distance (3 blocks west, 2 blocks south) of the Plaza. The user is referred to the Kitt Peak User's Handbook for more details.

The National Optical Astronomy Observatories provides shuttle bus service which can provide transportation from the downtown offices to the mountain on the morning of the first day of the observing run. The first convenient bus of the day leaves Tucson at 8:30 a.m. and arrives on site at about 10:00 a.m.


In general, an NSO staff member will provide full-time technical support during data acquisition. This instrument operator will have sufficient experience with the spectrometer to take full advantage of its versatility for a particular observation. Further, instrumental system problems can be recognized and evaluated, should they develop. As a result, the visiting investigator will be able to give full attention to the laboratory source, sample handling apparatus, et cetera. All details of the observing run should be communicated to NSO prior to the run.

Some fraction of the first scheduled day of the run will be spent on source setup and other instrument preparation. Data acquisition can begin as soon setup is complete.

In any event, visitors should plan to spend all nights of the scheduled run in mountain dormitory rooms and eat all meals in the mountain dining facilities. Questions relating to both Tucson and Kitt Peak accommodations should be directed to the Kitt Peak Observing Support Office (520/318-8397) or the NSO Director's Office (520/318-8294).


The number of spectra recorded on a given day depends on the resolution, the type of source, and signal-to-noise ratio desired. A typical full resolution scan requires about eight minutes to complete and usually a minimum of eight such scans are co-added, requiring a total of about one hour. In one hour, signal-to-noise ratios on the order of 104 for emission and 103 for absorption spectra can be obtained in many instances if the source has sufficient intensity and stability. Past experience would indicate that four to eight spectra of this type can be collected in an average day.


Some visitors prefer to leave Tucson late in the afternoon of the day following their run, while others spend one more night at the Plaza International before departing. Shuttle buses depart the mountain four times per day on week days and twice daily on weekends providing transportation back to the Tucson offices or to the Tucson Airport if requested.


The interferograms will be transformed on site at the time the observations are taken. The data format that is supported at the FTS is FITS and the media supported is 8mm exabyte tape.



2.1.1 Optical Overview

Figure 2.1 (below) shows the optical layout of the one-meter Fourier Transform Spectrometer. Click on the image for a higher resolution view.

It is a folded Michelson interferometer with a total internal path of approximately 12 meters from inputs to outputs. In practice, only one input and both outputs are utilized. The two outputs have complementary phases so that the difference in the signals produced by two matched detectors is used to record the interferogram. The sum of the signals is almost unmodulated and can be used to ratio out small d.c. source fluctuations when appropriate.

The source is focused onto the plane of a circular aperture (generally 8 mm in diameter) and is accepted at the collimator as an f/50 beam. After passing through the interferometer, the collimated beam is reimaged onto the detectors with about a 15% image scale reduction.

2.1.2 Vacuum Operation

The 12 meter path within the FTS between the input port and detector mounts is housed in a vacuum chamber. Vacuum within the tank during operation falls between 10-2 and 10-1 Torr. Vacuum operation has many advantages including reduced atmospheric absorption, vacuum wavenumber results without index of refraction corrections, improved acoustic decoupling of the interferometer, and the elimination of convective currents.


The resolving power of a Fourier transform spectrometer is limited in general only by the total path difference obtainable with the spectrometer. Hence, a 1-meter FTS is, in principal, capable of a resolving power of 100s where s is the observed wavenumber in cm-1. This assumes a symmetric interferogram (the central fringe or zero-path-difference point occurs in the center of the interferogram).

In practice, however, the resolution of the McMath FTS is ultimately limited in the visible and near infrared by the number of data points which can be stored by the magnetic disk associated with the on-line computer. The current,approximately, 2 million point disk capacity in fact sets the resolving power limit at 1,048,000 for all symmetric unaliased interferograms shortward of 10000 cm-1 (1 micron).

For the UV, visible, and near IR up to 4 microns, a useful approach when more resolution is needed involves taking data in second or third order alias. If the passband of interest is sufficiently narrow, it is possible to undersample the interferogram, taking only every other or every third point, and still recover the spectrum completely. This technique can be used whenever

    Delta(o)/Sigma(max) is equal to or less than 1/2

Where Delta(o) is the total width of the passband and Sigma(max) is the largest wavenumber (shortest wavelength) detected. In this, way resolving powers up to 3.0x106 can be obtained in the "disk limited" regime while maintaining symmetry in the interferogram.

The catseye mirrors which make up the moving arms of the FTS are generally configured such that the zero path difference point occurs in the center of their travel. Hence, in this mode it is only possible to obtain path differences up to 0.5 meters since the total effective travel is only 1 meter. This makes little difference in the disk-limited resolution regime, but at longer wavelengths (above 5 microns) the FTS is often configured in the "offset catseye" mode to allow the full 1-meter pathlength to be used. This, of course, precludes the acquisition of symmetric interferograms, but is the normal mode of operation beyond 2000 cm-1 (5 microns).

Figure 2.2 (below) summarizes the resolving power limitations of the instrument. The sloping lines represent the pathlength limit for symmetric and offset catseyes, while the horizontal lines are the limits imposed by the data storage capacity.


The FTS is currently capable of obtaining spectra from approximately 550 to 40,000 cm-1 (18 microns to 2500 A). Several different optical and detector configurations are necessary to accomplish this, however. Although detector changes can be accomplished in minutes, changes made to the instrument's optics are complex and time-consuming and are thus scheduled in blocks of several weeks. Internal optical changes are not made during the course of an individual's run.

Figure 2.3 (below) summarizes the coverage provided by the various beam splitters and detectors available. The wavenumber ranges indicated represent typical operating regions.

2.3.1. Beam Splitters

    The available beam splitters are as follows:

  1. UV - Ultrasil fused silica with Aluminum coating (6500 - 40,000 cm-1)
  2. Visible - Dynasil fused silica with Silver coating (5,000 - 27,000 cm -1)
  3. CaF2 - Calcium flouride with GaP coating (1300 - 10,000 cm-1)
  4. KCl - Potassium chloride with GaP coating (500 - 4,000 cm-1)

2.3.2 Detectors

  1. Large Diodes - Diodes from United Detector Technology, used with manufacturer's windows removed. (2500 A to 1.1 microns).
  2. Midrange Diodes - Diodes from United Detector Technology. Their smaller active area makes them more effective at lower light levels (2500 A to 1.1 microns).
  3. Indium Antimonide (InSb) - Photodiodes from Cincinnati Semiconductor , used at liquid nitrogen temperatures (~7000 A to 5.5 microns).
  4. Arsenic doped Silicon - Photoconductor used at liquid Helium temperatures (to 20 microns).

Other detectors are currently being evaluated for possible use below 3000 A.


The current data reduction computer used in the FTS lab is a UNIX based SUN SPARCstation IPX. It is of sufficiant capability to make full band-width, full resolution transforms of the FTS's interferograms practical.

The data reduction computer, Corona, has 4 Gbytes of disk storage and has an EtherNet conection to the InterNet (IP#

It takes Corona about 1 1/2 minutes to perform a transform on a 2 million point interfrogram.


Although interferometers have the advantage of being able to measure data over a large optical bandwidth in a single observation, it must be remembered that the shot noise from all photons incident on the detector contributes to each data point in the recovered spectrum. It is therefore advantageous to optically isolate the regions of the spectrum containing the information desired. A sufficiently narrow bandwidth has the additional advantage of permitting data to be aliased (see Section 2.2).


Filters are placed either in the input beam or just in front of the detectors to limit the optical passband of the system. The input filter holder accepts 2 x 2 inch filters up to about 8-mm thick and can tip the filters between 00 and 450.

Visitors are encouraged to bring their own filters to supplement the sets available. Whenever possible, these filters should be wedged to prevent channel spectra or "fringing." A 5-milliradian wedge is generally sufficient. Tipping can shift the passband blueward so it is recommended that interference filters be selected which pass the desired wavelengths when tipped 150 or so (higher angles are needed for high index materials).


A double-pass zero-dispersion monochromator is available as an alternative to filters. An intermediate slit of variable width selects a bandpass down to a few percent of the central wavenumber. Four prisms must be used to achieve coverage from UV to about 13 microns. The prefilter does not work well around 1 mm and does not work at all above 13 microns.

The prefilter, when used, has about 16 meters of internal path. Although the volume can be purged (see section 4.9), these additional 16 meters of path will likely impact the overall purge quality. Prism changes will interrupt purging in a profound way, and it takes roughly two hours to recover to previous purge effectiveness.



A six-meter multipass White-type absorption cell is available for use by visitors.

The cell has a volume of roughly 600 liters and in its current configuration can work at pressures from zero to nearly 2000 Torr absolute. Cell pressure can be monitored by a 0-88 inch (Hg) Bourdon type Heise gauge, Baratron capacitance manometers or a 0-1000 milliTorr thermocouple gauge (see Section 4.7).

The pathlength of the six meter cell can be varied from 24 to 432 meters in multiples of 24 meters (pathlength = 24n where n = 1,2,3,...18). Pathlength changes are accomplished externally via a vacuum feed-through coupling.

The cell is evacuated with a large rotary mechanical vacuum pump coupled to the cell through approximately 6 meters of 2" copper tubing. Several hours are required to pump the cell from one atmosphere down to sub-milliTorr pressures when the cell is clean. Overnight pumping is often necessary when sample gasses are changed.

Figure 3.1 (below) is a schematic representation of the White cell illustrating the pumping, pressure monitoring and gas handling connections.

The gas inlet port to the cell is a 1/4" male Cajon VCO connector. (Refer to Section 4.6 of this manual for further information relating to gas supplies, regulators, et cetera.)

The vacuum gauge port is a 1/2" O.D. tube, compatible with the 1, 10, 100, and 1000 Torr Baratron gauges described in Section 4.7. A visitor-supplied gauge needs to be configured with either a 1/2" O.D. tube, a 1/2" Cajon Ultra-Torr connector or a 1/2" female pipe thread in order to couple into the system.

The nearly 10 meters of external path between the exit port of the White cell and the FTS can be purged with dry nitrogen. This can reduce the effects of atmospheric absorption by as much as a factor of 30. Section 4.9 discusses the purging system which is used.

Figure 3.1 The Six Meter White Cell



Many facilities and supplies commonly found around university chemistry or physics departments are not available on Kitt Peak. It must be emphasized that the spectrometer is located on the mountain 50 miles from Tucson, and the resources available there are limited and directed primarily towards the operation and maintenance of astronomical instrumentation.

The equipment which is available for laboratory support has been developed with versatility in mind, but it is impossible to anticipate all the diverse needs of our users. Any visitors planning to bring a source or a sample with them from their home institution should carefully review the descriptions and inventories in the paragraphs which follow. It is important to inform the support staff of your needs during the pre-run discussions and to seek clarification about anything which is unclear.

Experience has shown that much time is often wasted trying to lash up vacuum systems or gas manifords during a two- or a three-day run and, in the end, major fractions of the program are compromised. If the facilities described below cannot be easily adapted to your needs, strong consideration should be given to bringing or shipping the critical elements of the experiment from home.


The so-called "front end" of the FTS (Figure 4.1) consists of a rectangular surface 30" above the floor and 30" below axis. Its dimensions are 48" x 30".

A large disk, approximately 35" in diameter, surrounds the entrance aperture of the FTS. This is a part of the solar image guider, and has an array of 1/4"-20 tapped holes at 4" intervals.

The platform shown in Figure 4.1 can be mounted at the "front end" of the FTS. Its top surface is 10" below the optcal axis and has surface dimensions of 18" x 24" with an array of threaded holes (1/4"-20) at one-inch intervals. The platform is typically kinematically mounted with its center line three inches from the optical axis in the horizontal plane. Its long (24") dimension is parallel to the optical axis.

Triangular optical benches in 1/2-meter and 1-meter lengths are available along with a few bench carriers and lens holders. Lattice clamps, three-prong clamps and a selection of 1/2" aluminum rods are also available.

Figure 4.2 is a floor plan of the FTS room showing the FTS tank, the front end, stationary electronics racks, furniture, etc. The cross-hatched areas are the tops of sunken cable trays. Vibrating equipment (vacuum pumps, etc) cannot be set up on top of these tray covers as this causes mechanical vibrations within the instrument.


  1. Sylvania DZE - 150 Watt Quartz iodine projection lamp.
  2. Ribbon Filament Lamp - 15 amp with Sapphire window.
  3. IR Source - 450 Watt.

Figure 4.1


  1. Optronics Model 550 15 ampere radiance standard with calibration supplied from 2500 A to 6.0 micrometers.
  2. Argon Miniarc with calibration supplied from 2000 A to 3000 A.


  1. Sargent-Welch 1397 2-stage rotary pump. Free air displacement of 500 liters/minute.

    This pump is an integral part of a vacuum and gas handling station which includes a liquid nitrogen trap in the vacuum line. Associated plumbing contains glass, stainless steel, aluminum, brass, and viton.
  2. Sargent-Welch 1376 2-stage rotary pump. Free air displacement of 300 liters/minute.
  3. Sargent-Welch 1402 2-stage rotary pump. Free air displacement of 160 liters/minute.
  4. Vacuum Interfacing
    Options at pump end (all pumps):
    1. Leybold-Heraeus KF 32, 20, 10 flanges.
    2. 1/4", 1", 3/8" rubber hose connection.
    3. 1/2" Cajon Ultra-Torr connector.

    Rubber hose is available as well as some hose clamps. Two 36" lengths of stainless steel Cajon flexible tubing is available for use if the vacuum port on the application is one of the following:
    1. 1/2" female pipe thread.
    2. 1/2" O.D. tubing projecting at least 1".
    3. Leybold Heraeus KF 20 Flange.
    4. 1/2" Cajon Ultra-Torr connector.


4.6.1 Bottled Gas Supplies

NSO can supply visitors with gas samples if sufficient lead time is given to allow delivery from the commercial suppliers. Our suppliers are Matheson, Airco, and Line. We reserve the right to backcharge the visitor for expensive samples.

4.6.2 Regulators

The following step-down gas regulators are available:

Regulator         CGA Cylinder Connection          Common Applications

Matheson 3803               580                    Ar, He, Kr, Ne, N2, Xe
Matheson 3104A              580                    Ar, He, Kr, Ne, N2, Xe
Matheson 3102A              580                    Ar, He, Kr, Ne, N2, Xe
Matheson 3102               350                    CO, D2, H2, CH4
Matheson 3102               320                    CO2
Matheson 3500               326                    N2O
Matheson 3503               540                    O2
Matheson 3320               170                    Many Matheson lecture bottles
Matheson 1PA                510                    C2H2
Matheson 3332               110                    Many Matheson lecture bottles
Matheson 3501               660                    NH3, AsH3, GeH4

All regulators are equipped with Cajon 1/4" VCO male connectors on the output side. (See Section 4.5.4 for other outlet options.)

4.6.3 Gas Handling Station

Figure 4.3 is a schematic representation of the gas-mixing portion of the vacuum and gas-handling station. Stainless steel tubing, valves and connectors are used throughout. The valves are described in Table 4.1 (next page).

Figure 4.3

Inlet connections are 1/4" Cajon VCO male connectors. The outlet can be 1/4" Cajon VCO for high pressure applications (> 1 atm.) or 1/4" Cajon Ultra-Torr for low pressure (< 1 atm.). The evacuation port is a 1/2" Cajon Ultra-Torr connector, compatible with any of the vacuum pumps described in Section 4.5.

4.6.4 Gas Carrying and Connection Options

The primary gas-handling connection in the FTS lab is the 1/4" Cajon VCO connector. The outlet connection on all regulators listed in Section 4.6.2 is a male VCO connector.

Heavy flexible stainless steel hose is available (one 6' length and two 3' lengths) with female VCO connections on both ends. These hoses are effective from vacuum to 2500 psi.

For pressures below one atmosphere, four 24" lengths of 1/4" Cajon stainless steel flex-line are also available.

Adapters are available to convert any VCO (male or female) to 1/4" Swagelock male or female connectors or to Cajon 1/4" Ultra-Torr.

In order to conveniently utilize the FTS lab's regulators, gas-handling station, hoses, etc., the gas inlet port of a visitor-supplied source or absorption cell should be configured in one of the following ways:

  1. 1/4" Cajon VCO male connector
  2. 1/4" Swagelock male or female connector
  3. *1/4" O.D. tubing projecting at least 1" from application
  4. *1/4" Cajon Ultra-Torr connector
  5. 1/4" Female pipe thread

*Options (3) and (4) are not appropriate for positive pressure application.


4.7.1 Thermocouple Gauge

There is a Veeco DV-1M thermocouple vacuum gauge tube available which can be coupled into a visitor supplied system having a 1/8" female pipe thread connection. A Veeco TG-70 three channel thermocouple gauge control is available for readout. This gauge is primarily useful for leak-checking the system, and the vacuum pumps described in Section 4.5 are already equipped with thermocouples for this purpose.

4.7.2 Baratron Gauges

Three Baratron model 220B capicitance manometers are available with ranges 0-1, 0-10, 0-100, and 0-1000 Torr. The manufacturer specifies accuracies of 0.15% of reading 1 zero/span. The heads can be used with any gas compatible with 316 stainless steel and inconel.

The unit in use is configured with two valved input ports, each with 1/2" Cajon Ultra-Torr connectors. The second port can be used to pump down the isolated head to check the gauge zero.

In order to use these gauges with a visitor-supplied system, the gauge port should be one of the following:

  1. 1/2" O.D. section of tubing projecting at least 1"
  2. 1/2" Cajon Ultra-Torr connector
  3. 1/2" female pipe thread


4.8.1 Visitor-Supplied Absorption Cells

Visitors bringing their own absorption cells should note carefully the three preceding sections of this manual which detail the interfacing requirements for vacuum pumps, gas handling, and pressure-monitoring equipment. The cell windows should be wedged at least 5 milliradians to suppress channeling.

4.8.2 Absorption Cells at NSO

Two cells having lengths 1.5 meters and 25 cm. are available for visitor use. These cells are on loan from JPL and do not include windows. Visitors wishing to use these cells should supply windows to the following specifications:

              diameter:     2"
              thickness:    0.25"
              wedge:        5 milliradians

	      material:     Fused silica for UV    
		            CaF2 for greater than 1250 cm-1
		            BaF2 for greater than  800 cm-1
		            KCl  for greater than  600 cm-1

We suggest the following vendor primarily because of its rapid response:

              Janos Optical Corporation
              Route 35
              Townshend, Vermont 05353
             (802) 365-7714


Although the 12 meters of path within the interferometer are evacuated, the external path from the source to the FTS tank must be purged with dry nitrogen to reduce atmospheric absorption. For most lab applications, this external path will be on the order of 2 meters.

The source of dry nitrogen used for this purpose is either bottled dry N2 gas or blow-off from a 50 liter liquid N2 dewar containing a 100 watt heater.

If purging is needed, please inform the FTS staff during pre-run communications so that the necessary supplies can be obtained


4.10.1	Harrison 6268 A regulated

			0-40	Volts
			0-30 	Amp

4.10.2	Harrison 6448 B regulated

			0-600	Volts
			0-1.5  	Amp

4.10.3	Current regulated
			0-40 	Volts
			0-40  	Amp

4.10.4	Current regulated

			0-120	Volts
			0-5        Amp

4.10.5	EMI SCR 600-3 regulated

			0-600 Volts
			0-3        Amp

4.10.6	Ballast resistors

			0-5500 	Ohms	0.30 	Amps
			0-1300 	Ohms	0.57 	Amps
			0-680 	Ohms	0.8 	Amps
			0-340 	Ohms	1.12 	Amps
			0-170 	Ohms	1.6 	Amps
			0-85 	Ohms	2.2 	Amps
			0-40 	Ohms	3.2 	Amps

4.11 A.C. POWER

The standard FTS lab circuits are wired for 120 Volt, 20 amp, 60 Hz service.

In addition, the following configurations are available:

			240 V	3 phase	20	Amps
			240 V	single phase	60	Amps
			208 V	single phase	20	Amps
			480 V	3 phase	60	Amps

Not all of these options are available simultaneously and some lead time may be
required to do minor wiring changes in advance of the observing run.


Raytheon PGM10 150-watt microwave generator.


4.13.1 2 Liters/Min

Carried in 1/4 I.D. tygon tubing.

4.13.2 0-4 Gal/Min, 0-40 psi

Carried in garden hose with garden hose connections for supply and return. This is a recirculating system, and during the winter months it contains a mixture of water and antifreeze.