Tài liệu The Proposer''''s Guide for the Green Bank Telescope: GBT Support Sta ff pdf

The Proposer’s Guide for the Green Bank
Telescope
GBT Support Staff
December 19, 2012
This guide provides essential information for the preparation of observing
proposals on the Green Bank Telescope (GBT). The information covers
the facilities that will be offered in Semester 13B.
i
ii
Important News for Proposers
Deadline Proposals must be received by 5:00 P.M. EST (22:00 UTC) on Friday, 1 February 2012.
Technical Justification is Required All GBT proposals must include a Technical Justifica-
tion section (see Section 8.2)). Any proposal that does not include a technical justification may
be rejected without consideration.
VErsitile GBT Astronomical Spectrometer (VEGAS) We will accept shared-risk ob-
servations using the new VErsitile GBT Astronomical Spectrometer (VEGAS) which is an FPGA
based backend (see Section 3.3.2)).
PF1/450 Feed RFI Digital TV signals at frequencies above 470 MHz will make observing very
difficult with this receiver. Available RFI plots do not show the strength of these signals very well
as they overpower the system. Observers should consult the support scientists before submitting
a proposal for this feed.
PF1/600 Feed RFI Digital TV signals at frequencies covering most of this feed will make observ-
ing very difficult with this receiver. Available RFI plots do not show the strength of these signals
very well as they overpower the system. Observers should consult the support scientists before
submitting a proposal for this feed.
C-band Receiver The C-band receiver will be upgraded to include the 6-8 GHz frequency range.
We will consider shared-risk proposals for the 1 February 2013 deadline for observations in the 6-8
GHz range.
Ku-wideband Receiver The Ku-wideband receiver has nominal frequency range to cover 12.0
- 18.0 GHz. We will consider shared-risk proposals for this new feed (Ku-wideband) at the 1
February 2013 proposal deadline. When proposing, please use the nominal system temperature for

the ”old” Ku receiver. Please note that this feed was built for continuum and pulsar observations
and is expected to have very poor baseline structures for spectral lines. The feed does not have a
noise diode so close attention must be paid to calibration.
Pulsar Proposals All proposals requesting pulsar observations should use the GBT Sensitivity
Calculator available at ui.html to estimate
their observing times.
Sensitivity Calculator New All proposers should use the new and improved GBT Sensitivity Cal-
culator. Please see the GBT Sensitivity Calculator available at />ui/war/Calculator ui.html for further instructions. The new Sensitivity Calculator results can be
cut and pasted into the Technical Justification section of the proposal. This will streamline the
creation of your Technical Justification and will increase your chances of getting a positive technical
review.
The Dynamic Scheduling System (DSS) The GBT will be scheduled by the DSS during
the 13B semester. Further information on the GBT DSS can be found at: />Large Proposals Large Proposals (more than 200 hours) will be accepted for the 13B semester.
Large proposals will be accepted for the fully commissioned hardware only.
New Ph.D. Support Policy Proposer’s are reminded of the NRAO policy related to the sup-
port of Ph.D. dissertations using NRAO facilities. The policy can be found at
/>iii
Contents
1 Introduction to the GBT 1
2 Submitting a proposal 2
2.1 Latest Call for Proposals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.2 Joint Proposals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.3 Travel Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.4 Student Financial Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.5 Observing Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.6 Page Charge Support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
3 GBT Instruments 4
3.1 Antenna . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3.1.1 Resolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3.1.2 Surface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

3.1.3 Efficiency and Gain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3.2 Receivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3.2.1 Prime Focus Receivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3.2.2 Gregorian Receivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3.2.3 Receiver Resonances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.3 Backends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.3.1 GBT Spectrometer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.3.2 VErsitile GBT Astronomical Spectrometer . . . . . . . . . . . . . . . . . . . . . . 17
3.3.3 Spectral Processor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.3.4 DCR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.3.5 Guppi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.3.6 CCB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.3.7 Mark5 VLBA Disk Recorder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.3.8 User Provided Backends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4 GBT Observing Modes 21
4.1 Utility modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.2 Standard Observing Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.3 Switching Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.4 Spectral Line Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.4.1 Sensitivity and Integration Times . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.5 Continuum Modes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4.6 Polarization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.7 VLBI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
iv
5 Defining Sessions 26
6 Estimating Overhead Time 27
7 RFI 27
8 Tips for Writing Your Proposal 28
8.1 Items To Consider . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
8.2 Advice For Writing Your Technical Justification . . . . . . . . . . . . . . . . . . . . . . . . 28

8.3 Common Errors in GBT Proposals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
9 Further information 30
9.1 Additional Documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
9.2 Collaborations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
9.3 Contact People . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
A Appendix 31
A.1 GBT Sensitivity to Extragalactic 21 cm HI . . . . . . . . . . . . . . . . . . . . . . . . . . 31
A.2 Useful Web Links . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
List of Figures
1 HA, Dec and Horizon Plot for the GBT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2 Predicted aperture efficiencies for the GBT. . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3 Expected Tsys for the GBT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4 GBT SEFDs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
List of Tables
1 GBT Telescope Specifications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 GBT Receiver resonances. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3 GBT Receivers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
4 Commonly configured GBT Spectrometer Wide Bandwidth, Low Resolution Modes. . . . 14
5 Commonly configured GBT Spectrometer 50 MHz Bandwidth, High Resolution Modes. . 15
6 Commonly configured GBT Spectrometer 12.5 MHz Bandwidth, High Resolution Modes. 16
7 VEGAS Large Bandwidth, Few Spectral Window Modes. . . . . . . . . . . . . . . . . . . 17
8 VEGAS Small Bandwidth, Few Spectral Window Modes. . . . . . . . . . . . . . . . . . . 18
9 VEGAS Small Bandwidth, Many Spectral Window Modes. . . . . . . . . . . . . . . . . . 18
10 Spectral Processor Specifications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
v
11 GBT Spectral Processor Modes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
12 Allowed bandwidths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
13 K
1
values. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

14 GBT Contacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
15 Useful Web Sites for Proposal Writers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
1
1 Introduction to the GBT
Location Green Bank, West Virginia, USA
Coordinates Longitude: 79
◦
50

23.406

West (NAD83)
Latitude: 38
◦
25

59.236

North (NAD83)
Track Elevation: 807.43 m (NAVD88)
Optics 110 m x 100 m unblocked section of a 208 m parent paraboloid
Offaxis feed arm
Telescope Diameter 100 m (effective)
Available Foci Prime and Gregorian
f/D (prime) = 0.29 (referred to 208 m parent parabola)
f/D (prime) = 0.6 (referred to 100 m effective parabola)
f/D (Gregorian) = 1.9 (referred to 100 m effective aperture)
Receiver mounts Prime: Retractable boom with
Focus-Rotation Mount
Gregorian: Rotating turret with

8 receiver bays
Subreflector 8-m reflector with Stewart Platform (6 degrees of freedom)
Main reflector 2004 actuated panels (2209 actuators)
Average intra-panel RMS 68 µm
FWHM Beamwidth Gregorian Feed: ∼ 12.60/f
GHz
arcmin
Prime Focus: ∼ 13.01/f
GHz
arcmin (see Section 3.1.1)
Elevation Limits Lower limit: 5 degrees
Upper limit: ∼ 90 degrees
Declination Range Lower limit: ∼ −46 degrees
Upper limit: 90 degrees
Slew Rates Azimuth: 35.2 degrees/min
Elevation: 17.6 degrees/min
Surface RMS Passive surface: 450 µm at 45
◦
elevation, worse elsewhere
Active surface: ∼ 250 µm, under benign night-time conditions
Pointing accuracy 1σ values from 2-D data
5

blind
2.7

offset
Table 1: GBT Telescope Specifications.
The Green Bank Telescope is a 100-m diameter single dish radio telescope. The telescope has several
advanced design characteristics that, together with its large aperture, make it unique:

• Fully-steerable antenna 5–90 degrees elevation range and 85% coverage of the celestial sphere
1
• Unblocked aperture reduces sidelobes, Radio Frequency Interference (RFI), and spectral standing
waves
• Active surface allows for compensation for gravity and thermal distortions, and includes near real-
time adjustments to optics and pointing.
• Frequency coverage of 290 MHz to 100 GHz provides nearly 3 decades of frequency coverage
for maximum scientific flexibility
1
Because the GBT is an alt-az mounted telescope it cannot track sources that are near the zenith.
2
• Location in the National Radio Quiet Zone ensures a comparatively low RFI environment
The GBT is operated by the National Radio Astronomy Observatory, a facility of the National
Science Foundation operated under cooperative agreement by Associated Universities Incorporated. The
GBT is intended to address a very broad range of astronomical problems at radio wavelengths, and is
available to qualified observers on a peer-reviewed proposal basis. It is run primarily as a facility
for visiting observers, and the NRAO provides extensive support services including round-the-clock
operators.
Technical specifications for the telescope are given in Table 1.
Source rising and setting times can be estimated using Figure 1.
Figure 1: Plot of elevation vs azimuth, with lines of constant Hour Angle (HA; cyan lines) and Declination
(DEC; brown lines) for the GBT. The horizon (magenta line) is shown at 5 degrees elevation, except for
the mountains in the west and the 140–foot (43-m) telescope at azimuth = 48
◦
. The lines of constant
DEC are shown in increments of ± 10
◦
, while the lines of constant HA are in increments of ± 1 hour.
2 Submitting a proposal
General proposal information is available at The NRAO proposal

submission tool ( should be used to submit all GBT proposals.
3
2.1 Latest Call for Proposals
The latest call for proposals can be found at />2.2 Joint Proposals
If you are submitting a joint proposal, you must explicitly state this in your proposal abstract. Proposals
requiring GBT participation in VLBA or global VLBI observations should be submitted to the VLBA
only, not to the GBT. Proposals for joint GBT and VLA observations must be submitted for each
instrument separately.
If you are planning to use the GBT as part of a co-ordinated program with other observatories, you
should follow these links:
For FERMI joint proposals see .
For CHANDRA joint proposals see />For SPITZER joint proposals see .
2.3 Travel Support
Some travel support for observing and data reduction is available for U.S. investigators on successful
proposals. Information can be found at
office/nonemployee observing travel.shtml.
2.4 Student Financial Support
Financial support for graduate and undergraduate students performing research with any NRAO tele-
scope is available through the Student Support Program. Awards of up to $35,000 are possible. Informa-
tion about the program can be found at />Your application for Student Financial Support should be included as part of your NRAO observing
proposal.
2.5 Observing Policies
The policy for observing with the GBT, including a description of the restrictions concerning remote
observing, can be found at />2.6 Page Charge Support
NRAO provides page charge support for U.S. authors for any paper that presents original data obtained
with any NRAO telescope. See charges.shtml for more details.
4
3 GBT Instruments
3.1 Antenna
3.1.1 Resolution

The resolution of the GBT is given by
FWHM = (1.02 + 0.0135 ∗Te(Db))
λ
100 m
rad (1)
where FWHM is the Full-Width at Half-Maximum of the symmetric, two-dimensional Gaussian shaped
beam and Te(Db) is the edge taper of the feed’s illumination of the dish in decibels. The edge taper
varies with frequency and polarization for all of the GBT feeds. For the Gregorian feed the edge taper
is typically 14 ± 2 Db which results in
F W HM
>1GHz
=
12.46 → 12.73

f
GHz
=
747.6 → 763.8

f
GHz
(2)
For the prime focus receivers the edge taper is typically 18 ± 2 Db which results in
F W HM
<1GHz
=
12.73 → 13.29

f
GHz

=
763.8 → 797.4

f
GHz
(3)
3.1.2 Surface
The GBT surface consists of 2004 panels mounted on 2209 computer-controlled actuators. Below 4 GHz,
use of the active surface makes a negligible change to the telescope efficiency, and it is disabled to avoid
unnecessary wear on the actuators.
Above 4 GHz, the active surface is automatically adjusted to compensate for residual non-homologous
deformations as the gravity vector changes with changing elevation. The corrections are a combination
of predictions from a Finite Element Model (FEM) of the GBT structure plus additional empirical
corrections derived from Out-of-focus (OOF) holography measurements. The OOF measurements are
parametrized as low-order Zernike polynomials. The FEM plus OOF corrections are automatically
calculated for the elevation of the mid-point of a scan, and are applied prior to the start of the scan.
3.1.3 Efficiency and Gain
A graph of the anticipated and measured aperture efficiencies for the GBT appears in Figure 2.
The proposer should also read the memo
/>by Ron Maddalena for more details on the characteristics and performance of the GBT.
3.2 Receivers
GBT receivers cover frequency bands from 0.290-49.8 GHz and 80-100 GHz. Table 3 summarizes the
receivers and their properties (nominal frequency ranges, efficiencies, etc.). If you would like to know
about any receiver’s performance outside of the nominal frequency ranges you should contact one of the
GBT Observational Support Scientists (see Table 14).
5
Figure 2: Predicted aperture efficiencies for the GBT. Values below 5 GHz are based on a surface RMS
of 450 µm and 300 µm for frequencies above 5 GHz. The beam efficiencies are 1.37 times the aperture
efficiency.
3.2.1 Prime Focus Receivers

The prime focus receiver is mounted in a focus-rotation mount (FRM) on a retractable boom. The boom
is moved to the prime focus position when prime focus receiver is in use, and retracted when Gregorian
receivers are required. The FRM has three degrees of freedom: Z-axis radial focus, Y-axis translation
(in the direction of the dish plane of symmetry), and rotation. It can be extended or retracted at any
elevation. This usually takes about 10 minutes.
As the FRM holds one receiver box at a time, a change from PF1 to PF2 receivers requires a
box exchange. Additionally, changing frequency bands within PF1 requires a change in the PF1 feed.
Changes of or in prime focus receivers are usually made during routine maintenance time preceding a
dedicated campaign using that receiver.
Prime Focus 1 (PF1)
The PF1 receiver is divided into 4 frequency bands within the same receiver box. The frequency
ranges are (see Table 3) 290 - 395 MHz, 385 - 520 MHz, 510 - 690 MHz and 680 - 920 MHz. Each
frequency band requires its specific feed to be attached to the receiver before that band can be used.
The receivers are cooled FET amplifiers. The feeds for the first three bands are short-backfire dipoles.
The feed for the fourth is a corrugated feed horn with an Orthomode transducer (OMT) polarization
splitter.
A feed change is required to move between bands. This takes 2-4 hours, and is done during routine
maintenance days (see above).
The user can select one of four IF filters in the PF1 receiver. These have bandwidths of 20, 40, 80
and 240 MHz.
6
Prime Focus 2 (PF2) (0.910 - 1.23 GHz)
PF2 uses a cooled FET and a corrugated feed horn with an OMT. The user can select one of four
IF filters in the PF2 receiver. These have bandwidths of 20, 40, 80 and 240 MHz.
3.2.2 Gregorian Receivers
The receiver room located at the Gregorian Focus contains a rotating turret in which the Gregorian
receivers are mounted. There are 8 portals for receiver boxes in the turret. All 8 receivers can be kept
cold and active at all times.
More information on individual Gregorian receivers follows, which includes design types and internal
switching modes, i.e., those switching modes activated inside the receiver (e.g., frequency, beam, or

polarization). External switching such as antenna position switching is always available. The Gregorian
subreflector can be used for slow chopping.
The Gregorian receivers all have the following components unless explicitly stated below. Each of the
Gregorian receivers is a cooled HFET amplifier and every feed/beam has a corrugated horn wave-guide.
All calibration for the Gregorian receivers are done via injection of a signal from a noise diode.
For information on how T
sys
varies with weather conditions and telescope elevation, please see
/>L-Band (1.15 – 1.73 GHz)
This receiver has one beam on the sky, with dual polarizations. The feed has a cooled OMT producing
linear polarizations. The user can select circular polarization which is synthesized using a hybrid (after
the first amplifiers) in the front-end. Allowed internal switching modes are frequency and/or polarization
switching. The user can select one of four RF filters: 1.1-1.8 GHz, 1.1-1.45 GHz, 1.3-1.45 GHz, 1.6-1.75
GHz. A notch filter between 1.2 and 1.34 GHz is available to suppress interference from a nearby Air
Surveillance Radar. There is choice between two noise diodes with different levels (∼ 10% or ∼ 100% of
the system temperature) for flux calibration.
S-Band (1.73 – 2.60 GHz)
This receiver has one beam with dual polarizations. The feed has a cooled OMT producing linear
polarizations. The user can select circular polarization synthesized using a hybrid (after the first ampli-
fiers) in the front-end. Internal switching modes include frequency switching. The user can select one
of two RF filters: 1.68-2.65 GHz, 2.1-2.4 GHz. There is choice between two noise diodes with different
levels (∼ 10% or ∼ 100% of the system temperature) for flux calibration.
A superconducting notch filter is permanently installed immediately after the first amplifiers and
covers the range 2300–2360 MHz. This filter suppresses interference from the Sirius and XMM satellite
radio transmissions.
C-Band (3.95 – 6.1 GHz)
This receiver has one beam, with dual polarizations. The feed has a cooled OMT producing linear
polarizations. The user can select circular polarization synthesized using a hybrid (after the first am-
plifiers) in the front-end. The allowed internal switching mode is frequency switching. There is choice
between two noise diodes with different levels (∼ 10% or ∼ 100% of the system temperature) for flux

calibration.
If funds and resources become available the frequency range of this receiver may be extended up to
8 GHz.
X-Band (8.0 – 10.0 GHz)
1
1
The frequency range of the X-band receiver has been extended up to 11.6 GHz. However, users are cautioned that
above 10 GHz, the polarization purity degrades, and the low level noise-diode strength drops off.
7
This receiver has one beam, with dual circular polarizations. The feed has a cooled polarizer pro-
ducing circular polarizations. The internal switching modes are frequency switching and polarization
switching. The user can select IF Bandwidths of 500 or 2400 MHz. There is a single noise diode (∼ 10%
of the system temperature) for flux calibration.
Ku-Band (12.0 – 15.4 GHz)
This receiver has two beams on the sky with fixed separation, each with dual circular polarization.
The feeds have cooled polarizers producing circular polarizations. Internal switching modes are frequency
and/or IF switching (the switch is after the first amplifiers). The user can select IF Bandwidths of 500
or 3500 MHz. The two Ku-band feeds are separated by 330

in the cross-elevation direction. There is a
noise diode for each beam (∼ 10% of the system temperature) for flux calibration.
Ku-Wideband (11.0 – 18.0 GHz)
This receiver has one beam on the sky with dual linear polarization. The receiver will cover 11 to
18 GHz simultaneously in both polarizations for pulsar observations. Spectral line observations will also
be possible with this receiver but the observer should be aware that spectral baselines are not expected
to be very good.
K-Band Focal Plane Array (18.0 – 27.5 GHz)
The K-band Focal Plane Array has seven beams total, each with dual circular polarization. Each
beam covers the 18-27.5 GHz frequency range with fixed separations on the sky. The feeds have cooled
polarizers producing circular polarization. The only internal switching modes is frequency switching.

The seven feeds are laid out in a hexagon with one central feed. The hexagon is oriented such that the
central feed is not at the same cross-elevation or the same elevation as any of the other beams. There
is a noise diode for each beam (∼ 10% of the system temperature) for flux calibration. The maximum
instantaneous bandwidth for the receiver is currently 1.8 GHz.
Ka-Band (26.0– 39.5 GHz)
This receiver has two beams, each with a single linear polarization. The polarizations of the two
beams are orthogonal and are aligned at ±45
◦
angles to the elevation (and cross-elevation) direction.
The receiver is built according to a pseudo-correlation design intended to minimize the effect of 1/f gain
fluctuations for continuum and broadband spectral line observation. 180
◦
waveguide hybrids precede
and follow the low noise amplifiers. Phase switches between the amplifiers and the second hybrid allow
true beam switching to be used with this receiver.
The Zpectrometer and CCB use the full 26–40 GHz range of the Ka-band receiver. For other
backends, the receiver is broken into three separate bands: 26.0-31.0 GHz, 30.5-37.0 GHz, and 36.0-
39.50 GHz. You can only use one of these bands at a time, except for the CCB and Zpectrometer
backends which can use the full frequency range of the receiver. For backends other than the CCB and
Zpectrometer, the maximum instantaneous bandwidth achievable with this receiver is limited to 4 GHz.
There is a noise diode for each beam (∼ 10% of the system temperature) for flux calibration. The feeds
are separated by 78

in the cross-elevation direction.
Q-Band (38.2–49.8 GHz)
This receiver has two beams with fixed separation, each dual circular polarization. The feeds have
cooled polarizers producing circular polarizations. The internal switching mode available is frequency
switching. The IF Bandwidth is 4000 MHz. Calibration is by noise injection and/or ambient load. The
feeds are separated by 57.8


in the cross-elevation direction.
W-band 4mm (67–93.3 GHz)
A W-band 4mm two pixel receiver is currently under development. The receiver will cover the fre-
quency range 67–93.3 GHz. Please see 12b.shtml for more details.
Note that the receiver’s instantaneous bandwidth is limited to 1280 MHz. An ongoing update to
the receiver may change this limitation. Please see 12b.shtml for
the latest bandwidth limitation information.
8
The IF system for the 4mm system is broken into four separate bands: 67-74 GHz, 73-80 GHz, 79-86
GHz, and 85-93.3 GHz, and you can only use one of these bands at a time.
Mustang (80–100 GHz Bolometer Array)
Mustang, built by a collaboration that includes the University of Pennsylvania, NRAO, GSFC,
NIST, and Cardiff University, is a 64 pixel bolometer array which operates with a 20 GHz bandwidth
centered at 90 GHz. As of March 2009, the demonstrated sensitivity of MUSTANG on the GBT yields
a 0.4 mJy RMS in one hour of integration time mapping a 3

x3

region. The noise scales as the square
root of the integration time, and with the square root of the area covered. Mapping smaller areas is
not efficient in terms of noise performance. For significantly larger areas, faster scanning will reduce the
noise by up to ∼ 35%. For photometry of compact (D = 1

or less) objects, center-weighted “daisy”
mapping scans may be used which further reduce the RMS by a factor of two in the central region.
Finally, smoothing will reduce the map RMS by a factor of ∼ (FWHM/4

), where F W HM is the
full-width at half max of the smoothing kernel (the default gridding and pixel size parameters provide
an effective 4


smoothing of the map). Proposals must explicitly state a target map RMS in order to be
evaluated for scheduling.
Extended emission on scales of 30

to a few arcminutes can be imaged with reasonable fidelity, but
faint emission more extended than this may be difficult to detect. Bright emission (> 20mJy/beam) is
easily reconstructed over scales of many arcminutes. The angular resolution of Mustang on the GBT is
typically 9

(FWHM) and the instantaneous field of view is 40

x40

.
Allowing for weather, calibration and observing overheads, the typical observing efficiency realized
on the telescope is ∼ 50%. Daytime observing at 90 GHz is currently not advised; only observations
collected between 3 hours after sunset, and sunrise, are consistently useful. The Mustang receiver
noise will increase below 30
◦
elevation due to increased vibrations and below 19
◦
elevation the receiver
cryogenics will no longer function. Elevation constraints should be noted in the “constraints” section
within the Proposal Submission Tool.
For further information please refer to
or contact Brian Mason at NRAO ()
3.2.3 Receiver Resonances
The GBT receivers are known to have resonances within their respected band-passes. These are fre-
quencies where the receiver response is non-linear. The resonances arise in the ortho-mode transducers

(OMTs) which separate the two polarizations of the incoming signal. Although valid data can be ob-
tained within the receiver resonances, the observer should be aware that this might not always be the
case. As a general rule, polarization observations will be affected much more strongly than total intensity
observations in the regions of the resonances.
The receiver resonances have been measured in the lab and are listed in Table 2. However, these
data should not be taken as complete as there may be resonances that could not be detected in the lab
due to sensitivity limits. The center frequencies of the resonances are determined with an accuracy of
only a few MHz at best. The widths of the resonances are typically less than 5 MHz.
9
Receiver Frequency FWHM
MHz MHz
PF1 796.6 2.09
PF1 817.4 3.29
PF2 925.9 0.17
PF2 1056.0 –
PF2 1169.9 3.28
L 1263.0 0.60
L 1447.0 0.68
L 1607.0 0.90
L 1720.0 –
S 1844.0 2.00
S 2118.0 0.96
S 2315.0 –
S 2561.0 0.91
C 4163.0 4.7
C 4747.0 28.6
C 5150.0 –
C 5248.0 4.3
C 5680.0 5.0
X 9742.0 6.7

X 10504.0 118.0
X 11415.0 46.4
Ku 12875.0 8.1
Ku 12885.0 7.1
Table 2: GBT Receivers resonances. This list is not necessarily complete. The FWHMs are from
Gaussian fits. Typical resonances have wings that are broader than Gaussian profiles. Resonances with
FWHM listed as “–” were not seen in the astronomical data.
10
Receiver Band Frequency Focus Polarization Beams Polarizations Beam FWHM Gain Aperture Maximum Instantaneous
Range per Separation (K/Jy) Efficiency Bandwidth
(GHz) Beam (MHz)
PF1 342 MHz .290 395 Prime Lin/Circ 1 2 —— 36

2.0 70% 240
450 MHz .385 520 Prime Lin/Circ 1 2 —— 27

2.0 70%
600 MHz .510 690 Prime Lin/Circ 1 2 —— 21

2.0 70%
800 MHz .680 920 Prime Lin/Circ 1 2 —— 15

2.0 70%
PF2 —— .910-1.23 Prime Lin/Circ 1 2 —— 12

2.0 70% 240
L-Band —— 1.15-1.73 Greg. Lin/Circ 1 2 —— 9

2.0 70% 650
S-Band —— 1.73-2.60 Greg. Lin/Circ 1 2 —— 5.8


1.9 70% 970
C-Band —— 3.95-6.1 Greg. Lin/Circ 1 2 —— 2.5

1.85 70% 2000
X-Band —— 8.00-10.0 Greg. Circ 1 2 —— 1.4

1.8 70% 2400
Ku-Band —— 12.0-15.4 Greg. Circ 2 2 330

54

1.7 70% 3500
Ku-wideand —— 11.0-18.0 Greg. Linear 1 2 54

1.7 70% 7000
KFPA —— 18.0-27.5 Greg. Circ 7 2 96

32

1.5 67% 1800
Ka-Band MM-F1 26.0-31.0 Greg. Circ 2 1 78

26.8

1.5 56-64% 4000
MM-F2 30.5-37.0 22.6

MM-F3 36.0-39.5 19.5


Q-Band —– 38.2-49.8 Greg. Circ 2 2 58

16

1.0 47-56% 4000
W-Band 4mm MM-F1 67-74 Greg. Circ 2 2 TBD 10

TBD 35% 1280
MM-F2 73-80 Greg. Circ 2 2
MM-F3 79-86 Greg. Circ 2 2
MM-F4 85-93.3 Greg. Circ 2 2
Mustang —– 80-100 Greg. —- 64 —- —- 10

—- 35% 20000
Table 3: GBT Receivers’ parameters. Beam efficiency is 1.37 times the aperture efficiency. See Figure 3 for information on how T
sys
varies with
frequency. See Figure 4 for information on how the System Equivalent Flux Density (SEFD) varies with frequency. Note that for lower frequencies
the observer will have to add an estimate for the Galactic background emission, T
bg
to the system temperatures in order to get realistic values for
sensitivity and noise limits.
11
Figure 3: Expected Tsys the GBT for typical weather conditions.
12
Figure 4: System Equivalent Flux densities the GBT for typical weather conditions.
13
3.3 Backends
3.3.1 GBT Spectrometer
The GBT Spectrometer provides the observer with a remarkable variety of spectral line observing modes,

intended to optimize the scientific return on experiments. The Spectrometer is a modular system, with
four quadrants. Quadrants may be used independently or grouped together into banks of 1, 2 or
4 quadrants. This provides the observer with 1 to 3 different levels of spectral resolutions for each
observing mode, as described below. When the 4 quadrants are independently operated, they can be
configured to acquire data at up to 8 different frequencies.
The spectrometer performs auto correlations of the input signals. The input signals may be a) both
polarizations in a spectral window (i.e the selected bandwidth centered on a specified spectral line), b)
both polarization inputs from different feeds of multi-feed receivers, or c) combinations of the preceding
in different spectral windows.
The spectrometer modes are divided into two major types, wide bandwidth, low resolution and
narrow bandwidth, high resolution.
The spectrometer has a dynamic range of about ±2.5 dB from its optimal balance point
2
Off/On
observations of bright continuum sources may be affected by this.
The GBT IF system limits the number of spectral windows available to a maximum of eight. Our
observing software assumes that polarization pairs (i.e. both polarizations) will be routed to the spec-
trometer. Consequently, it is best to write your proposal assuming that you will use both polarizations.
Using only one polarization for an observation requires the observer to be a GBT expert, and requires
very long setup times.
Wide Bandwidth, Low Resolution
The spectrometer can be configured to produce 1, 2, or 4 spectra simultaneously, each with up to
800 MHz bandwidth. Hence the maximum total spectral coverage is 3200 MHz (4 spectral windows each
with 800 MHz bandwidth and no overlap) using both polarizations. The maximum spectral resolution
is dependent on the number of spectral windows and the number of quadrants used. For a given number
of spectral windows up to three different spectral resolutions are possible.
A sub-mode of the wideband operation is the 200 MHz bandwidth option, which provides increased
spectral resolution.
Table 4 shows the possible spectral resolutions with both the 800 MHz and 200 MHz bandwidths.
Narrow Bandwidth, High Resolution

The narrow bandwidth, high resolution spectrometer mode can produce bandwidths of either 12.5
or 50 MHz bandwidth for 1, 2, 4, or 8 spectral windows. The maximum spectral resolution is dependent
on the number of spectral windows and the number of quadrants used.
The narrow bandwidth operation supports 3 and 9 levels of analog to digital (A/D) sampler resolu-
tion. The 3-level mode allows greater spectral resolution and the 9-level mode reduces the sensitivity to
RFI and also slightly increases the sensitivity to radio astronomical sources. The 3-level option provides
a factor of 4 better spectral resolution compared with the 9-level operation. The 3-level mode is ∼ 83.5%
as sensitive to radio astronomical sources as the 9-level mode as shown by the K
1
values in Table 13.
Tables 5 and 6 show the possible spectral resolutions with both the 50 MHz and 12.5 MHz band-
widths.
2
The optimal balance point of the spectrometer is determined by the input power level that provides the maximum
sensitivity of the output, quantized digital data stream when the input analog signal is digitally sampled. 1 dB is a measure
of the change in power levels and is defined as 1 dB = 10 (log(P
1
) − log(P
2
)) where P
1
and P
2
are two power levels being
compared.
14
Bandwidth Polarization Number of Number of Lags - Approximate Resolution
(MHz) Cross-Products Spectral Beams Low Medium High
Windows
800 No 1 1 2048 – 390.6250 kHz 4096 – 195.3125 kHz 8192 – 97.6563 kHz

800 No 2 1 2048 – 390.6250 kHz 4096 – 195.3125 kHz 4096 – 195.3125 kHz
800 No 1 2 2048 – 390.6250 kHz 4096 – 195.3125 kHz 4096 – 195.3125 kHz
800 No 4 1 2048 – 390.6250 kHz 2048 – 390.6250 kHz 2048 – 390.6250 kHz
800 No 2 2 2048 – 390.6250 kHz 2048 – 390.6250 kHz 2048 – 390.6250 kHz
800 Yes 1 1 1024 – 781.2500 kHz 2048 – 390.6250 kHz 4096 – 195.3125 kHz
800 Yes 2 1 1024 – 781.2500 kHz 2048 – 390.6250 kHz 2048 – 390.6250 kHz uncommissioned
800 Yes 1 2 1024 – 781.2500 kHz 2048 – 390.6250 kHz 2048 – 390.6250 kHz uncommissioned
800 Yes 4 1 1024 – 781.2500 kHz 1024 – 781.2500 kHz 1024 – 781.2500 kHz uncommissioned
800 Yes 2 2 1024 – 781.2500 kHz 1024 – 781.2500 kHz 1024 – 781.2500 kHz uncommissioned
200 No 1 1 8192 – 24.4141 kHz 16384 – 12.2070 kHz 32768 – 6.1035 kHz
200 No 2 1 4096 – 48.8281 kHz 8192 – 24.4141 kHz 16384 – 12.2070 kHz
200 No 1 2 4096 – 48.8281 kHz 8192 – 24.4141 kHz 16384 – 12.2070 kHz
200 No 4 1 8192 – 24.4141 kHz 8192 – 24.4141 kHz 8192 – 24.4141 kHz
200 No 2 2 8192 – 24.4141 kHz 8192 – 24.4141 kHz 8192 – 24.4141 kHz
200 Yes 1 1 4096 – 48.8281 kHz 8192 – 24.4141 kHz 16384 – 12.2070 kHz
200 Yes 2 1 2048 – 97.6563 kHz 4096 – 48.8281 kHz 8192 – 24.4141 kHz uncommissioned
200 Yes 1 2 2048 – 97.6563 kHz 4096 – 48.8281 kHz 8192 – 24.4141 kHz uncommissioned
200 Yes 4 1 4096 – 48.8281 kHz 4096 – 48.8281 kHz 4096 – 48.8281 kHz uncommissioned
200 Yes 2 2 4096 – 48.8281 kHz 4096 – 48.8281 kHz 4096 – 48.8281 kHz uncommissioned
Table 4: Commonly used configurations of the GBT Spectrometer in its Wide Bandwidth, Low Resolution Modes. All modes use 3 level sampling.
15
Bandwidth Polarization Level Number of Number of Lags - Approximate Resolution
(MHz) Cross-Products Sampling Spectral Beams Low Medium High
Windows
50 No 3 1 1 32768 - 1.5259 kHz 65536 - 0.7629 kHz 131072 - 0.3815 kHz
50 No 3 2 1 16384 - 3.0518 kHz 32768 - 1.5259 kHz 65536 - 0.7629 kHz
50 No 3 1 2 65536 - 0.7629 kHz 65536 - 0.7629 kHz 65536 - 0.7629 kHz
50 No 3 4 1 16384 - 3.0518 kHz 32768 - 1.5259 kHz 32768 - 1.5259 kHz
50 No 3 2 2 32768 - 1.5259 kHz 32768 - 1.5259 kHz 32768 - 1.5259 kHz
50 No 3 8 1 16384 - 3.0518 kHz 16384 - 3.0518 kHz 16384 - 3.0518 kHz Single Beam Receivers Only

50 No 3 4 2 16384 - 3.0518 kHz 16384 - 3.0518 kHz 16384 - 3.0518 kHz
50 Yes 3 1 1 16384 - 3.0518 kHz 32768 - 1.5259 kHz 65536 - 0.7629 kHz uncommissioned
50 Yes 3 2 1 8192 - 6.1035 kHz 16384 - 3.0518 kHz 32768 - 1.5259 kHz
50 Yes 3 1 2 32768 - 1.5259 kHz 32768 - 1.5259 kHz 32768 - 1.5259 kHz uncommissioned
50 Yes 3 4 1 8192 - 6.1035 kHz 16384 - 3.0518 kHz 16384 - 3.0518 kHz
50 Yes 3 2 2 16384 - 3.0518 kHz 16384 - 3.0518 kHz 16384 - 3.0518 kHz uncommissioned
50 Yes 3 8 1 8192 - 6.1035 kHz 8192 - 6.1035 kHz 8192 - 6.1035 kHz uncommissioned, Single Beam Receivers Only
50 Yes 3 4 2 8192 - 6.1035 kHz 8192 - 6.1035 kHz 8192 - 6.1035 kHz uncommissioned
50 No 9 1 1 8192 - 6.1035 kHz 16384 - 3.0518 kHz 32768 - 1.5259 kHz
50 No 9 2 1 4096 - 12.2070 kHz 8192 - 6.1035 kHz 16384 - 3.0518 kHz
50 No 9 1 2 16384 - 3.0518 kHz 16384 - 3.0518 kHz 16384 - 3.0518 kHz
50 No 9 4 1 4096 - 12.2070 kHz 8192 - 6.1035 kHz 8192 - 6.1035 kHz
50 No 9 2 2 8192 - 6.1035 kHz 8192 - 6.1035 kHz 8192 - 6.1035 kHz
50 No 9 8 1 4096 - 12.2070 kHz 4096 - 12.2070 kHz 4096 - 12.2070 kHz Single Beam Receivers Only
50 No 9 4 2 4096 - 12.2070 kHz 4096 - 12.2070 kHz 4096 - 12.2070 kHz
50 Yes 9 1 1 4096 - 12.2070 kHz 8192 - 6.1035 kHz 16384 - 3.0518 kHz uncommissioned
50 Yes 9 2 1 2048 - 24.4141 kHz 4096 - 12.2070 kHz 8192 - 6.1035 kHz
50 Yes 9 1 2 8192 - 6.1035 kHz 8192 - 6.1035 kHz 8192 - 6.1035 kHz uncommissioned
50 Yes 9 4 1 2048 - 24.4141 kHz 4096 - 12.2070 kHz 4096 - 12.2070 kHz uncommissioned
50 Yes 9 2 2 4096 - 12.2070 kHz 4096 - 12.2070 kHz 4096 - 12.2070 kHz uncommissioned
50 Yes 9 8 1 2048 - 24.4141 kHz 2048 - 24.4141 kHz 2048 - 24.4141 kHz uncommissioned, Single Beam Receivers Only
50 Yes 9 4 2 2048 - 24.4141 kHz 2048 - 24.4141 kHz 2048 - 24.4141 kHz uncommissioned
Table 5: Commonly configured GBT Spectrometer 50 MHz Bandwidth, High Resolution Modes.
16
Bandwidth Polarization Level Number of Number of Lags - Approximate Resolution
(MHz) Cross-Products Sampling Spectral Beams Low Medium High
Windows
12.5 No 3 1 1 32768 - 0.3815 kHz 65536 - 0.1907 kHz 131072 - 0.0954 kHz
12.5 No 3 2 1 16384 - 0.7629 kHz 32768 - 0.3815 kHz 65536 - 0.1907 kHz
12.5 No 3 1 2 65536 - 0.1907 kHz 65536 - 0.1907 kHz 65536 - 0.1907 kHz

12.5 No 3 4 1 16384 - 0.7629 kHz 32768 - 0.3815 kHz 32768 - 0.3815 kHz
12.5 No 3 2 2 32768 - 0.3815 kHz 32768 - 0.3815 kHz 32768 - 0.3815 kHz
12.5 No 3 8 1 16384 - 0.7629 kHz 16384 - 0.7629 kHz 16384 - 0.7629 kHz Single Beam Receivers Only
12.5 No 3 4 2 16384 - 0.7629 kHz 16384 - 0.7629 kHz 16384 - 0.7629 kHz
12.5 Yes 3 1 1 16384 - 0.7629 kHz 32768 - 0.3815 kHz 65536 - 0.1907 kHz uncommissioned
12.5 Yes 3 2 1 8192 - 1.5259 kHz 16384 - 0.7629 kHz 32768 - 0.3815 kHz
12.5 Yes 3 1 2 32768 - 0.3815 kHz 32768 - 0.3815 kHz 32768 - 0.3815 kHz uncommissioned
12.5 Yes 3 4 1 8192 - 1.5259 kHz 16384 - 0.7629 kHz 16384 - 0.7629 kHz
12.5 Yes 3 2 2 16384 - 0.7629 kHz 16384 - 0.7629 kHz 16384 - 0.7629 kHz uncommissioned
12.5 Yes 3 8 1 8192 - 1.5259 kHz 8192 - 1.5259 kHz 8192 - 1.5259 kHz uncommissioned, Single Beam Receivers Only
12.5 Yes 3 4 2 8192 - 1.5259 kHz 8192 - 1.5259 kHz 8192 - 1.5259 kHz uncommissioned
12.5 No 9 1 1 8192 - 1.5259 kHz 16384 - 0.7629 kHz 32768 - 0.3815 kHz
12.5 No 9 2 1 4096 - 3.0518 kHz 8192 - 1.5259 kHz 16384 - 0.7629 kHz
12.5 No 9 1 2 16384 - 0.7629 kHz 16384 - 0.7629 kHz 16384 - 0.7629 kHz
12.5 No 9 4 1 4096 - 3.0518 kHz 8192 - 1.5259 kHz 8192 - 1.5259 kHz
12.5 No 9 2 2 8192 - 1.5259 kHz 8192 - 1.5259 kHz 8192 - 1.5259 kHz
12.5 No 9 8 1 4096 - 3.0518 kHz 4096 - 3.0518 kHz 4096 - 3.0518 kHz Single Beam Receivers Only
12.5 No 9 4 2 4096 - 3.0518 kHz 4096 - 3.0518 kHz 4096 - 3.0518 kHz
12.5 Yes 9 1 1 4096 - 3.0518 kHz 8192 - 1.5259 kHz 16384 - 0.7629 kHz uncommissioned
12.5 Yes 9 2 1 2048 - 6.1035 kHz 4096 - 3.0518 kHz 8192 - 1.5259 kHz uncommissioned
12.5 Yes 9 1 2 8192 - 1.5259 kHz 8192 - 1.5259 kHz 8192 - 1.5259 kHz uncommissioned
12.5 Yes 9 4 1 2048 - 6.1035 kHz 4096 - 3.0518 kHz 4096 - 3.0518 kHz
12.5 Yes 9 2 2 4096 - 3.0518 kHz 4096 - 3.0518 kHz 4096 - 3.0518 kHz uncommissioned
12.5 Yes 9 8 1 2048 - 6.1035 kHz 2048 - 6.1035 kHz 2048 - 6.1035 kHz uncommissioned, Single Beam Receivers Only
12.5 Yes 9 4 2 2048 - 6.1035 kHz 2048 - 6.1035 kHz 2048 - 6.1035 kHz uncommissioned
Table 6: Commonly configured GBT Spectrometer 12.5 MHz Bandwidth, High Resolution Modes.
17
Bandwidth Number of Number of Channels - Approximate Resolution Mininum Integration Notes
(MHz) Spectral Windows Beams (kHz) Time (sec)
1500 1 or 2 1 1024 – 1464.844 0.5 1st priority mode

1500 1 2 1024 – 1464.844 0.5 1st priority mode
1000 1 or 2 1 2048 – 488.281 0.7
1000 1 2 2048 – 488.281 0.7
800 1 or 2 1 4096 – 195.313 1.3
800 1 2 4096 – 195.313 1.3
500 1 or 2 1 8192 – 61.035 2.5
500 1 2 8192 – 61.035 2.5
400 1 or 2 1 16384 – 24.414 5.0
400 1 2 16384 – 24.414 5.0
Table 7: VEGAS Large Bandwidth, Few Spectral Window Modes.
Minimum Integration Times with the Spectrometer
The spectrometer can nominally handle integration times as small as 1 − 2 seconds. The exact
nominal minimum integration time depends on the total number of spectral channels being recorded.
It is possible to reduce the integration below time 1−2 seconds; however, this requires special setups
and more time for the project. Such setups will be allowed only for well justified circumstances. Please
contact a support scientist if you desire shorter integration times than 1 −2 seconds.
3.3.2 VErsitile GBT Astronomical Spectrometer
A new FPGA based spectrometer is being build as a collaboration between NRAO and UC Berkeley.
The VErsitile GBT Astronomical Spectrometer (VEGAS) will have several modes delivered in the 12B
semester and some may become available for shared-risk observations. The current priority for the modes
is shown in Tables 7, 8 and 9. The priority for development will be reconsidered depending on proposal
demand. There will be minimal software support for both observations and data reduction. People
proposing for the shared-risk VEGAS should be willing to spend a significant amount of time helping
improve the software to use VEGAS and the software pipeline for data reduction.
For the 12B semester, only two beams of the KFPA receiver can be used with VEGAS. No resolution
is gained with the VEGAS by using only one beam of a multi-beam receiver.
VEGAS will have a larger dynamic range than the GBT spectrometer and will be able to retain a
linear response in the presence of stronger RFI than the GBT spectrometer. VEGAS will use 256 level
(8 bit) sampling of the input data.
VEGAS will be able to produce Full Stokes parameters or normal outputs (RR and LL or XX and

YY) without a loss in spectral resolution.
Minimum integration times are listed in Tables 7, 8 and 9.
The modes that will be available for this proposal call are listed in Tables 7, 8 and 9.
3.3.3 Spectral Processor
The Spectral Processor is an FFT spectrometer primarily designed for high time resolution observations.
Because of its wide dynamic range it is also useful for spectral line observations at low frequencies where
strong interference is a problem. It contains two FFT engines, each with 1024 channels over a maximum
bandwidth of 40 MHz which may be divided into 1, 2, or 4 separate pass bands. The two FFT engines
are synchronous and their outputs may be cross-multiplied to measure polarization. The most commonly
used types of observing with the spectral processor either total power or frequency-switched spectral
line observations.
Table 10 gives the general specifications for the spectral processor, and Table 11 shows the possible
combination of IF’s, frequency channels per IF, and bandwidth per IF.
18
Bandwidth Number of Number of Channels - Approximate Resolution Mininum Integration Notes
(MHz) Spectral Windows Beams (kHz) Time (sec)
250 1 or 2 1 32768 – 7.629 10
250 1 2 32768 – 7.629 10
100 1 or 2 1 32768 – 3.052 10
100 1 2 32768 – 3.052 10
50 1 or 2 1 32768 – 1.526 10
50 1 2 32768 – 1.526 10
25 1 or 2 1 32768 – 0.763 10
25 1 2 32768 – 0.763 10
10 1 or 2 1 32768 – 0.305 10 3rd priority mode
10 1 2 32768 – 0.305 10 3rd priority mode
5 1 or 2 1 32768 – 0.153 10
5 1 2 32768 – 0.153 10
1 1 or 2 1 32768 – 0.031 10 4th priority mode
1 1 2 32768 – 0.031 10 4th priority mode

Table 8: VEGAS Small Bandwidth, Few Spectral Window Modes.
Bandwidth Number of Number of Channels - Approximate Resolution Mininum Integration Notes
(MHz) Spectral Windows Beams (kHz) Time (sec)
30 8 or 16 1 4096 – 7.324 10
30 8 2 4096 – 7.324 10
15 8 or 16 1 4096 – 3.662 10 2nd priority mode
15 8 2 4096 – 3.662 10 2nd priority mode
10 8 or 16 1 4096 – 2.441 10
10 8 2 4096 – 2.441 10
5 8 or 16 1 4096 – 1.221 10
5 8 2 4096 – 1.221 10
1 8 or 16 1 4096 – 0.244 10
1 8 2 4096 – 0.244 10
Table 9: VEGAS Small Bandwidth, Many Spectral Window Modes.
Number of spectrometers (FFTs) 2
Number of IF’s per spectrometer 1, 2, or 4
Number of frequency channels per spectrometer 1024
Total bandwidth per spectrometer 40 MHz
Shortest time resolution 12.8 µs
Accumulation memory per spectrometer 256K, 32-bit
IF unwanted sideband rejection > 30 dB
Spectrum dynamic range to narrow band signal > 45 dB with taper
Input A/D dynamic range 10 dB over system noise power
Sensitivity: Average over 1 frequency channel 0.77
Wideband 1.0
Input IF range 70 MHz - bandwidth to
500 MHz + bandwidth
Accumulation modes: Synchronous with front-end switches
Synchronous with pulsar including
Doppler tracking

De-dispersed time series
Table 10: Spectral Processor Specifications.
19
Bandwidth (MHz) Number of Spectral Windows Multiplied by Number of Channels
1 × 1024 2 × 512 2 × 256 4 × 256
40 Auto, Cross
20 Auto, Cross Auto, Cross, Both Auto, Cross, Both
10 Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both
5 Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both
2.5 Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both
1.25 Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both
0.625 Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both
0.3125 Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both
0.15625 Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both
0.078125 Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both Auto, Cross, Both
Table 11: GBT Spectral Processor Modes. Columns are Number of Spectral Windows by Number
Channels while rows are for each available bandwidth. Entries indicate availability of auto-correlations
(Square mode), cross-correlations (Cross mode), or both (Sqrcross mode). When configuring the GBT,
note that: a) High resolution modes are the 1 × 1024 and 2 × 512 modes; and b) Low resolution modes
are the 2 × 256 and 4 × 256 modes.
3.3.4 Digital Continuum Receiver (DCR)
The digital continuum receiver is the GBT’s general purpose continuum backend. It is used both for
utility observations such as pointing, focus, and beam-map calibrations, as well such as for contin-
uum astronomical observations including point-source on/offs and extended source mapping. It has the
following specifications and characteristics:
• Number of input channels: 32 in two banks of 16, only one bank usable at a time.
• Switching modes: A user defined mode and Four pre-defined modes: total power with and without
continuous calibration and switched power, with and without continuous calibration.
• Maximum number of switching phases: 10 (determined by software).
• Minimum phase time: 1 millisecond. Switching frequency is the reciprocal of the sum of the phase

times in use.
• Phase time resolution: 100 nanoseconds
• Minimum integration time: 100 milliseconds.
• Maximum integration time per switch phase: 250 seconds at nominal input level and 25 sec-
onds at maximum input level.
• Blanking: At the beginning of each switching phase, 100 nanosecond resolution. Blanking time
may be different for each switch phase, but the current user interface software assumes the same
blanking time on each phase.
• Integrator type: Voltage-to-frequency converters into 28-bit counters.
20
Table 12: Bandwidths for the DCR for different receivers.
Signal From Receiver Possible Bandwidths (MHz)
IF Rack Prime Focus 20, 40, 80, 240
IF Rack Rcvr1 2, Rcvr4 6, Rcvr8 10, Rcvr12 18 20, 80, 320, 1280
IF Rack Rcvr2 3, Rcvr18 26, Rcvr40 52 80, 320, 1280
Analog Filter Rack Any 12.5, 50, 200, 800
3.3.5 Green Bank Ultimate Pulsar Processing Instrument (GUPPI)
Guppi has one hardware mode and many software modes. Guppi can be used with any receiver – with
the exception Mustang. Only one polarization would be available for the Ka-band receiver.
GUPPI currently has the following specifications and characteristics:
• 8-bit sampling: Provides dramatically increased RFI resistance, a high dynamic range and much
more accurate pulse shapes than spectrometer/spigot
• Polarizations: 2 polarizations and full stokes parameters are available
• Bandwidths: 800, 200 and 100 MHz
• Number of spectral channels: 2048 / 4096
• Minimum integration time: 40.96 µs using an 800 MHz bandwidth. (20.48 µs is possible when
on-line folding)
Details of the this backend are available at
/>3.3.6 Caltech Continuum Backend (CCB)
The Caltech Continuum Backend (CCB) is a sensitive, wideband backend designed exclusively for use

with the GBT Ka-band receiver over the frequency range of 26–40 GHz. It provides carefully optimized
RF (not IF) detector circuits and the capability to beam-switch the receiver rapidly to suppress instru-
mental gain fluctuations. There are 16 input ports (only 8 can be used at present with the Ka-band
receiver), hard-wired to the receiver’s 2 feeds x 2 polarizations x 4 frequency sub-bands (26-29.5 , 29.5-
33.0; 33.0-36.5; and 36.5 - 40 GHz). The CCB allows the left and right noise-diodes to be controlled
individually to allow for differential or total power calibration. Unlike other GBT backends, the noise-
diodes are either on or off for an entire integration (there is no concept of “phase within an integration”).
The minimum practical integration period is 5 milliseconds; integration periods longer than 0.1 seconds
are not recommended. The maximum practical beam-switching period is about 4 kHz, limited by the
needed 250 micro-second beam-switch blanking time (work is underway to reduce the needed blanking
time). Switching slower than 1 kHz is not recommended.
Under the best observing conditions (clear, stable, and few or no clouds) the combination of the
Ka-band receiver and the CCB deliver a photometric sensitivity of roughly 0.2 mJy for a single one-
minute, targeted nod observation. The median sensitivity is 0.4 mJy RMS. These numbers apply to the
most sensitive (33-36 GHz) of the four frequency channels; averaging all channels together will slightly,
but not significantly, improve performance because of noise correlations and variations in the receiver
sensitivity between channels. The analogous noise performance figures for sensitive mapping projects
are still to be determined.