srt.hlp

###Help Data File for the SRT Help window
### Original by Eric Evarts in 2005
### Each set of information is started with ##command_name
### Each set is ended with ##

##home
Welcome to the SRT Help Window

Please click on the buttons above to get more information about the features of the SRT software.

Plot: Click on spectrum to generate a PostScript Plot  with filename yyyy_ddd_hr_mn.ps

srt.cat keywords (see also sample srt.cat file):

STATION lat_deg_North lon_deg_West name 
AZLIMITS azlim1_deg azlim2_deg - for site in north suggest 5  355   for site in southern hemisphere 185  175
                               - azlim2 < azlim1 centers display to the north for sites in southern hemisphere 
ELLIMITS ellim1_deg ellim2_deg
*STOWPOS  - stow position - recommend using default which is stow at azlim1 ellim1 limits 
*RMSCALC - calculates rms voltage in units of digits and displays as srms otherwise smax is displayed
         - displaying smax is recommended a value above 2047 and 127 indicates ADC saturation for pci card dongle respectively
NOISECAL - strength of noise injection from antenna vertex if available 
TCAL - temperture of absorber 
CALMODE - recommend 3
SIMULATE ANTENNA RECEIVER FFT
MAINTENANCE - disables a stow position used for testing
PLOTSEC - not used
NOCLEARINT - prevents automatic clear of integration on source change
NBSW - number integration periods per beamswitch cycles
NBLOCK - number of data acquisition block per display cycle
NUMPOLY - number of terms in polynonial fit to the bandpass
BEAMWIDTH deg - antenna beamwidth used to set spacing of 25 point scan 
TOLERANCE - not used
CASSIMOUNT - no longer supported
H180MOUNT - no longer supported
ALFASPID - antenna mount type for old controller
AZCOUNTS - for old contoller
ELCOUNTS - for old controller
AZELPORT - for old controller
COMMAND - if used this will start running this file when program is started
TSYS - used in calmodes 2 and 3 - is calculated when calmode 20 is performed
FREQUENCY - frequency set at startup
RESTFREQ - for vlsr calculations
FREQCORR - frequency correction for dongle in MHz
BANDWIDTH - MHz bandwidth used - recommend 2 MHz 
NUMFREQ - recommend 256 for dongle and up to 1024 for PCI card
DATADIR - data directory - defaults to directory with the executable
COUNTPERSTEP - for old contoller
RECORD - number of periods per output add [SPEC] to include spectrum [RCLR] to clear at start of each block 
NODISPLAY - can be used when running from command file
NOPRINTOUT - reduces information written to screen
DEBUG - adds information written to screen
SPEED_UP - speeds up time for simulation
START_DATE 2014:110:12:0:0  // date for simulation - use SPEED_UP to advance
SOU - source ra dec epoch   // source names Sun and Moon are keywords and have zeros for ra and dec
GALACTIC - lat long
AZEL - fixed positions 
RFISIGMA - level at which to report RFI default = 6 
RFI - used to remove RFI from data which cannot be removed at the source
also see comments in srt.cat

##

##mode
The SRT operates in a total power mode. The bandpass is normalized by clicking the cal botton.
NPOINT does 5x5 point scan to find peak power and pointing offsets. BEAMSW can be used for sources
which have suitable reference spectra on either side of the source. the beam throw equals the BEAMWIDTH 
set in srt.cat. NBSW sets the number of integration periods in each beam position. 

##

##npoint
NPoint

pointing adjustment:
run npoint on the sun and subtract the azimuth offset from the start of the azimuth limit
subtract the elevation offset from the starting elevation limit

screen saver:
the screen saver may cause the srt program to stall
possible solutions are:
a) minimize srt program when not using
b) turn off screen saver

error calculations:
the noise in each spectral bin for a single response from
the receiver (a single line in the output file) is given
by tsys / sqrt(resolution * intg. period)
the resolution for the digital receiver is equal to the spacing.
However in many cases the errors are determined by systematic effects. 
For example the errors in a beamswitch continuum observation are estimated from the variance in the data from many on/offs
##

##beamsw
Beam Switch - click on beamsw to start and stop
total power result is displayed at bswpwr. The Moon should give about 1 K. 

##

##freq
Frequency in MHz - optional arguments are bandwidth and number of frequencies

##

##offset
Offset - click on offset and enter az el offets in degrees with space between entries

##

##record
Record - click on record to start and stop 
RECORD 10 in srt.cat sets interval to 10 averaging periods
this writes the continuum record and is an independent output each 
averaging period  
RECORD 10 SPEC adds the accumulated spectrum to the output
click on clear to clear the spectrum and clear the spectrum integration.
The accumulated spectrum is normalized by the number of integration periods. It can be
unnormalized by multiplying by the number of integration periods so that subtraction of the
previous unormalized spectrum gives the spectrum without accumulation which then can be used to
remove periods of RFI from the accumulated spectrum. 
RECORD 10 SPEC RCLR  adds the accumulated spectrum to the output and clears the integration each output

##

##cmdfl
Command file rules:
(default filename cmd.txt) 
The COMMAND keyword in srt.cat file can be used to start a command file automatically when
the program is started. If you want to be able to enter a command file name remove cmd.txt
to force a prompt allowing a different name to be entered.

1] reads one line at a time skipping blank lines, and lines which start with * and lines with past times
2] stops at and executes any line with current or future time time format:
line starts with yyyy:ddd:hh:mm:ss and is followed by cmd
alternate format:   LST:hh:mm:ss   cmd
current time format:   :   cmd
current time plus n seconds format:   :n   cmd
each line MUST start with time or :  or * for comment

command keywords:
items in [ ] are optional
sourcename,mode,radec,azel,galactic,offset,stow,calibrate,noisecal,record,freq,roff
 sourcename (any name in catalog) [mode]
 mode n(for 25_point scan)    b(for beamswitch)
 radec hh:mm:ss  [sign]dd:mm:ss [epoch] [mode]
 azel az_deg el_deg
 galactic glat_deg glon_deg
 offset azoff_deg eloff_deg
 stow
 vplot (writes Postscript plot to disk with filename yyyy_ddd_hh_mm.ps)
 calibrate (uses mode specified in srt.cat)
 clearint (clears the spectrum integration)
 record [optional filename]
 roff (turns off record)
 freq frequency [num] 
Note that 25_point scan runs only once while beamswitch until next cmd
For example:
:  Sun
:60
:600 Sun n
:  stow
goes to the sun,waits 60 seconds,does 25 point
and after 600 seconds goes to stow
Note that the seconds to wait is right next to the ":"
and otherwise there is a space before the command

 A command file can be checked by running in simulate mode  -see help on cmdline options
another name for a command file can be entered at the prompt if the default doesn't exist
##

##cal
Calibration - various modes are available enter desired mode
in srt.cat file using CALMODE keyword. Suggest CALMODE 2 or 3
The basic method is to get the system temperature from the "Y-factor"

   Y-factor = power_on_hot_load / power_on_cold_load
                = (tsys + hot_temperature) / (tsys + cold_temperature)

   solving   tsys = (hot_temperature - y*cold_temperature) / (y-1)

   power_on_hot_load  = with absorber over the feed  or pointing at the trees
   power_on_cold_load = looking at cold sky
   hot_temperature = ambient ~ 290 K
   cold_temperature = 3 K from CMB

Since the LNA noise temperature is very stable is in not really necessary
to get the system temperature every time you point the telescope. What is
needed is to get the power looking at an absorber or pointing at the trees fairly
frequently as the gain of the LNA + amplifiers + dongle change with time.
CALMODE 3 is appropriate if you can point at the trees. In this case just click
on cal and the software will divide by a reference spectrum and convert the power to kelvin
assuming the power while still pointed at the trees corresponds to Tsys + Tcal. 
CALMODE 2 does the same by waits for the operator to hit enter when the absorber
has been placed over the feed.

SRT tests:
 Check the y-factor and if high run tests with LNA disconnected to check "back-end" noise 
contribution.
 Observe the Sun to check pointing and beamwidth. The Sun is variable but you should get
at least 800 K.
 It is possible to observe the moon using the beamswitch and you should get about 1 K.
 Observe the H-line - results can be compared with srt info online 
 - typical antenna temperature you should observe in the Galactic plane is about 65 K.  

RFI:
  RFI is becoming more of a problem as the spectrum becomes more crowded and we have more
electronic devices. The 1400 - 1427 MHz band is reserved for passive use only worldwide.
However strong transmitters not in this band can cause interference it very nearby and in
some cases may require the use of a cavity filter between the antenna probe and the LNA.
Low level signals in the 1400 - 1427 MHz band are produced by PCs, printers, etc. and it may be
necessary to use a handheld receiver like the AR8200-MK3 to locate these signals and shut them down,
shield them with absorber, or mode the SRT away from these sources. 
A good choice of absorber for calibration at 1.4 GHz is eccosorb AN-77

##