hf xmiter setup notes
Some Notes on setting up the hf transmitter for
testing and data taking.
Most of the examples assume you are sitting at the hfctrl computer
in the hf building..
Some topics are:
receivers for monitoring:
- cover alfa
- vnc session to control 305m telescope, datataking. .. this is
normally done during testing.. for an hf run, you won't do
- using megs3 vnc sesssions:
- vncviewer 184.108.40.206:1 (megs3) (need phil's
password)... should probably setup hfuser on dataview?)
- open terminal
- ssh -Y observer2 -l dtusr (need password from
- ask operator to exit the cor vw% window running on observer2
- datataking cor
- The vw% commands will then be entered in this window.
- vw% commands:
- commands below are entered in vw% window we opened above.
- monitoring with only 430 ch
- gr430protect. closes all shutters
- cover alfa
- ifsetup ch
- dopset vel 0 top .. dopsetrfonly .. dopsetfrq 430
.. dopsetdoit .. do set the hdr freq to 430
- monitoring with dome 430 reciever
- do dual beam
enable so we can close shutters but leave 430 rcvr
- ifsetup 430
- dopset vel 0 top .. dopsetrfonly .. dopsetfrq 430 ..
- Start interim correlator running
- cor "cor_bwnum 2 cor_lagc 9 cor_dmpsper 1 cor_brd 6 "
- adjpwr if1
- adjpwr if2
- Note tha`t polB will be on load.. so you only get data
- disc on
- cor 9999 ... starts recording data to
/share/olcor/corfile .. 1
monitoring of ch with interim correlator
- this is done during hf testing, for an hf campaign you won't
- need to start correlator running (see above).
- Idl is not on hvctrl, so you need a vncsession on another
- in vnc session open a terminal on megs3, then
- you may want to set the hor, vertical scales (rather than
- ctrl-c, retall
- hor, minvall,maxval
- then restart monitoring.
Start synth control gui
- At hfctrl, click on button named synth xxx (need to
get the name).
- click file, load one of the setup files. rhcp, 5 or 8
- check that tx is ready to go, broad bands on
- raise harware synth enable switch
- clock send to hardware
- clock enable rf.
monitoring with rf port switch and anritsu
- running the anritsu to record the power and running idl to
display the data is done on galfas2.. so we need to open a vnc
- Open vnc session on galfas2
- On hfctrl:
- vncviewer galfas2:4
- It uses the standard guest password.
- the Hf_Anritsu and Hf_idl button are in this window display.
hf tx power with anritsu and rf port switch.
- This uses the rf multiplexor, the anritsu spectrum analyzer,
and the anacom program to drive the anritsu
- clicking the Hf_anritsu button will:
- execute the script in /share/megs/phil/x101/hf/hfmon.sc.
This script will:
- set the command file used to
- this file has the information to drive the anacom program.
- It then starts the anacom program.
- source: /share/megs/phil/svn/aosoft/src/anacom/anacom.c
- this program controls the anritsu and the rfport switch.
- It runs as a 32 bit program (because of the visa
libraries is uses).
- The data will be written to
- yymmdd is the ast date when you started the program.
- By editing the command file, you could send the data
- the program will output the current port it is working on
to the terminal screen
- if there problems with i/o (usually the rfport) it
will retry about 3 times before giving up..
power data with idl:
- the Hf_idl button will display the anritsu data from the
- The button executes the script: /local/guest/monhfpwr.sc
- The button does not provide a date, so the script uses the
current date for the files to monitor.
- Once idl starts, it will start monitoring using 5 MHz as the
- The upper plot is the spectra, the bottom plots is the peak
power vs time.
- hitting any any key will bring up a menu that lets you
- the band 5 or 8
- horizontal vertical display for the spectra or the total
power vs time
- the menu is:
- Cmd CurVal function
band to use. 5 or 8
high set spc hor plot range ..no args
high set spc vertical plot
range ..no args --> autoscale
high set pk hor plot range..1 arg->
startpos,no args --> autoscale
high set pk vertical plot
range ..no args --> autoscale
delay 1.00 plot delay
offset correction. add to data
number of file to use
track cursor position
avg,max,min to plot
0 step mode.
stop in procedure (for debugging)
- Idl Examples:
- You need to hit 1 key each time to get the menu. After
you type in a command it goes back to monitoring.
- switch to band 8
- band 8
- you probably then want to also do:
- hspc 0 20 .. to change the
scale 0 to 10 Mhz to 0 to 20 Mhz.
- Blowup vertical scale on the total power vs time to
- Change the pwr vs time to start plotting at 13.5
- Change to single step mode
- s 1
- after each plot, it will stop in the menu. Hit a key
to got to next plot . to got back to continuous
- s 0
- Move back or forward in the
- num fileNum
- The filenums get incremented after each 6 spectra are
collected. Unfortunately there is currently an easy way
to move in time.. I need at this...
- Note: the idl program can be run from any computer with
access to idl.
- for a terminal:
- /share/megs/phil/x101/hf/hfmonpwr.sc yymmdd
- if the program aborts, i think the screen will disappear
before you can read the error message (buttons are wonderful :).
- a kludge work around will be to change the button to have
the window stick around after exit
- right click on button -> properties ->
application->advanced options-> (do not close on exit)
- The Hf_Anritsu program selects the current data yymmdd when
started as the output directory. If you cross midnight it will
continue to write to the previous day's directory.
- The monitor program (see below) also defaults to the current
- if the monitor program (Hf_idl) is started after midnite, but
the Hf_Anritsu is still running from the previous day, then
Hf_idl will just wait for data in the new directory .. which
- the fix is to restart the Hf_Anritsu program.
Synchronizing the HF tx with the 430 Tx
- The agilent signal generators have an external trigger input
that can be driven by the 1 or 10sec tick
- If the length of the buffers driving the hf tx are a multiple
(or sub multiple) of 1 second, then after starting on a
tick, the buffers will remain synchronized to the tick (or you
could retrigger on each tick).
- The ipp for the 430Tx is not normally synchronized to the 1
second tick. You must manually do it before the hf run.
- once you run the synchronization command, the ipp will
remain synced to the 1 sec tick as long as the ipps divide
evenly into 1 second.
- To synchronize the 430tx to the 1 second tick
- go to the vxworks control computer in the receiver room (in
front of the oscilloscopes).
- In the window that has the vw% prompt:
- This will execute the sps config file in
- this will synchronize the ipp to the 1 sec tick
- Don't do this with the tx pulses in remote (switch
to local).. since the routine will cutoff the pulses to the
430tx until the tick arrives.
- The buffer sent for synchronization is a copy of a sbclp
- The ctrlStart field in the file has been changed from:
- ctrlStart 6 ... start on
endOfBuffer, recirculate on end of buffer
- ctrlStart 4 .. start on
tick, recirculate on end of buffer.
- To verify that the ipp is synced
- On the oscilloscope:
- chan1 1 sec tick
- chan2 : tx ipp from sps
- trigger on chan 1
- set time base to a few 10's of usecs
- the two pulses should be within a few usecs.
- After synchronizing.. you should reload the sps buf you want
to use (eg .. run rd, send whatever buf you want).
- rebooting the da crate (in the control room) will probably
cause the synchronization to be lost,.
- synchronization will rely on the ipps of the 430 to remain
divisible into 1 sec.
<- page up