Specifications: Folding mode

Will provide two (XX, YY) or four (XX, YY, real(XY), imag(XY)) products. Maximum of 2048 phase bins for 2048 frequency channels for four products, each 32 bits. The number of products, phase bins or number of channels can vary by factors of two with a maximum of 2048 x 8192. Normalization count to be recorded indicating number of output samples in a particular phase bin. Pulse period range (for 2048 frequency channels per product and 2048 phase bins) 4 ms - 20 s. Shorter folding periods to 0.5 ms possible with proportionally reduced phase bins Minimum integration time = 10 s. Fractional folding period precision < 1.e-12 The polycos required for folding will be defined using a new polyco definition (the definition of which is currently being determined). Maximum data output rate = 6.4 Mbyte/s. Output data files in PSRFITS format. Synchronous TTL signal for calibration driver, adjustable duty cycle and phase.

Setting up PDFB2

It is likely that changes in the set-up requirements will occur frequently and so check first with the Parkes staff and update this wiki if necessary.

Software setup

Setting up the PDFB2

bourbon% ssh corr@pkccc1
pkccc1% pdfb2

Running SPD monitoring software

bourbon% ssh corr@pkccc1
pkccc1% spd
spd: mode b
spd: a - bandpass
spd: sel pp11 - pulsar phase vs amplitude
spd: sel biaa (or bibb) - two-dimensional pulsar phase - frequency display

Running tcs

pavo% tcs

Select "expert mode" and "PDFB2".

Hardware setup

The PDFB2 can process up to 1GHz of bandwidth. Currently at Parkes we use the PDFB2 at 10cm with a bandwidth of ~1GHz and a bandwidth of 256 MHz at 20cm. It is currently necessary to change the wiring setup to change between these systems (ADD INFO ABOUT THIS HERE)

(DESCRIBE SELECTING RECEIVERS ETC.)

Analysis of PDFB2 observations

Files produced by the PDFB2 have filenames starting with an "r". The raw observations are stored in /nfs/PKCCC1_1/. These observations are subsequently copied to Epping and can be accessed from the @@$DFB2$$ directories. The observations can be processed and analysed using the PSRCHIVE suite of software. Typical processing commands include:

Bad channels and subintegrations can be deleted using

> pazi file.rf

In order to frequency, polarisation and time scunch the profiles, use:

> pam -e FT -FTp *.rf

which will produce *.FT files

To add profiles together and get a frequency and time scrunched summed profile, use

> psradd -f added.F *.FT
> pam -T -e FT added.F
To get a standard profile

>pas -s added.FT

in pas you can zap baselines or smooth the data. When finished press 'S' to save, with filename e.g. '20cm.std'

To obtain TOAs for all your .FT files:

>pat -f tempo2 -s 20cm.std *.FT > mytim.tim

and then time as normal using your .tim file.

View profiles etc. using

>pav -DFTp myprofile.FT

Use the -h option (e.g. pam -h, psradd -h) to get a full list of options.

After each observing session

John Sarkissian starts the web-page checklist and writes report providing information on the last observing session and data processing. The aim here is to highlight bad observations/hardware errors/amount of RFI etc. John Sarkissian checks whether the data have been correctly archived to DLT4. NOTES REQUIRED FOR HOW THIS PROCESS IS CARRIED OUT. John Sarkissian copies WBC/DFB data from correlator disks to lagavulin by running: > /psr1/TAsort_lagavulin_dfb.csh > /psr1/TAsort_lagavulun_wbc.csh John Sarkissian pre-scrunches the data files and looks for bad subintegrations and recording whether the files are bad and should simply be ignored (i.e. the observation did not start correctly etc.). He produces a text file containing the processing commands required: /psr1/jms/processed.txt John Sarkissian copies the raw data files (DFB/WBC) from Lagavulin to Epping. > cd /data/LAGAVULIN_4/pulsar > cd DFB > rsync -av . pulsar@zwicky:/pulsar/archive06/DFB/ > cd ../WBCORR > rsync -av . pulsar@draco:/pulsar/archive04/WBCORR/ This procedure will typically take a few hours. John Sarkissian copies the processing commands to Epping using > scp processed.txt pulsar@tycho:/psr/preProcessing/. (Note: this file is automatically backed up as it is on /psr disk) After processing has been completed by the Swinburne group the CPSR2 data are copied to Epping

> ssh pulsar@pegasus > cd /import/cpsr1/archives/____ (name of most recent session) > rsync -av . pulsar@zwicky:/pulsar/archive05/CPSR2/. > change the name of the directory containing the data to append _epp John Sarkissian completes the web-page checklist. Dick Manchester: After confirmation from John S. that the DLT4 backup and file transfer to Epping are complete, the files on the Parkes disks can deleted. The process for deleting data is given at the bottom of this page.