[evlatests] 4-antenna tests of 3-bit samplers

[Rick Perley]

    A short (5 minutes) observation of 3C84 and a nearby blank field 
were taken this afternoon, with the 4 antennas outfitted with the 3-bit 
samplers.  Full polarization in a single IF,  128 MHz BW, and 64 
channels.  (This is an OSRO mode). 

    Everything looks quite good.  Amplitude stability was good, other 
than a small drift which is perhaps due to pointing (no referenced 
pointing was applied, and the observation was at K-band).  Phase 
stability is appalling, exactly as expected for a summer afternoon with 
high humidity and big clouds.  Bandpasss looked normal. 

    I self-calibrated the observation, and looked at the noise 
histograms.  For the blank field, all histograms (RR, RL, LR, LL) are 
the same, with about a 1 Jy dispersion -- this is about right, presuming 
SEFD ~ 1000, 2 MHz channel width, and 1 second integration. 
    For 3C84, after self-calibration, the 'real' histogram has the same 
width as the noise field, but the 'imaginary' one is about twice as 
high, indicating unresolved phase fluctuations, probably from the delay 
clunking.  (I didn't take these out). 
    Images made of the noise field show the rms going down as expected 
with time and bandwidth. 
    An image of 3C84 gave 1000:1 DR -- this with 3 minutes of data and 6 
baselines.  Probably as good as we can expect. 

    All looks good so far ...