Issues and help dealing with Q5/Q6 in the upgraded SWARM
Higher Noise at IF frequencies 12-16 GHz
The sensitivity of the SMA system falls off at the higher IF frequencies covered by the newly commissioned "Q5" and "Q6" units of the SWARM digital backend. The figure below shows an example of the reduced sensitivity in the 12-16 GHz part of the IF that covers sky frequencies 333-345 GHz (LSB) and 353-365 GHz (USB) in Q5 (green) and Q6 (light blue); the noise will be about 20-30% higher in these windows, and up to 2-3 times higher at the upper edges of these bands.

Figure 1: Amplitude gain and system equivelent flux density (SEFD; sensitivity) across the full band of the "400" reciever for Antenna 1. Q1 is shown in dark blue, while Q5 and Q6 are green and light blue respectively.
Higher Noise at the upper edges of SWARM Q5 and Q6
The fall off in sensitivity is also seen within the Q5 and Q6 bands themselves. In the figure above you can see the sensitivity difference between the lower and upper ends of Q5 and Q6 is significant. As such, care should be used when setting up your observations to avoid putting lines in the upper IF regions of Q5 and Q6 if you cannot avoid putting them in those bands completely.
Dealing with limitations of Q5/Q6.
1) As a result of the lower sensitivity of Q5 and Q6 in general, it is best to avoid placing important spectral lines in these windows if possible. It is especially important to avoid placing spectral lines in the upper ~15-20% of the IF in Q5 and Q6.
2) For continuum imaging, it is important to check the upper IF edges of Q5 and Q6 for high amplitude spikes. The reduced sensitivity in these parts of the band often result in spikes caused during the bandpass calibration stage owing to "division by almost zero" parts of the process. These spikes can cause ringing in your final images.
Please feel free to contact us at sma [hyphen] propose [at] cfa [dot] harvard [dot] edu if you have any questions.
Last updated: 2021-Jan-06 by Glen Petitpas
|