New Allen Telescope Array Development

June 05, 2018 - received the mini circuits attenuator and 8 port RF switch. Two of each.

Note: That is a quarter coin in the center for size reference.

Plugged into VM server ngs-work1 (sonata@10.1.49.174, polaris)

1. Attenuator RUDAT-6000-30: https://www.minicircuits.com/WebStore/dashboard.html?model=RUDAT-6000-30
2. RF Switch USB-1SP8T-63H: https://www.minicircuits.com/WebStore/dashboard.html?model=USB-1SP8T-63H

Created 2 programs, rfswitch and atten to control these units over USB. See https://github.com/jrseti/ATARFSwitchUtils for a full explanation and the source code.

Setting up Proxmox USB ports

root@nsg-headnode:~# qm monitor 101
qm> info usbhost
  Bus 3, Addr 5, Port 14.1, Speed 1.5 Mb/s
    Class 00: USB device 0557:2419
  Bus 3, Addr 8, Port 12.4.4, Speed 12 Mb/s
    Class 00: USB device 20ce:0023, Mini-Circuits
  Bus 3, Addr 7, Port 12.4.2, Speed 12 Mb/s
    Class 00: USB device 20ce:0022, Mini-Circuits
  Bus 3, Addr 10, Port 12.4.1.3, Speed 12 Mb/s
    Class 00: USB device 20ce:0022, Mini-Circuits
  Bus 3, Addr 9, Port 12.4.1.1, Speed 12 Mb/s
    Class 00: USB device 20ce:0023, Mini-Circuits
    

Note the port numbers (like Port 12.4.4)

In Proxmox, assign the usb ports as shown:

The head node needs to be rebooted to make the USB changes work.

The RF switch and attenuator units are installed at the top of the FXA Rack just to the right of the RFCB rack:

Various feed measurements using the spectrum analyzer. Using feed 4jx. May 09, 2018

In all these pictures please pay attention to the dB and bandwidth scaling, these are not the same in all pictures.

Turns out that without the filter there is 600MHz bandwidth to play with.

With 200MHz filter, pointed at az 330, el 30:

Without 200MHz filter:

Pic of the filters, and the 200MHz filter bypassed for feed 4jx:

For reference, here is the original RFCB drawing I found:

CasA at 1420, looks like it always does:

Did not see the bump at 1665MHz.

1613.9MHz:

1667.348MHz:

1718.74MHz

Using the spectrum analyzer I discovered there is a spike every 8MHz from 1000 to 2016MHz. ONLY on the new feeds. No matter where i pointed. The old feeds do not have this spike.

On feed 3Lx it was very strong, this is a 120MHz badwidth shot: