[evlatests] Ka band power level pulsing (fwd)

[Ken Sowinski]

I thought the following message deserves a wider audience.
The context is that Rick and Bob found a ten second glitch
in TP of IF B while using the conversion path of the T303
and doing tipping scans.  The origin turns out to be in how
the M301 is commanded when the conversion path is requested;
a solution is proposed.


---------- Forwarded message ----------
Date: Tue, 10 Jun 2008 14:29:55 -0600 (MDT)
From: Ken Sowinski <ksowinsk at aoc.nrao.edu>
To: Ken Sowinski <ksowinsk at aoc.nrao.edu>
Cc: Barry Clark <bclark at aoc.nrao.edu>, Robert Long <rlong at aoc.nrao.edu>,
     Robert Hayward <rhayward at aoc.nrao.edu>, Rick Perley <rperley at aoc.nrao.edu>
Subject: Re: Ka band power level pulsing

Staring at the block diagram and a moment of thought explained
what is happening.

When selecting K, Ka or Q band the M301 sets the T303 output
switches to replicate the direct connection into both outputs,
so that A is copied into B and C into D.  When the conversion
path is to be used one of the switches must be reset to route
the mixer output to IF B and D and this is done with a separate
command to the M301 after the band select command.  It is this
sequence of commands which causes a glitch in TP at every scan
initiation.  In pointing or tipping scans, as currently implemented,
this is every ten seconds.

If the signal interruption is deemed a problem we can make it
go away by inventing three more band codes for the M301 to
represent K, Ka and Q band using the T303 conversion path.
U band is not at issue because the M301 already sets the switches
correctly in that case.