[Gb-ccb] CCB Integrations

[Brian Mason]

All- I have reviewed the change in the CCB design to a 10 MHz ADC
sample rate, and if my understanding is correct (please check me as
below) the consequences are not problematic. Nor is there a clear need
for configurable manager level integration.  I do have one question
however: is sample_dt now a dummy variable always having a value of 1?
(with units of 100 ns)

It is the case that the CCB can be configured to overflow the
integration buffers rather easily. This is fine, but we'll need a
rule, probably at the manager level, to prevent it.  I have worked out
the rule to be

 samp_per_state * integ_period < 2^20 =~ 10^6

This rule is independent of the number of phase switches in
use. Martin and/or Tim, can you check me on this? This is mostly to
validate my understanding (of the document, and our emails...).

At this limit-- the maximum integration time without register
overflow-- you get physical integration durations of 0.1, 0.2, and 0.4
seconds for 0,1, and 2 switch phase switching. Here 0,1,2 refer to the
number of phase switches per radiometer that are active, hence in Ka
band there are really 0,2 or 4 switches actually toggling since there
are 2 radiometers in the receiver (nominally for the 2 polarizations
though they are mixed).

 -Brian
-- 
--------------------------------------------------------------------
Brian Mason                           | office: +1(304)456-2338
Assistant Scientist                   | fax:    +1(304)456-2229
National Radio Astronomy Observatory  | mail:   PO Box 2
bmason at gb.nrao.edu                    |         Green Bank, WV 24944
http://www.gb.nrao.edu/~bmason/       |
--------------------------------------------------------------------