[evla-sw-discuss] total power

[Jim Jackson]

The Total Power detectors used in ALMA perform the 2ms integration in the
A/D converter chip (an Analog Devices AD7731).  It appears they can be
configured to integrate for up to at least 20ms. I would like to see the
MIB designed to handle the worse case we expect to see, which I believed
will be defined by Solar Observation or RFI excision.  If 10ms is fast
enough for all appplications, then we certainly can use it but if any
special modes require faster rates, we need to provide for that. We will
really need the scientists to tell us what is an acceptable maximum rate.
The integration (or output rate) of this A/D is software configurable. 

Cheers,
Jim

At 04:31 PM 1/23/2002 -0700, Bill Sahr wrote:
>During the LO/IF PDR it became clear that total
>power measurements may be used more widely in 
>the system than previously thought.  I would like
>to connect this impression to a statement made 
>by Barry - that TP measurements made every 2 ms
>are really not necessary for the EVLA.  Every
>10ms or so would be adequate.
>
>It would be useful, and may be important, that
>the MIB not be asked to deal with TP measurements
>every 2 ms if 10ms is sufficient.  The 10 ms
>interval would leave a great deal more spare time
>for the MIB to attend to other tasks.  The issue
>here is not one of network traffic.  If measurements
>were needed only every 10ms, but arriving at the MIB 
>every 2 ms, the MIB could average (or discard) and
>send out the values at 10ms intervals.  Rather,
>the issue is one of not occupying the MIB with 
>work that does not need to be done.  The current
>plan is to use the ALMA device, unmodified, for
>the TP measurements for the EVLA.  Would it be simple
>or complex, time-consuming or trivial, expensive
>or cheap to modify the ALMA device to produce TP
>measurements at a rate appropriate to the needs
>of the EVLA ?  Would it be desirable to do so ?
>_______________________________________________
>evla-sw-discuss mailing list
>evla-sw-discuss at listmgr.cv.nrao.edu
>http://listmgr.cv.nrao.edu/mailman/listinfo/evla-sw-discuss