[evlatests] 700KHz interference from deformatters.

[Mike Revnell]

The jumper that bonds the shell of the SMA output connector to the board 
ground got left out of the assembly procedure. It looks like most of the 
boards are working ok without it. A couple things to check.. Be sure 
that the screws securing the boards in the bins are secure and that the 
shields of the cables are bonded at the rack panels. If this doesn't fix 
the problem the jumpers can be installed. Installing the jumper requires 
the board to be removed from the bin and a piece of wire soldered in. 
It's enough work that it isn't justified for the ones that aren't a problem.

Mike.

Jim Jackson wrote:
> There is also an RF Attenuator installed on the output of the 
> deformatter to reduce the level a little more. I believe it is a 6dB 
> part.  If this happens to be missing, the birdie will be a higher but 
> the RF level might otherwise seem ok since the T5 ALC will probably 
> compensate.  Someone may want to check to see that these couple 
> deformatters have that attenuator installed on the output jack.  If 
> these boards were swapped out for repair at some point, the 
> attenuators may have been missed.
>
> Cheers,
> Jim
>
> At 02:48 PM 1/6/2009, Vivek Dhawan wrote:
>   
>> This message is mostly to Mike Revnell (or delegate) since
>> he fixed this last time.
>>
>> A survey was done on all 4 IFs on the available EVLA antennas
>> 1 2 3 4 5 11 13 14 16 17 18 19 21 23 24 25 26 28.
>> X and C bands were surveyed at 2 BW settings.
>>
>> The birdies are strong on 3L and 28R; 3L is weaker.
>> IFs 1 and 2 are the same. Other antennas/IFs were clean.
>>
>>
>>
>> Background for the record (memories are mercifully short):
>>
>> The interference is in the backend, not related to any sky signal.
>>
>> It can be seen in autocorrelation, does not correlate between antennas.
>> (It might correlate in RL self-spectra, I did not think to take the
>> data to check this and do not consider it worth doing.)
>>
>> It comes from the deformatter switching power supply, and appears at
>> ~700KHz away from the LO edge (top of X band, bottom of C and other
>> bands) in any bandwidth 780KHz or more, at any sky frequency.
>>
>> It is narrow (few KHz), and sometimes the 2nd harmonic may be seen
>> in addition to the 700.
>>
>> It is believed to be purely a transition problem, the signals are
>> getting into the analog path that is re-constituted from the digital
>> EVLA signals. Once the deformatters are in the WIDAR correlator this
>> goes away.
>>
>> It is fixed as follows: (Revnell email 2005 december 31) There is a
>> jumper on the line driver output that goes to the VLA baseband system.
>> Leaving the jumper off makes the birdie 10's of dB stronger.
>>
>> _______________________________________________
>> evlatests mailing list
>> evlatests at listmgr.cv.nrao.edu
>> http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests
>>     
>
>
> _______________________________________________
> evlatests mailing list
> evlatests at listmgr.cv.nrao.edu
> http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests
>
>