[evlatests] No Wobbles at S-Band!!!

[Jim Jackson]

Actually, I think there are two separate problems.

The one we know about and have a solution for causes the 1408 MHz 
birdie in L-band and is related to the 512MHz comb mixing with the 
L301 synthesizer output to create birdies that get through the T302 
LSC converter.  It is a 512 MHz comb, not 128 MHz.  We are certain of 
that one because we can see it on a spectrum analyzer and putting a 
high pass filter on the output of the L301 fixes the problem. It also 
removes other 512 MHz birdies in S and C band we can see in the lab 
but that no one has ever actually complained about.  We have been 
patiently waiting for the RFI sweep before buying these filters but 
I'm about done with that and am ready to just buy the filters, 
install them on the LO input to the T302 and solve the problem we know about.

The second problem is an intermittently reported 128 MHz comb that 
apparently is intermittently seen on some unspecified group of 
antennas in S-band only. We have never seen it. Every time I have 
asked about it I get an inconsistent answer as to whether its even 
still there or not and no one has seemed particularly interested in 
worrying about it.  If it's real and causing problems, someone needs 
to tell us exactly what antennas it's seen on, when it is seen, and 
how the antennas are set up at the time. If it is a 128 MHz comb line 
radiating in the system and is seen all the way across S-band, it 
seems rather surprising it does not also appear in at least part of L 
and/or C bands. Other than the receiver (and it's M&C and power 
connections), the signal path is identical. I suppose it could be 
possible it is getting into some S-band receivers on power, ground or 
M&C signals due to some wiring or grounding problem, but everything 
should be reasonably filtered to prevent that.

We'll talk about it in the meeting on Monday.

Cheers,
Jim


At 04:41 PM 3/11/2011, Rob Long wrote:
>We can spend money to eliminate the known problem at the risk of needing
>to "re-spend" money to fix the as of yet undiscovered problems. We were
>waiting on the RFI sweep to ensure we get the correct filters to resolve
>all the issues.  We know what we think we need to solve the 128 birdies,
>but what if there are other issues that the filter we buy doesn't
>correct. Not my decision of course, but that was our rationale.
>
>Rob
>
>Rick Perley wrote:
> >     The origin of the 'wobbles' seen yesterday is now understood.
> >
> >     It is due to an RFI spike (actually, an internal 'birdie') at 2944
> > MHz, strong enough to beat with the continuum emission of 3C147, and
> > visible only on the north-arm baselines because of the (relatively) slow
> > fringe rate at the time.
> >
> >     A bit of numerology:  2944 = 23 * 128.  This is the 23rd harmonic of
> > a 128 MHz fundamental tone.  The problem is well known -- indeed,
> > perusal of the spectrum shows a harmonic in *every* S-band subband
> > (conveniently, each is 128 MHz wide, so it's easy to find).  The odd
> > multiples are stronger, as expected.
> >
> >     Not only is the origin of this problem known, but so is its
> > solution.  We have been waiting for a long time to survey the spectrum
> > for other such internal birdies, but it seems that this 'RFI Sweep'
> > survey is still some time away.   Perhaps it's time to get on with
> > eliminating this known problem?
> >
> >
> > _______________________________________________
> > 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