[evlatests] T302 Linearity Test

[Jim Jackson]

One thing to clarify before anyone gets too worried.  This was a 
special test to measure the LO power at the T302 outputs (which we 
know is there) and how good the T304 input filter is at rejecting it. 
These plots are taken with a spectrum analyzer on "max-hold" while 
cycling the L301 through all of it's possible settings.  There really 
isn't a comb like that on the T302 outputs.

There is a separate issue open about apparent 128MHz comb lines on 
L,S and C band in a few antennas that seems to come from the 
T302.  We've just identified the cause of that today. I'll report on 
it at one of the meetings next week.

Cheers,
Jim

  10:34 AM 5/6/2011, Keith Morris wrote:
>I measured this recently in the lab.  LO-filter.pdf shows the LO 
>leakage out of the T302, with (red) and without (blue) the T304 
>input filter. The second file, noise-plus-LO.pdf, shows the filtered 
>LO leakage superimposed on the nominal T304 input noise power.
>
>I injected into the T304 a signal consisting of x-band noise plus 
>the L301 at the levels shown in the plots.  For any of the most 
>common L, S, or C-band L301 tunings, I was unable to measure any 
>gain compression in the T304.
>
>
>
>Paul Harden wrote:
>>Bob Hayward wrote:
>>>If the non-linearity is somehow coming from the LO used in the 
>>>T302 leaking into the T304 and compressing the first RF amplifier, 
>>>then switching in the Solar 20 dB Attenuator wouldn't affect the 
>>>strength of the LO birdy (since the Solar Attenuator is in front 
>>>of the T302 mixer).
>>The T302 is an upconverter using high-side LO injection.  For 
>>C-band, the LO is in the order of 16 GHz or higher.  All T302s 
>>received extensive bench testing to ensure they were compliant to 
>>specifications.   Test data sheets exist on each T302 module build.
>>The amount of LO power and birdies was one of the things tested on 
>>each IF.  I believe the specs were LO leakage in the IF must be at 
>>least 40dB below the LO power, and -70dBm minimum on the IF 
>>outputs.  The average IF output power was around -35dBm.  I do not 
>>recall ever having a problem with achieving the LO leakage/spurious 
>>specs.  There was almost scarcely a hint of the LO in the IF 
>>outputs.  That was measured with the 8-12 GHz IF filters on the 
>>T302 output (even though further downstream in the system).  It 
>>seems to cause compression problems in the T304 amplifiers, the LO 
>>leakage would have to be tens of dB above the nominal -35 or -40 dBm IF power.
>>Still, it may be a good idea to look at the C-band inputs to the 
>>T304 to ensure nothing has changed, or we don't have LO leaking in 
>>from some other source (bad cable or something).
>>Eric Chavez, who now takes care of the T302s, can better answer 
>>what these power levels are.
>>Paul Harden
>>
>>_______________________________________________
>>evlatests mailing list
>>evlatests at listmgr.cv.nrao.edu
>>http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests
>
>--
>Keith Morris
>National Radio Astronomy Observatory
>1003 Lopezville Rd.
>Socorro, NM 87801
>575-835-7060 (phone)
>575-835-7027 (fax)
>
>
>
>
>_______________________________________________
>evlatests mailing list
>evlatests at listmgr.cv.nrao.edu
>http://listmgr.cv.nrao.edu/mailman/listinfo/evlatests