[evlatests] Ku-band Polarizer problems

[Wes Grammer]

Hi Rick,

Have the cross-pol issues noted below been seen in the past (and not 
necessarily on the same antennas)? It would be helpful to know whether 
the magnitude of cross-pol has changed over time, as well as the number 
of problematic receivers.

In the EVLA Ku-band receiver chain, the most likely culprits are the 
waveguide phase shifter and OMT. As these are purely passive components, 
mechanical failures or misalignment are likely behind these issues. Some 
comments on this, below:

  * Manufacturing yield on the phase shifters wasn't that good, from
    what I gather. However, these were all individually characterized on
    the bench before installation into a receiver, weeding out any bad
    units. In addition, the axial ratio of both polarizations was also
    measured on the fully assembled receiver, though I think it may have
    been done on a limited number of frequencies across the band.
    Nevertheless, I think it's pretty unlikely these receivers had these
    issues from the beginning.

  * A weak spot in the phase shifter are the waveguide flanges, which
    are separately-machined parts brazed onto the electroform. We've
    seen some of these develop cracks over a long period after a lot of
    temperature cycling; these could cause reflections and possibly
    resonances. Fortunately these are fairly evident on inspection, and
    can be repaired.

  * The waveguide OMT's Achilles Heel is the septum, a thin piece of
    metal clamped in between the two halves of the OMT housing. Provided
    there are no gaps in the split waveguide channels, and the septum
    alignment is maintained, everything works well. However if the
    screws in the housing loosen over time and temperature cycling, a
    gap can then develop. This could create multiple problems: resonant
    spikes in the passband, added loss, and degraded cross-pol.

  * Another potential issue is the alignment of the waveguide flanges
    between the polarizer components. The most troublesome ones are the
    square waveguide interfaces, which are at three separate locations
    (at each end of the phase shifter, and at the OMT input). These are
    screwed together, but we've had past instances where they've
    loosened enough over time to shift slightly, creating resonant
    spikes and possibly degraded cross-pol as well.


-Wes


On 7/8/2026 4:44 PM, Rick Perley via evlatests wrote:
> Observations made over the past two months show some antennas have 
> problems with the Ku-band polarizers.
>
> Most of these are at the high frequency end — between 16 and 18 GHz.
> Bad antennas of note:
>
> ea03:  cross-polarization of 10 to 20% between 16.7 and 17.6 GHz.
>
> ea07:  10 to 40% between 16.2 and 18 GHz.  Also shows a resonance-like 
> bump to 40% at 14.4 GHz.
>
> ea14;  10 to 50% above 16.6 GHz.
>
> ea15;  10 to 25% above 17.2 GHz.
>
> ea02 and ea05 show a single high spike in cross-pol (to 40%) around 
> 17.8 GHz.
>
> All others are generally less than 5% over the entire band.
>
> rick
>
> _______________________________________________
> evlatests mailing list
> evlatests at listmgr.nrao.edu
> https://listmgr.nrao.edu/mailman/listinfo/evlatests
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listmgr.nrao.edu/pipermail/evlatests/attachments/20260708/7b474306/attachment-0001.html>