[Gbt3mmrx1] 3mm receiver development activity since PDR

[Mike Stennes]

Summarized here are highlights of the design and procurement activity
since October 2, 2001, on the GBT 3mm Receiver project.


Frequency Conversion:  Available Bandwidth
------------------------------------------
It was decided that the GBT 3mm receiver module 1 will have a 2-26 GHz IF,
with a somewhat limited LO tuning range of approximately 2 GHz.  While the
initial configuration of the downconverter will provide for only two, 3.2
GHz wide IF channels, we will maintain provisions for two more, and also
for the addition of future wideband spectrometers.  The current plan is to 
assign the two initial IF channels one to each beam, and to orthogonal
polarizations.


Amplitude Calibration
---------------------
Stennes and Petencin, through testing of W-band amplifiers at CDL,
confirmed that saturation would indeed occur (in the second LNA) when a
warm load is placed in front of the feed.  We are now focusing on
alternative calibration schemes, as well as the option of providing
continuum detection at IF (rather than at RF).


Biphase Modulators
------------------
Pacific Millimeter Products (PMP) was successful in producing a modulator
for 68-92 GHz.  There is a slight degradation in phase shift at
frequencies above 90 GHz, but this will not significantly affect the
performance of the receiver.  Two modulators were purchased from PMP, 
and tested at NRAO.  There were resonances found (see
http://www.gb.nrao.edu/~mstennes/S21bias004.htm), but they should not
degrade receiver performance significantly.


Vacuum Windows
--------------
Optics analysis showed that a window 3.0 inches in diameter will be large
enough to minimize truncation effects, assuming a distance of 0.75 inch
between the feed aperture and the quartz dewar window.  Discussions were
held with Dan Koller regarding costs and fabrication/test scheduling.


OMTs
----
Testing by Wes Grammer and others has shown that the ALMA W-band OMT is
well suited for this application.  We plan to purchase four units as part
of the next production run, with delivery expected by mid-July, 2002.


Cryogenic LNAs
--------------
As stated in the PDR summary, Stennes and Webber were to look at the
existing CDL amplifier designs (MAP W-band, the "broadband" version, and
the E-band prototype) and choose the one best suited for this receiver.
Recently, Marian Pospieszalski stated that there are several W-band 
designs that would work well down to 68 GHz, but recommends that the 
amplifier's input and output transitions be changed from WR-10 to WR-12 
(since all of the other components in the differencing assembly are in 
WR-12). Marian believes that the relatively small effort to redesign the 
transitions would have significant benefits in terms of receiver  
performance; by eliminating the need for waveguide tapers that could  
possibly introduce asymmetries between channels in the differencing  
assembly, we will stand a much better chance of meeting the gain and phase
balance goals.


Magic Tee
---------
Magic tees covering 68-92 GHz are not commercially available, however,
Quinstar and MRI were willing to develop this custom component.  An order
was placed with MRI (the lowest bidder) for two units.  On March 18, MRI
notified NRAO that all attempts to build these units have failed.  Stennes
reviewed MRI's test data and verified that the performance is very far
from meeting specifications.  We are now investigating other purchasing
options, as well as an in-house design (Kerr).


Other Components
----------------
Bandpass filters (68-92 GHz) were purchased and tested.  

Two isolators (68-92 GHz), designed for 300K operation, were purchased
from MRI and tested at NRAO.

A mixer assembly (downconverter) was ordered from Spacek Labs.  The
downconverter will convert the RF input (68-92 GHz) to an IF of 2-26 GHz
with 40 dB image rejection.  Delivery is expected by 5/30/02.  

AutoCad drawings for feed horn, circular-to-square waveguide transition,
and a circular-to-rectangular transition have been submitted to the NRAO
GB machine shop for fabrication.



------------------
M. Stennes, 4/1/02