U2 Anisotropy Experiment:
U2 Anisotropy Experiment:
This instrument was designed to measure the isotropy and detect the first order anisotropy in the cosmic microwave background (CMB) that was due to our motion in the Solar System's orbit around the Galaxy. The instrument was flown aboard the NASA Ames U2 jet aircraft.
Some interesting photographs:
DMR in the U2
close-up with the hatch cover and protective metal fairing removed.
This photograph shows the 54 GHz radiometer as it was set up for
testing in the laboratory before it was repackaged and placed
into the U2 apparatus.
You can see many features of a Differential Microwave Radiometer here.
There are two antennas followed by the black switch.
The antennas here are standard gain horns that feed into V-band waveguide.
The waveguide takes the CMB photons (electromagnetic waves)
gathered by the antenna into the black-colored switch.
The switch is a magnetically shielded, switchable ferrite
which we purchased from Electromagnetic Sciences.
We operated it at a switching rate of 100 Hz.
The output of the switch is connected by waveguide to an isolator.
The isolator passes the signal but does not allow signals from the
54 GHz receiver next in line get back to the switch.
The next stage is a balanced mixer with a 54 GHz local oscillator (gold-color).
The mixer down converts the signal by beating it with the 54 GHz
from the local oscillator.
Its output goes to two successive intermediate frequency (IF) amplifiers
that amplify the signal by a very large (100,000,000) factor.
That signal goes to a square law detector which converts the electic
field power in to a voltage.
After that, but not shown, is a synchronous detector that amplifies signal
that is synchronous and in phase with the 100 Hz switching of the input
switch.
The synchronous detector output is proportional to the difference in
input power received by the two antennas.
It can be calibrated directly in terms of differnce in antenna temperature.
I.e. the difference in temperature observed by the two horns.
Part of our preflight check out was to compare the temperature
of our hand to an ambient temperature load - each of which were
held in front of the horns.
If that showed roughly the right value,
we then used targets cooled with liquid nitrogen and ambient
targets to calibrate and measure the zero point of the DMR.
It was later used to check the Western Test Range
launch site for the COBE Delta at Vandenberg Air Force base.
We were concerned that there might be high levels of electromagnetic
interference that could damage the COBE DMRs at the launch site.
So we (George Smoot and Giovanni De Amici)
brought various radiometers at the COBE frequencies from our lab
to the site a month or so before the COBE equipment was to arrive
and operated them at the checkout facility and at the Delta launch tower
as well as using a spectrum analyzer to look for severe interference.
We had no other operational receiver near 53 GHz so took the this unit
from the U2 apparatus. I (George Smoot) eventually returned it to its
place in the U2 apparatus but with some modifications recorrected
by John Gibson and myself.
Peru Campaign Photographs and Information
Click here for information and photographs
of the U2 pilot and flight preparations.
George Smoot took these photographs during the campaign in Peru
so that there was ready access and difficult conditions.
References
54 GHz DMR used primarily as an atmospheric monitor.
It operated at 54 GHz (0.56 cm wavelength) where the atmospheric
emission was many times stronger than at 33 GHz (0.9 cm wavelength).
It was able to show that the U2 flew level to within one sixth of a degree
which was much more precise than necessary.
the U2 DMR can removed from the hatch.
The diameter of the metal can was set by the size of the maximum diameter
rotation bearing we could put into the U2 hatch.
The can had to be modified to have a dog-leg in it to accomodate the
parametric amplifier we used for part of the experiment.
To work the apparatus had to be snuggled into its can.
Then the closed system was lowered into the hatch very vertically.
Once we reached the dog-leg we had to slide it side ways and then
could continue lowering it.
The whole can was technically (topologically) outside of the U2,
even if it looks like it was physically inside.
There was another can that was the air seal inside the hatch
making the U2 like a limo with a swimming pool.
Our can sat in that swimming poole with a rotation bearing
mounted to the hatch.
We then had a simple cover for aerodynamics and additional RF shielding
making it look like the equipment was inside.
It was not. It was in the near vacuum of the upper atmosphere
and not an essential part of the integrity of the U2 in the
sense that the rest of the U2 skin is.
the U2 DMR instrument can removed another view.
the U2 33 GHz DMR instrument This photograph shows the 33 GHz U2
DMR in the parametric amplifier configuration.
Notice the very large form-fitting aluminum block that provided
thermal stability and thermally shorted the two signal paths (antennas).
the U2 33 GHz DMR parametric amplifier version
with the aluminum block and amplifier cover open.
One can easily see many of the tuning devices we installed on the front end.
There are triple screw tuners on each antenna and transition to waveguide
to ensure that the reflection was very low and the transmission very high.
There are also balancing screws on each arm in the black switchable
circulator. This allows us to make sure that the DMR was extremely
well balanced so that gain variations were not so significant.
This balancing was very important for the parametric amplifier
configuration since the gain stability of a parametric amplifier
is not a ideal as one would like. Later when the technology and sensitivity
of balanced mixers improved, we purchased a new wide-bandwidth, high
sensitivity mixer and retired our expensive parametric amplifier.
the U2 DMR instrument 33 GHz antennas .
This photograph shows the conical, corrugated-horn design developed
for this experiment. This technology was critical to making sure
that our signal came from the CMB and not the Earth or aircraft.
It was the pioneering use for what is nearly a standard approach now.
The COBE DMR's used this design.
U2 with experiment inside outside the hanger in preparation
for flight. Part of the preflight checkout required going outside
both to run on the U2 flight systems and to see the sky with the equipment. 
experiment in stored position
This photograph shows experiment in the stored - also take off
and landing position. Two redish plastic disks match to the surface
of the upper hatch for good aerodynamics. The disks are surrounded
with a track holding brush bristles to seal against dust and dirt
being blown into the equipment. 
experiment in observing position
The two antennas point outwards through the two holes in the upper hatch
and are essentially flush with the surface of the aircraft skin.
Through the center hole protrudes an aluminum housing
for the atmospheric monitoring radiometer. 
experiment in observing position with the upper hatch cover off.
You can see the protecting aluminum can, the antennas, and the sealing
plastic disks as well as the system for rotating the experiment.
The rotation is both for storing and observing. During observing the
antennas were interchanged about once per minute by rotating plus
and then minus 180 degrees. In this picture and the next one can
see the chain drive, microswitches, stops, etc. that were necessary
to accelerate smoothly rotate around and stop at the same precise angle
each time. 
experiment in observing position with the upper hatch cover off.
Better view of the chain drive and associated features of the rotation
system. 
DMR in the U2 with the hatch cover and protective metal fairing
removed. The metal fairing was a mechanical and electrical protection
for the equipment. One can see the two radiometers and
other associated electronics. This was very much a prototype
of two of the COBE DMR radiometers. The two radiometers worked
at 33 and 54 GHz while the COBE radiometers worked at 31.5 and 53 GHz.
The 90 GHz concept was tested in the laboratory and by balloon-borne
experiments. (see balloon-borne CMB anisotropy.)
Return to
U2 CMB Anisotropy Page
Projects Page
