Transcript hockey--dougx - University of Maryland

Effect of Temperature on Magnetic Field Measurements
Doug
1
Hockey ,
1
Brendan Van
1
Hook ,
Ryan
2
Price
University of Maryland, 2 Joint Quantum Institute
12
Introduction
Circuit Designs and Considerations
Analysis and Conclusion
A Bose-Einstein condensate (BEC) is a collection of atoms which
occupy the lowest quantum level when cooled near absolute zero.
In order to control the quantum spin state of a BEC, it is
necessary to reach a stable resonance frequency. The magnetic
field around the BEC is used to tune the frequency. Errors in
measurements of the magnetic field will prevent stabilization.
Because many electrical components are temperaturesensitive, changes in temperature are capable of affecting the
electronics. This could change the current in the electromagnets
and thereby change the resonant frequency of the atoms.
We used the AD590, a temperature transducer which outputs
1 µA per Kelvin.
Requirements of the circuit were:
•Multiple output gains to measure the temperature at different
sensitivities.
•Single sided printed circuit board (PCB) for ease of
construction.
•A zero offset close to room temperature.
•Output as a voltage between plus and minus 10 volts.
We set up two different temperature sensors at two different
points near the electronics which control and measure the
experiment. With the data we have gathered it appears that the
temperature could be a cause of the disruption in the field.
Even though the data is not as clear as we would like it to be, a
correlation is able to be seen between temperature measuring
device and the strength of the field. While it may take more
data collection to determine if temperature is the sole factor
affecting the field, there does appear to be a trend. If that is the
case, the next step will be determining how to counteract
changes in temperature.
The trend line shows a change of about 0.34 milliGauss (mG)
per degree Celsius. This can be quite substantial as our
resonance is around 5 mG wide and one or two mG is capable
of changing the fraction between the two spin states. For scale,
the magnetic field of the Earth is around 500 mG.
Background
At our lab Doppler cooling is used to trap and cool rubidium
atoms. In Doppler Cooling, light is emitted and then absorbed by
the rubidium atoms. The excited rubidium atoms then emit a
photon in a random direction and the rubidium loses momentum.
Since light can be emitted from nearly every direction the
rubidium atom will eventually slow down until it hits it's limit of
150 microkelvin.
At this point, the atoms are trapped in an optical dipole trap at
around 1 microkelvin. Evaporation techniques then cool the
atoms further by allowing more energized atoms to escape. The
result is a collection of atoms around 100 nanokelvin which
have reached the Bose-Einstein Condensate state of matter.
𝑉𝑜𝑢𝑡
Multiple output gains were required because the level of
temperatures drifts was unknown and we needed to fit our
output in +- 10 V. We can therefore see drifts on the scale of a
few millikelvin or as large as several kelvin. The following
gains were chosen:
•1/3 K per volt for measuring very small changes in
temperature.
•5 K per volt for measuring a wider temperature range.
•1 K per volt for a middle ground between the two.
Our design was quite simple and made use of two operational
amplifiers. By varying ratio between two resistors, we are able
to amplify our signal to a desired output gain.
𝑅2
= −𝐼𝐴𝐷590 𝑅1
𝑅1
= −𝐼𝐴𝐷590 𝑅2
Figure 1: The basic design of the circuit.
Figure 2: The unpopulated circuit board.
Figure 3: The full design of the circuit board.
Sponsored by the Department of Physics, University of Maryland, College Park