Graduate Exam Abstract
Sam Wright
M.S. Final
January 20, 2016, 1:30 pm - 3:30 pm
Atlas Studio, Scott Bioengineering
Sub 1 fF Capacitive Sensor with PVT Compensation and Wide Dynamic Range
Abstract: High sensitivity capacitive sensors
have been of interest in many fields of
study including, but not limited to,
electronics, chemistry, biology and
medicine. A high accuracy and high
precision sensor of this type would
allow for faster, safer, and cheaper
research solutions. The research
presented targets an easy to use
device focused on label free affinity
sensing for biological target
detections in a laboratory
environment and in the field using a
mobile platform.
Many tests targeted
towards pathogen detection, including
polymerase chain reaction, require a
great amount of time and attention to
detail from a professional to produce
accurate results. Even then, additional
tests are often requested to verify the
results. Tests such as fluorescence
detection rely on the binding of
fluorescent tags bound to targets of
interest, and optically confirming the
presence of light emission from said
target. Introducing foreign bodies
such as these can cause
complications, and often result in
expensive tests. Capacitive sensors
are a relatively new means of
detection with the introduction and
widespread application of CMOS
processes. Unfortunately, this
approach has not been developed
mature enough, and specifically, its
reliability needs to be further
improved.
A compact, low power, high
sensitivity capacitive sensor could be
used with micro-arrays for multi-target
detection in real time. Comprehensive
designs would allow for easy-to-use
and affordable solutions to the
previously mentioned approaches.
Devices of such nature implemented
in a nomadic platform could then be
used to detect pathogens in the food
industry, water supplies, first-aid, and
law enforcement.
The capacitive sensor
presented contains matched
oscillators whose frequency is a direct
function of the target capacitive value,
designed in two separate processes
with minor differences. Dual
oscillators allow for a single level of
calibration against a controlled
oscillator to isolate minute changes.
An additional stage of calibration can
be carried out in real time to account
for variations in temperature, and
variability introduced during
manufacturing. Combined, the dual
stage calibration provides a detection
sensitivity of less than 1 fF relative
change by eliminating the effects of
process variation, voltage, and
temperature (PVT), produced in a
matter of seconds. The first design,
completed in 180 nm, takes
advantage of temperature behaviors
of integrated resistors to compensate
for environmental changes. Active
body biasing, and digital calibration
account for active PVT correction.
The second design contains a 6.35
pF dynamic range control oscillator
capacitive adaptation can
compensate for changes in the
environment during testing to retain
the target sensitivity. Digital calibration
alone accounts for PVT
compensation. Both designs have
been implemented in silicon layout
using commercial 180nm and 600nm
CMOS processes with sub-1fF
sensitivity.
In addition to the integrated
circuit, a mobile platform was
developed to provide an easy to use
device to perform similar tasks. The
bench top sensor has the advantage
of using sensitive test equipment,
computers, and their peripherals. A
mobile device can provide a hand
held lab environment, mimicking the
likes of an at-home blood sugar meter
used by diabetics. Sensitivity, power,
and size were the focus of such a
device, in addition to ease of use that
would be expected from any
consumer product. Utilization of low
power microcontrollers and off the
shelf, high sensitivity capacitive
sensors housed in a 3D printed case
provide a foundation for a functional
prototype capable of 1.5 fF sensitivity.
Adviser: Dr. Tom Chen
Co-Adviser: Dr. Edwin Chong
Non-ECE Member: Dr. Chuck Henry, Chemistry
Member 3: N/A
Addional Members: N/A
Publications:
A PVT-Compensated Capacitive sensor with sub 1 fF Sensitivity: MWSCAS 2015
Program of Study:
534
512
520
526
538
562
571
544
