path: root/methodology.tex

\begingroup
\clearpage% Manually insert \clearpage
\let\clearpage\relax% Remove \clearpage functionality
\vspace*{-16pt}% Insert needed vertical retraction
\chapter[METHODOLOGY]{METHODOLOGY}
\endgroup

\section{General Aircraft System}
\section{Configuration One}
  \subsection{Data Acquisition}
  The turboprop engine control unit
(ECU) connects to a laptop-based program. This program outputs
various metrics of the engine and can be used to record data visually.
Unfortunately, the controller area network bus protocol used by the
engine is proprietary and the manufacturer did not provide software
to record data directly. Thus, all engine data was recorded from
video capture of real-time graphical read-out, and figures were
created during postprocessing.
The main data acquisition system consists of three Arduino
Megas. Two of the microcontrollers were utilized to record most of
the sensor inputs. The third Arduino only measures voltage and was
located inside the aircraft with the pilot. The two main boards utilize
the printed circuit boards detailed in Fig. 9. The board was designed
to accommodate the numerous sensors present on the aircraft,
though this paper will focus on the power data because most of the
recorded data was thermistor data and not particularly interesting.
All the specifics of the relevant DAQ components are listed in
Table 2.
The locations of the hall effect current sensors can be seen in the
main electrical system diagram shown in Fig. 5. The aircraft testbed
consists of 2 LEM DHAB S-133 current sensors that can read up to
750-A and are accurate to within 2% of the actual value. They are
located after the generator and before the crowbar circuit. A third
DHAB S-124 current sensor was also used before the battery, and it
can read up to 500-A accurately. This sensor is slightly more
accurate than the S-133 because it has a smaller current rating. The
amount of power expended from the battery was not expected to
exceed 300-A. These three sensors allow for current in the system to
be fully accounted for and monitored. The current sensors are
bidirectional and so can determine the direction electrical current is
flowing. All power production (battery and generator) and power
utilization (battery, wings, and crowbar) can be accurately measured
in real-time.
A voltage divider was utilized to reduce the 120-V to a 5-V range.
This scaled range was then read by the analog port on an Arduino.
For the voltage sensor, a separate dedicated Arduino was used to
allow for an increased response rate and ease of electrical isolation.
  \subsection{Experimental Procedure}
  Before hybrid-power testing
commenced, the system went through a series of preliminary tests to
reduce the technical risk of a full hybrid-power system test. These
tests were important for informing the test matrix design. First, the
turboprop engine was tested to ensure the engine was operating
nominally, before adding the generator [18]. Next, a check run was
performed using only the batteries, ensuring the power system and
electric motors worked properly. The test confirmed that the data
acquisition system and electric motors worked correctly. Before full
system testing commenced, the generator was plugged into the
power system and the engine was slowly brought up to speed to find
the point where the generator voltage exceeded the battery voltage.
During this testing, it was discovered that the generator had a k V of
27 instead of the expected value of 23.
The test procedure for the ground test rig consists of first charging
and balancing the batteries. The batteries are then installed into the
aircraft, with one battery connector left disconnected until just prior
to the test. The aircraft strut is filled with air, the engine oil level is
checked, and the brakes are checked. The aircraft is rolled out of its
hanger and brought to an open field. The aircraft is then secured with
a chain to the ground. The precharge circuit is then activated, and
batteries are plugged into the main bus. The precharge circuit is then
disabled. Two people then get into the aircraft, a pilot and a data
recorder with a laptop. A third person records a video with a fire
extinguisher on standby. The data recorder confirms that all sensors
are working. Then the engine is started and brought to idle. Once the
engine has reached operating temperature and is ready, the pilot
designates the beginning of testing and brings up the throttle to the
test point. The data recorder then brings up the electric motors to
the respective test points. Finally, after the data have been recorded,
the engine is brought down to idle and then turned off. The engine is
cooled, the main power battery is disconnected, and the data are
exported to the laptop hard drive.
The test matrix was designed to accommodate the 27-k V value
generator. If the voltage of the battery is higher than the output
voltage of the generator, then no power is generated and the
generator spins freely. This amounts to an all-electric configuration
in practice if the generator is outputting below the battery voltage.
The output voltage of the generator is around 111.1-V at the
maximum power turbine shaft speed. The minimum battery voltage
that was deemed safe is around 20% of the useful capacity.
This meant that the cell voltage of the batteries was brought down
to 3.81-V per cell. This brought the 28S battery to a total of 106.7-V,
which meant that the generator would only be operating slightly
above the battery voltage. The maximum voltage of the battery at
4.2-V per cell is 117.6-V. This has practical advantages as it makes it
difficult to overcharge the battery if the generator’s output voltage is
less than the total maximum voltage of the battery. The net effect of
the 27-k V generator was that the generator voltage would only
exceed the battery voltage at full throttle. This meant that the full
engine throttle test point was the only engine test point of interest.
The maximum charging current of the battery specified by the
manufacturer is 85-A and should not be exceeded. Based on this
value and the known resistance values of the batteries, Eq. (2) was
used to determine that a voltage difference of 7-V between the
battery and generator would be needed to exceed the maximum
current rating. However, the largest value that the generator could
output was 111.1-V and the minimum battery voltage was 106.7,
which lead to a maximum difference of 4.3-V, which is well below
7-V. In practice, because the maximum RPM is not normally
achieved under normal operation and there is a voltage drop across
the rectifier the full 111.1-V, generator output will not be achieved.
The test matrix in Table 3 was designed to operate at the full
turboprop throttle position. The test procedure was developed to
capture steady and transient data. The engine would be brought up to
idle and then brought to full throttle. The low ESC throttle value was
intended to be around 30-A per motor based on previous electric
tests, with a medium value of around 80-A, and a high value intended
to be around 150-A per motor. Both electric motors are brought up to
a low value and then held for 5-s, before moving to the next value and
holding it for 5-s as well. The result provided data at different
relatively steady conditions, as well as the transient reaction of the
electromechanical system to changes in the electrical load.
\section{Configuration Two}
  \subsection{Data Acquisition}
  \subsection{Experimental Procedure}