Final Project Lab Report - Forming the Torque - Speed
Curve Given an Arbitrary Motor
Section: 3, Group: 7
Jack Boyd: 311-436
Sean Kowalik: 974-070
Eric Perkey: 835-625
MCEN 3047: Data Analysis and Experimental Methods
Date: 12/18/2017
1.0 Abstract
This project explored the use and application of an encoder, which was used in this project to
measure the rotational speed of three motors at different voltage levels, in order to recreate the
Torque - Speed curves of each respective motor while hoisting up masses. The encoder was
attached to the end of an 81:1 custom gearbox where a pulley was used to pull up increasing
masses. The motors were individually attached to the other end of the gearbox and were tested at
four different voltage levels each. This resulted in voltage measurements correlating to the
rotational speed of the motors. Current was measured using a simple multimeter. With the
speed and current measurements, the power and efficiency curves within the Torque - Speed
curves were then derived for the motors at each voltage level.
2.0 Introduction/Background Information
DC motors are commonly used in the engineering classroom. They are used in robots, model
cars, and other teaching applications in the education system. When one class uses a motor for a
project, they most likely return the motor and is re-used for future projects. After a couple years
of use, the specs of the motor are most likely lost. Some applications do not need a specific
motor, but some do need certain specifications. A group may need a high torque motor instead of
a high RPM. Every motor is made differently. Some have a higher speed while others provide
more torque. Regardless of the application, every DC motor will have a Torque-Speed curve.
Figure 1 is an example of the relationship between Speed, Torque, and Amperage. This plot was
created by inputting the no load and stall values of a known motor.
Figure 1: General Torque-Speed Curve of a Motor
With this in mind, the goal of this experiment was to be able to replicate the Torque - Speed
curve of any arbitrary DC motor. Setting up a testing apparatus that can measure the values of
speed, current, and torque allows us to plot the results as a Torque - Speed curve and allows us to
spec the motor tested. Stall torque, stall current, no load speed, and no load current are the four
most important values for a motor. The Torque - Speed curve allows us to categorize the motor
as well as show where the motor is most efficient.
1
3.0 Procedures
3.1 Voltage Levels Used
In this experiment, the Torque - Speed curves were created by holding the voltage constant.
Tests were done at 6V, 8V, 10V, and 12V. This allows us to see how, or if, the torque speed
curve changes at different voltage levels. Voltage was supplied by using a power supply at the
stations within the ITLL.
3.2 Motors Used
Three different motors were used in the experiment. One motor was checked out from Dan in
the ITLL while the other two were bought from SparkFun. The checked out motor had unknown
specs and is labeled Motor 1 throughout this report. Motor 2 and Motor 3 were bought and the
specifications were known and compared to the results from the experiment. Figure 2 shows
each motor used in this project.
Figure 2: From left to right: Motor 1, Motor 2, Motor 3
3.3 Encoder
A rotary encoder was used to measure the speed of the shaft in rotations per minute (RPM).
The encoder used in the experiment has a PPR (pulses per revolution) of 1024. This means there
exist 1024 ‘ticks’ for only one revolution of the shaft. This is very precise and allows for
accurate measurements at low speeds. The maximum RPM the encoder can measure is 6000
RPM. Because of this limitation, a gear box was made to slow down the shaft output of the
motor.
3.4 Data Acquisition
In order to read data from the encoder, a data acquisition instrument needed to be used. In this
experiment the USB-6009 National Instruments DAQ was used. These are the same DAQs given
to the group in previous lab experiments. The MATLAB data acquisition application was used to
record and save the data from the encoder.
3.5 Gearbox
In order to slow down the motor without changing the voltage input, a gearbox was made
to slow the output shaft. The motor was coupled with the lego shaft which connected to a
2
gearbox that geared down the motor 81:1. This allows for a motor spinning faster than 6000
RPM to be read successfully by the encoder.
3.6 Pulley System
To apply the torque to the shaft, a simple pulley system was used. At the end of the gear box,
a lego shaft connected to the encoder. On this shaft a circular lego piece was placed which was
able to wind up a cable that had a mass attached at the end. The radius of the circle was
measured and a weight was hung from it. This a simple way to apply a rotary torque to a shaft.
The gearbox helped slow down the motor which in turn allows enough cable to be wound up.
The data was captured over three seconds, so there had to be enough time for the data to be
captured and for the mass to not reach the end of the cable.
3.7 Testing Apparatus
The above systems were combined into one testing apparatus. The motor was connected to a
lego shaft by a coupler. The shaft entered the gearbox, increasing the ratio to 81:1. The output of
the gearbox was also a lego shaft where the pulley system was placed. The shaft continued and
was connected to the encoder shaft by another coupling. This is a full summary of the apparatus
used in the experiment. Figure 3 shows a picture of the lego apparatus and all its parts.
Figure 3: The lego testing apparatus.
3.8 Recording Process
With the testing apparatus fully functional, two values had to be recorded. Current and RPM.
The power supply supplied the constant voltage to the motor. For Motor 1, the power supply was
also used to measure current. The power supply displayed the voltage and current input. For
Motors 2 and 3 a multimeter was used instead. The difference in current was less dramatic than
current from Motor 1, so a more precise method needed to be used. The multimeters stationed at
3
the computers in the ITLL were hooked up in series in order to measure the exact current
running through the system. The RPM was measured with the encoder. Different torques were
applied by having different masses from the pulley system. The first test would start at the lowest
voltage level, 6V, and no mass was applied to the system (ignoring the mass of the cable.) The
power supply was turned on and the data acquisition tool ran for 3 seconds. The current
displayed by the power supply or multimeter was also recorded at this time. Then mass was
added to the system in increments of 100g until 1000g was reached. The process was repeated
for each voltage level and each motor.
4.0 Results and Analysis
4.1 Encoder Data
As discussed in the procedures section, the encoder was used to measure the speed of the
shaft in rotations per minute (RPM). Figure 4 shows how the encoder takes data. When the shaft
of the encoder rotates, the encoder outputs a pulse. Depending on the pulses per revolution (PPR)
of the encoder, more or less pulses will be created for one revolution. This information was used
in the MATLAB script created to count the amount of pulses recorded. The script counted each
pulse over time which plots a graph shown as Figure 5.
Figure 4 (left): Encoder output pulses over time.
Figure 5 (right): Amount of rotations over time
The slope of this line is the rotations per second. A simple conversion is used to get RPM.
The time axis window is very small in the graph. As stated before, the more PPR, the more
pulses exist. The axis is edited to be so small as to allow the viewer to visually see the pulses
over time.
4.2 Torque, Speed, and Current
4
The current was measured from a multimeter and a weight was hung from the pulley, giving
us three data values. At each increment of mass, the current and speed were measured while
keeping a constant voltage. Figures 6-8 show the scatter plot of each point at each applied torque,
for each motor at 10V.
Figures 6-8: Speed and Current plot comparison across all three motors.
Using another MATLAB script allowed us to plot linear lines formed by the current and RPM
points. With no torque applied, the values of no load speed and no load current are measured. As
more torque is applied to the motor, the slower the motor will spin. It can be seen that RPM
lowers in value as more weight was added to the pulley (a linear line with a negative slope.) It
can also be seen that as more voltage is applied, the faster the motor will spin overall. Figure 9
shows the current and speed lines at each voltage level.
Figure 9:Torque Speed curve at varying voltage levels
Eventually, the motor would stop spinning which is known as the stall torque. Current,
however, has an opposite relationship. As more torque is applied, the motor has to ‘try’ harder to
move the shaft. More power is needed so more current is supplied. We can see an increase in
current with increase in torque (a linear line with a positive slope.) At the point of stall torque,
the stall current is also known. Once these four values are known, they are inputs to the
5
MATLAB script mentioned in the introduction/background information section. The Torque -
Speed curve of each motor at 10V is shown below in Figures 10-12.
Figures 10-12: Torque - Speed Curve comparison between each motor
It can be seen that Motor 1 spun slower than Motors 2 and 3. However, Motor 1 was a truly
arbitrary motor; the specs were unknown. Motors 2 and 3 had internal gear boxes, which were
accounted for, and may have resulted in faster speeds. Motor 1 may have had an internal gear
box, but it cannot be certain. Table 1 compares the four values of interest.
Motor, 10V No Load Speed No Load Current Stall Current Stall Torque
Motor 1 4434.4 RPM 195.83 mA 731 mA 0.633 Nm
Motor 2 185,540 RPM 50.95 mA 1691.6 mA 1.23 Nm
Motor 3 18,110 RPM 37 mA 496.27 mA 0.408 Nm
Table 1: Each motor’s specifications from the experiment at 10V.
Motor 2 is shown the have the highest stall torque. This is somewhat surprising as it is the
smallest motor in size. Motor 2 spins much faster than expected. It had an internal gear box of
100:1 times a gear box made of legos of 3:1. This is what most likely led to the RPM being so
high. Different specs were seen at different voltages and these tables can be found in the
Appendix section.
5.0 Statistical Analysis
It is known that DC motor speed-torque curves and current-torque curves are linear. Because
of this, linear regression can be used to approximate a line of best fit. However, measured data
will never exactly follow this linear approximation, so the coefficient of determination (r2) needs
to be calculated to validate our measurements. Tables 2 and 3 list the values of r2 for the
speed-torque and current-torque curves respectively.
6
Speed-Torque 6 V 8 V 10 V 12 V
Motor 1 0.9307 0.8311 0.8400 0.6126
Motor 2 0.9990 0.9991 0.9978 0.9921
Motor 3 0.9999 0.9999 0.9998 0.9999
Table 2: r2 values for the speed-torque approximations
Current-Torqu 6 V 8 V 10 V 12 V
e
Motor 1 0.9472 0.8816 0.7986 0.2848
Motor 2 0.9946 0.9964 0.9950 0.9957
Motor 3 0.9956 0.9979 0.9961 0.9977
Table 3: r2 values for the current-torque approximations
By taking the mean and standard deviation from the columns of data, the spread of data can be
approximated for the measurements, given a specific voltage. Figure 13 displays this data in the
form of a bar graph with error bars. As we can see, the variation in the data increases as voltage
increases. This is likely due to the fact that the system that was built and used for testing data
would vibrate more violently as voltage increased.
Figure 13: Variance in measurements with respect to voltage
7
The mean and standard deviation can also be taken across the rows. This would show how the
measured data varies with respect to each motor. As shown in Figure 14, the data varies much
more with motor 1 than motors 2 or 3. This is likely due to the fact that more Lego gears were
used to gear down motor 1 than the other motors.
Figure 14: Variance in measurements with respect to motor
Overall, the measurements taken from motors 2 and 3 seems to be fairly precise, while motor 1
was far less reliable. This, however, does not mean that the data is accurate, as it only displays
the fact that the data taken fits a line well. In order to record more accurate data, a system made
of properly machined gears, shafts, and couplings should be used, as the efficiency of the Lego
gearbox is probably pretty low.
6.0 Conclusion
In this project, the Torque - Speed curves were created using three different motors, each at
four different voltage levels: 6V, 8V, 10V, and 12V. Each motor was hooked up to a gearbox
and pulley system to pull up incrementally increasing masses until each motor reached its stall
point or until the maximum mass available (1kg) was reached. An encoder was used to measure
how fast each motor spun in RPM while they pulled up the masses. A DAQ USB-6009 and
MATLAB data acquisition application were used to record and save the data collected from the
encoder. Current was measured using a multimeter each time a mass was hoisted up. The data
for both motor speed and current were then fitted, using linear regression, to obtain the
approximate speed and current lines in the Torque - Speed curves. From there, the power and
efficiency curves were plotted using their respective equations. At 10V, it has been shown that
Motor 2 had the highest no load speed, stall current, and stall torque of 185,540 RPM, 1631.6
8
mA, and 1.23 Nm respectively, and Motor 3 had the lowest no load current of 37 mA. A similar
trend has been observed the other voltages as seen in the Appendix.
7.0 Appendix
7.1 Spec Tables
Motor, 6V No Load Speed No Load Current Stall Current Stall Torque
Motor 1 2509 RPM 150.91 mA 521.36 mA 0.394 Nm
Motor 2 185,630 RPM 45.89 mA 1685 mA 1.231 Nm
Motor 3 18,111 RPM 25.36 mA 502.33 mA 0.408 Nm
Table 4: Each motor’s specifications from the experiment at 6V.
Motor, 8V No Load Speed No Load Current Stall Current Stall Torque
Motor 1 3406 RPM 166 mA 777.68 mA 0.532 Nm
Motor 2 185,640 RPM 50.7 mA 1542 mA 1.17 Nm
Motor 3 18,113 RPM 31.72 mA 499.71 mA 0.409 Nm
Table 5: Each motor’s specifications from the experiment at 8V.
Motor, 10V No Load Speed No Load Current Stall Current Stall Torque
Motor 1 4434.4 RPM 195.83 mA 731 mA 0.633 Nm
Motor 2 185,540 RPM 50.95 mA 1691.6 mA 1.23 Nm
Motor 3 18,110 RPM 37 mA 496.27 mA 0.408 Nm
Table 6: Each motor’s specifications from the experiment at 10V.
Motor, 12V No Load Speed No Load Current Stall Current Stall Torque
Motor 1 5336 RPM 244.36 mA 502.06 mA 0.712 Nm
Motor 2 185,620 RPM 60.41 mA 1662.8 mA 1.16 Nm
Motor 3 18,112 RPM 36.5 mA 501.93 mA 0.407 Nm
Table 7: Each motor’s specifications from the experiment at 12V.