Science/Probeware experiments with a Hovercraft

Seibun Mind Development (www.seibunUSA.com) produces a model hovercraft that is sold through their network of educational dealers for less than $30.00. This hovercraft is a working model that includes two motors and floats on a cushion of air less than 1 mm thick. It is a great tool for teaching concepts of air pressure, circuits (as you build the craft) or concepts of transportation. The hovercraft can also be used in place of an air table and pucks to teach concepts of collisions and acceleration. The craft, as shown on the right is made of transparent plastic which can be painted as part of a design activity.  The craft is powered by four double "A" batteries and can fly over any smooth surface including water.  (although the craft will not float)

With the addition of sensors and probeware such as CoachLab, ULAB, or CBL students can gather data from the hovercraft and learn concepts such as acceleration, velocity, instantaneous velocity, momentum, collisions, and more. 

You can use acceleration sensors, ultrasonic position sensors, photo gates, or physical switches in order to measure values relevant to physics experiments with the Hovercraft model.  This article will detail only the use of acceleration sensors and how to fix a related problem (noise induced by the vibrations of the hovercraft motor).

The Experiment

Step One: Remove the two batteries that power the hovercraft's top propeller and rearrange and rewire the craft so that the remaining two batteries are in the innermost battery holders.  (this reduces the weight of the hovercraft and also disconnects the top propeller's drive motor which normally drives the craft forward) Step Two: Using clear tape, affix two acceleration sensors to the flat part of the hovercraft (the top).  These sensors should be at right angles to each other, one pointing to the hovercraft's front, the other pointing to the right or left side of the hovercraft.  (See image to the right).  Make sure to use a clear tape that is easy to remove so that you can later remove the sensors from the craft.  This is most important if the craft is already painted.  Also make sure not to block the main air inlet that is above the internal propeller.

Step Three: Create an activity in Coach software that can graph the data from the acceleration sensors.  Record your data for 15 to 30 seconds and display the data as "Acceleration Front to Back vs. Time" and "Acceleration Right to Left vs. Time" {see this article for more information on how to author your own activities}

Step Four: Place the hovercraft onto a large smooth surface such as a table top or the floor.  Make sure the wires leading to the acceleration sensors are not tangled and have plenty of room to move or play-out.  Turn on the hovercraft and hold it in place once it starts to float.  Start Coach recording data.  Give the hovercraft a gentle push.  Observe and review the data.

Activity Ideas:

• Use the data from Coach to identify the instantaneous velocity of the hovercraft at any given moment. 
• Identify the X and Y (i.e. front/back and right/left) acceleration/velocity components involved in the movement of the hovercraft.
• Discuss the role that friction has on this vehicle model.
• Set up an elastic and inelastic collision barriers and measure the results when the craft bumps into them.
• Have your students devise a way to measure and describe the momentum of the hovercraft.
• Add a counter-mass on the opposite side of the hovercraft to the left/right acceleration sensor.  Determine the effects of mass on the crafts acceleration and its momentum.

Filtering Data

Sometimes data sources will impose "noise" along with their signal data.  This noise might be from electrical interference such as fluorescent lights or electric motors, or the noise might be caused by other factors such as vibrations.  In this example (measuring the acceleration of the hovercraft with a low-g accelerometer) a considerable amount of noise is present.  Luckily Coach software has a tool that you can use to eliminate this noise and leave a useable approximation of the signal data that you need.

The following data was observed using measurement settings of 15 seconds and 50 data points per second.

As you can see the data does not make sense, i.e. the hovercraft is not accelerating and decelerating at a regular interval of more than one per second.  So why the bad data?  Since the accelerometer is physically mounted to the hovercraft's body, vibrations from the hovercraft's motor and propeller are transmitted to the accelerometer and show up as this regular noise. 

The first step in getting rid of this noise might be a physical step, i.e. somehow isolating the accelerometer sensors from the vibrations.  However this is not necessary with Coach Software.  You can instead "filter" the data in order to trim out the noise and leave a more stable and understandable graph (in this case acceleration vs. time.)

The first step is to change the measurement settings to include more data points per second.  We choose 350/second for our experiment.  The more data points you can record the more it is possible to filter the data without deleting too many "real" data points.

To learn more about changing measurement settings click here for a related article.

Now you can record data a second time and see what effect the change has had. (see below for the newly recorded data)

You can now see that the vibration induced in the sensor is much much greater than one per second.  As a result your students might get confused.  So it is time to filter the data.

To do this right click on the graph window (or click on the window toolbar button) and choose "Process" and then "Filter Graph"

This will launch a data filter window where you can filter the data and either add the filtered data to the existing graph, replace the existing graph, or create a new graph.

First choose the data column that you wish to filter (in this case acceleration for "Front and Back")

Next choose the filter interval (we picked 10, experiment with values on your own data)

Finally give your new data set (the filtered data) a name under the "quantity" field.  In this case we choose the default 'filtered"original column name"' and then click on "Start"

The filter window will now show the original data along with the results of the filter in a dark red color.  If your results match your needs then you can pick "add graph", "replace graph" or "new diagram" and then click "OK"

The newly filtered data compared to the original data.

You can use the filter tool after you have recorded data as in this case, or you can build the tool into an activity so that data is automatically filtered as it is recorded.  Filtering data as it is recorded can lead to a lag-time between the real world data source and the display of the data on the screen but may be worth it if you are working with younger students and want to keep the filter details away from them in order to keep from confusing them.

To filter data in real time you can add the "filter" formula as a column on a graph and then make its original data an invisible column.  This is done in the "create/edit diagram" menu.  See below.

Data Range "C2" is made "Invisible" so that the original noisy data is not shown on the diagram.

Then Data Range "C3" is assigned a formula instead of a sensor input... to do this assign the connection to "Formula" and either manually enter the formula into the "formula" box or use the formula wizard located on the right side of the window.

In this case the formula is:

Filter([g (Front-Back)];10) where "[g (Front-Back)" is the data from connection "C2" and "10" is the filter interval. 

To learn more about the create/edit diagram window click here.

To learn more about authoring activities click here.