Experiments: Tests of the Apparatus
Test Set 1: Testing for Distance Differences & Microphone Sensitivities
Test# |
Result |
Increased sensitivity |
Increased + sensitivity |
1 |
Success 5 ft |
|
|
2 |
Success 10 ft |
|
|
3 |
Success 15 ft |
|
|
4 |
Success 20 ft |
|
|
5 |
Failure 25 ft |
Success 25 ft |
|
6 |
Failure (Muffled sound test) 25 ft (Failure to detect muffled sound is a success for the glove because I did not want it to detect muffled sound.) |
Failure 25 ft (Success for glove) |
Failure 25 ft (Success for glove) |
7 |
Success (Control test- reversed inputs) |
|
|
Analysis: The glove worked to the maximum distance that we could test it in the house. At 25 feet, I had to moderately increase the sensitivity of the microphones. Increasing the sensitivity of the microphones caused a problem, though: the microphones on the glove began to pick up the sound of the music coming from Duncan's headphones. (This was the music we were using to deafen Duncan.) I could not turn the music down, because then Duncan could hear the buzzer.
During this test, Duncan and I realized that Duncan might be subconsciously picking up on my movement around the room with the buzzer, so I decided to find a way to fix that.
The glove did not detect the buzzer when it was muffled under cushions. I determined that the glove works only for louder noises. This is not really a problem, because I don't want the glove to respond to every sound, only to sounds that require attention, like horns, crying, etc.
The glove passed the control test. In that test (#7), I switched the inputs on the circuit, meaning that Duncan should point away from the sound's direction, if he was not picking up the sound of the buzzer with his ears.
Improvements:
1) I built two long buzzer circuits, which would enable me to locate buzzers in different parts of the room and control them from the couch, without walking around.
2)Using Audacity, I made a recording of 3 different sine waves, that I hoped would do a better job of covering the buzzer sound. I also bought some ear plugs.
Test Set 2: Testing the Remote Dual Buzzer System and Audacity Sine Wave Soundtrack
Test # |
Result |
A |
Pinpointed 15 ft RDBS |
B |
Pinpointed 15 ft RDBS |
Analysis: The test subject was successfully able to pinpoint the location of the buzzer, even though I was sitting on the sofa and not moving around with them. However, the subject could not tolerate the sound of the Audacity Sine Wave Soundtrack.
Improvements: I made a new recording. This time I recorded the sounds of both buzzers, and blended them together in Audacity, and added some white noise as a mask. Because the sound of the buzzers in the recording was identical to the sound of the buzzers, this meant that the sound could be played through the headphones at a lower volume. This increased the comfort for the test subjects and also eliminated the problem of the microphones picking up the sound of the music instead of the buzzer, that I had in the first test series.
Test Set 3: Testing the Remote Dual Buzzer System and Audacity White-Noise & Buzzer Soundtrack
Test # |
Result |
1 |
Invalid -- motor defective |
2 |
Invalid -- motor defective |
3 |
Invalid -- motor defective |
4 |
Invalid -- motor defective |
5 |
Invalid -- motor defective |
6 |
Invalid -- motor defective |
7 |
Invalid -- motor defective |
8 |
Pinpointed 15 ft RDBS |
9 |
Pinpointed 15 ft RDBS |
10 |
Pinpointed 15 ft RDBS |
11 |
Pinpointed 15 ft RDBS |
12 |
Pinpointed 15 ft RDBS |
13 |
Pinpointed 15 ft RDBS |
14 |
Pinpointed 15 ft RDBS |
15 |
Pinpointed 15 ft RDBS |
16 |
Failure 15 ft RDBS |
17 |
Pinpointed 15 ft RDBS |
18 |
Pinpointed 15 ft RDBS |
19 |
Pinpointed 15 ft RDBS |
Analysis: The first several tests did not work (1 to 7) but it turned out that the wiring to the vibrating motor had become loose. Because it was not possible for the glove to work with the motor disconnected, I am not including these results in my analysis.
In this series of tests, I used the distant buzzer system and the new Audacity recording. Because Duncan was hearing both buzzers in his headphones, he could not detect the sound of the buzzers in the room. I was not walking around, so he could not detect my movements. I changed the location of the buzzers between tests.
The apparatus worked 11 out of 12 times. Duncan failed to pinpoint the direction of the sound on test #16. I do not know why this happened, because the equipment was the same. It's possible that the connection to the motor was becoming loose again and that it didn't work when Duncan had his hand in a certain angle. (I had that problem sometimes early on in the design of the apparatus.) It is also possible that my family was making noise somewhere else in the house that was louder than the buzzer at that time.
Analysis For all three test sets:
Conclusion for all Three Test Sets:
The success rate of the glove was so high on all three test days that I decided not to do any more tests. That is because I designed my glove to work in an indoor setting, and it was nearly always successful indoors.
If I wanted the glove to work outside, I would have to consider more variables. The distance from the glove to a car horn could be much greater than 25 feet. As well, there could be many kinds of loud noises, such as a dog barking or pedestrians laughing, and the apparatus would have to be able to do a better job of distinguishing between levels of sound. Instead of just vibrating or not vibrating, it would have to be designed to vibrate at different intensities.