This week the team continued finalizing the project in preparation for device testing.
The team met on Tuesday again at West campus in order to discuss implementation of the touchscreen as well as continue working on the DSP ANC program.
The team met on Wednesday to conduct some preliminary testing with the device in the testbed. They hoped to work out any issues that may arise before the more serious testing they plan to conduct on Sunday.
During this test, the team discovered that the test bed may be limited in it's usefulness for determining device efficacy. Particularly for more massive samples such as the double-paneled particle board and drywall/stud sample.
The test bed was made to be portable and has a relatively light weight construction. Thus it is more subject to resonant vibration than the more massive samples. This leads to the problem that, when the sample is firmly affixed to the end of the tube, nearly all of the sound energy is reflected back into the test bed. The sound measured at the sample is too low to allow for meaningful measurement of attenuation, and increasing the noise volume only serves to vibrate the test bed more.
However, the testbed may still be useful in the testing of the light weight door sample and will be used for this purpose on sunday.
As a solution the team has decided to simultaneously make measurements in real-world settings. John's house has ideal locations for conducting these trials.
The team met again on Thursday to test the Arduino and Touchscreen and prepare the housing.
The Team met for testing on Sunday at John's House. The team drilled a hole through the selected wall and finshed the reference microphone cable through using a long metal rod and some wire.
Once the wire was through, they connected both halves of the device and began to establish basic functionality. However, It was not long until they had to switch into troubleshooting mode as several different issues cropped up.
--
John
John continued to talk to DaveThib of Analog Devices as he develops the DSP program.
Further development of an ANC program using the MFxLMS
Another MFxLMS based program. This time with down sampling for increased frequency resolution
Hunter
This majority of the work this week was accomplished as a team as we strived to begin the test trials. There were some issues with the testbed and some additional troubleshooting required while planning and testing both the test bed and real world testing, as mentioned in the team section. With the odds seeming against the team at times, they did not give up, and the real world test trials were able to start, although there were some set backs at the start.
Any remaining time was focused on programming the microcontroller/touchscreen and, testing each program separately before applying them all together, to confirm that they function as expected incase problems arise when combing the programs, to work as one.
Below are two time-lapse videos of some programs being tested. Microphone sensitivity must be accounted for when implementing to the program. [1]
Testing program for measuring SPL-dB:
Although this test for Vrms seems invalid since rms voltage is only possible to test with AC voltage instead of DC. This "test" using a AA battery is simply to check that the calculations within the program for calculating Vrms are updating correctly to help prepare for the next meeting. Something like a signal generator that can connect directly to the analog input of the Arduino, in order do this type of calculation for an AC voltage with proper accuracy is not a device I own in home that can also provide an AC voltage that is less than the Arduino's max input voltage of 5V.
Testing program for measuring Vrms:
Hunter updated the websites weekly minutes with professor page.
References:
[1] “Jerad Lewis,” Understanding Microphone Sensitivity, [Online]. Available: https://www.analog.com/en/analog-dialogue/articles/understanding-microphone-sensitivity.html [Revised 2004].