Following the advice of Professor Notash, the team met on 10/25 to conduct another round of proof of concept testing.
The experiment was conducted as an extension of last week’s test, this time using a much more massive barrier material. The team also took measurements at more positions relative to the barrier surface.
The team met at the Sound and Music facilities on East campus. The room used for last week’s test, Studio A, was not available. Testing was instead conducted in Studio B.
The test was conducted in the same manner with the same equipment as before. However, the noise floor measured in Studio B was significantly higher than what was measured in Studio A thus, the lowest measurable SPL was limited to ~44.8 dB SPL C weighted.
John
This week John worked on designing and simulating the microphone preamp circuit. Simulation of the amplifier circuit shown in the microphone datasheet (1st order HPF w/ fc=3.38Hz) gave a group delay of 214.5uS @50Hz.
In previous research, John found that a Bessel-type filter would be best suited as it preserves the shape of the passband signal.
The group delay @50Hz can be decreased by further lowering the cutoff frequency. John followed “Op Amps for Everyone” and the Application Report from Texas Instruments “Analysis of the Sallen Key Architecture” in order to design a Second Order Active HPF.
Setting Q = 0.58 (2nd order Bessel), fc= 0.85Hz, C = 2uF, and K = 2.5 yielded resistor values of 130.918k and 66.948kohm.
AC Sweep analysis of Voltage dB and Group delay was performed for both the new 2nd order and original 1st order filter. Plots are shown on the next page.
Lowering the cutoff frequency reduced the group delay at 50Hz from 214.4uS to 98.2uS.
1st Order in Green and 2nd Order in Blue
The higher gain of 2.5 corresponds to a dB Voltage magnitude of:
20*log(2.5) = 7.958dB in the passband
This is confirmed by the PSPICE simulation
Input in Blue.
1st Order in Red and 2nd Order in Green
The 0.8V DC offset is successfully removed by both circuits
Future testing will be conducted to increase the gain such that maximum output level is optimized for CODEC analog inputs
Group
Second: Proof-of-Concept
The group met again on East campus to conduct a second round of proof-of-concept testing.
This round of testing aimed to show similar results on a more massive sample material
Although the increased noise floor of the testing environment limited the lowest measurable SPL to around , the data collected showed good correlation with last week’s test.
An initial round of tests was conducted to measure each transducers frequency response when attached to the barrier. The frequency responses from 50-1000 Hz differed by an average of <1.5%.
Hunter
produced the following line graphs using the recorded data obtained after completing the test results for this round of testing.
The RMS voltage outputs of each channel of the Class D amp were also measured independently. The level difference between output channels also differed by an average of <1.5%. However, once both loads were added the difference in channel RMS voltage increased to an average of 9.2%. This also occurred in last week’s test.
Despite the discrepancy in channel voltage, good attenuation was measured from 50-300 Hz at all measurement positions.
A test with the transducers offset by 6” was also conducted. Attenuation was still achieved, albeit to a lesser degree, from 50-300Hz.
Hunter
In addition to the second Proof-of-Concept conducted this week, updates were also made to the website, and the search for components and their data sheets began.
Hunter went to SkyCraft to see their new location and if they had any new supplies that might be useful for our project, either as a back-up or in general.Some items that I was specifically looking for consisted of ADC/DAC's (>=24bit/192kHz), switch regulators, push-pull op-amps, Multiple output DC-DC power supplies, and MEMS mics(specifically the ICS-40300 that will be used in the design). I was also attempting to find a small and cheap 2.1 stereo amplifier to set up a second location at my home to use for additional proof of concept tests to experiment with different ideas.
Comments