Review of Wolfson Microelectronics WM7110

Table of contents

RoadTest: Wolfson Microelectronics WM7110

Author: DAB

Creation date: 16 Aug 2011

Evaluation Type: Independent Products

Did you receive all parts the manufacturer stated would be included in the package?: True

What other parts do you consider comparable to this product?: I am building a system to record and analyze thunder so I wanted to see if the MEMs technology was better than the current Electret.

What were the biggest problems encountered?: The website provided all of the documentation I needed. However, my initial test setup was incomplete, so I am building up a second one to conduct the tests.

Detailed Review:

September 17, 2011

You can read my final review in the attached word document.

To summarize, the new MEMS technology was much superior to the current Electret technology.  If you are designing a new audio device, take a good look at the Wolfston line.  They have a great product.

Road Test Report on Wolfson Microphone

by

Donald Bertke

I would like to thank Element 14 and Wolfson for providing their MEMS microphone board for testing.

Summary

The Wolfson microphone performed as specified, providing a nice flat response from 20Hz to 20KHz.  The detection sensitivity remained flat across the angle to the sound source from perpendicular to parallel (0 to 90 degrees).  The sound level detection at 1KHz was a good 10dBm better than the standard commercial Electret microphone.

I was amazed that a device this small was such an improvement over the current Electret devices in use.  The actual detection area for the MEMS device was roughly 1-mm squared compared to the Electret area of about 18-mm squared.  The MEMS device was better even though it was 18 times smaller.  No doubt about it, the MEMS technology will replace the Electret devices.  They are really superior in performance.  If they are priced close to the current devices, they should quickly replace the Electret devices in all applications.

Test Plan

After reviewing the Wolfson specifications, which were easy to find on line, I decided to compare the microphone with the current Electret microphone installed in most current commercial audio devices.

What I wanted to see was how it stacked up in Sensitivity, Bandwidth and Direction of signal reception.  To accomplish the tests, I decided to use my new UNO32 connected to my Dell laptop.

I would use identical transistor amplifiers for both microphones (I found the circuit in the new book "30 Arduino Projects for the Evil Scientist by Simon Morese.) and use the UNO32 to measure the detected values and send them to the Laptop for display.

To generate the sound, I used the software TRUERTA (www.trueaudio.com, by John L. Murphy, version 3.5.2.

Test 1

To measure the frequency bandwidth of the microphones, I put them side-by-side in front of the Laptop speaker and ran the TrueRTA software to output a signal from 20Hz to 20,000 Hz.  For each frequency, I let the UNO32 sample for several seconds and took the peak value measured from each microphone.

Test 2

To measure the angle to microphone sensitivity, I put both in the same configuration as for Test 1 and then I used a protractor to change the angle to the Speaker from 0 to 90 degrees.  As in test 1, I moved the angle and let the UNO32 sample and display the peak value over several seconds.

Test 3

To measure the sensitivity, I used the same configuration as for Test 1.  I set the software up to output a 1KHz tone and adjusted the output level to -60dbm.  I first recorded a no tone background level measurement.  I then increased the output level until I reached 0.0dbm, which was fully detectable by both microphones.

Test data

Test 1

Figure 1 graphically shows the bandwidth response across the frequencies.  Table 1 is the raw data I recorded during the test.  Watching the data as it was being measured, the Wolfson device was much quieter than the Electret.  The Electret measurements would be plus or minus 20%, where the Wolfson measurements were well within 5%.

As you can see from the data, both had a mostly flat response over the frequency range tested.

One issue I noticed was that the TrueRTA software output was not consistent from frequency to frequency, so I manually adjusted the output level so that the microphones saw the same peak to peak level for each frequency during the test.

Note, I could not figure out how to insert my excel graphs, sorry.

Figure 1, Frequency Response Test

Table 1, Frequency Peak Value Measurements

Test 2

Figure 2 shows the signal detection by angle of alignment of the microphones to the speakers.  Since I used only one Frequency, there was no need to adjust the volume as I rotated the microphone position in relation to the speaker.

Again, the Electret measurements showed a decidedly high variation of signal measurement, where as the Wolfson held to a much tighter range of values.

Note, I could not figure out how to insert my excel graphs, sorry.

Figure 2, Angle to Source Measurements

Table 2, Angle to Source Measurements

Test 3

Figure 3 shows the signal level sensitivity of both microphones at 1KHz, starting at no signal and then raising the output from -60dbm to 0.0dbm.

As in Test 1 and Test 2, the Electret showed wildly varying values during the measurements, where the Wolfson remained fairly steady.

Note, I could not figure out how to insert my excel graphs, sorry.

Figure 3, Sound Level Detection Measurements

Table 3, Sound Level Detection Measurements

Test Results

In all three tests, the Wolfson MEMS microphone proved to be superior to a standard Electret microphone.  The MEMS technology was more sensitive, had a better angle to signal detection capability and demonstrated a flat response sensitivity from 20 to 20,000 Hz.

Based upon these measurements, the MEMS technology used by Wolfson has provided designers with a superior microphone in a much smaller package than is currently available with Electret technology.

August 2011

This is just my initial update.  After getting the two microphone chips, I researched the part and decided on a test plan.

1.  Check the sensitivity of the Wolfson device against a standard Electret microphone.

2.  Compare the frequency response for each microphone.

3.  Compare the angle of incidence sensitivity of each microphone.

I first tried to build my test rig without using amplifiers, but that did not work.  Therefore I have ordered the parts to build two equal amplifiers so that I do not inadvertantly give one of the microphones an advantage.  Once I get the parts in, I will run the tests and do the analysis.

So for now, all I can say is that the MEMs are much smaller than the Electrets, so it will be interesting to see if they are as good or better.

Thanks,

DAB