0

MAX4466 power supply noise
Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.

MAX4466 power supply noise

by ogored on Thu Jul 25, 2019 2:32 pm

Hello,

I purchased the Electret Microphone Amplifier - MAX4466 with Adjustable Gain and have noticed some significant noise in the output signal.

The setup: I am using the Feather WICED board to send audio signals (analog samples) every 700 data points wirelessly over WiFi at a sampling rate of 15,260 samples per second (i.e., each 700 data point buffer fires off every 0.046 seconds).

Please see attached PDF for figures. I tried adding a 10uF and 0.1uF from the positive supply to ground on the MAX4466 but it made no difference. I can tell this is noise on the power lines, but what is a good solution? This is part of a wearable so adding separate battery source is inconvenient.
Attachments
audio.pdf
(629.04 KiB) Downloaded 11 times

ogored
 
Posts: 7
Joined: Sun Jul 22, 2018 2:42 pm

Re: MAX4466 power supply noise

by adafruit_support_mike on Fri Jul 26, 2019 12:03 am

The noise in the output is roughly in sync with the 0.046s burst rate you mentioned.

The noise while the MAX4466 is powered from the Feather could be related to power spikes caused by the Wifi radio, but it looks like you might be getting some EMF noise too.

Post a photo showing your hardware and connections and we'll take a look. 800x600 images usually work best.

adafruit_support_mike
 
Posts: 59172
Joined: Thu Feb 11, 2010 2:51 pm

Re: MAX4466 power supply noise

by ogored on Fri Jul 26, 2019 11:26 am

Here is a photo of the setup, the only difference is I am using a pair of twisted wire that came from headphones (cut to about 4 inches in length) as the wires connecting the mic to the feather board to help reduce EMI on the power lines.
Screen Shot 2019-07-26 at 10.16.27 AM.png
Screen Shot 2019-07-26 at 10.16.27 AM.png (229.09 KiB) Viewed 121 times

ogored
 
Posts: 7
Joined: Sun Jul 22, 2018 2:42 pm

Re: MAX4466 power supply noise

by adafruit_support_mike on Sat Jul 27, 2019 12:33 am

Post a photo of the physical hardware please. None of the things that cause line noise get represented in schematics.

adafruit_support_mike
 
Posts: 59172
Joined: Thu Feb 11, 2010 2:51 pm

Re: MAX4466 power supply noise

by ogored on Mon Jul 29, 2019 5:10 pm

Here are photos of the physical hardware:
IMG_6720.jpg
IMG_6720.jpg (138.92 KiB) Viewed 89 times

IMG_6721.jpg
IMG_6721.jpg (145.98 KiB) Viewed 89 times

ogored
 
Posts: 7
Joined: Sun Jul 22, 2018 2:42 pm

Re: MAX4466 power supply noise

by adafruit_support_mike on Tue Jul 30, 2019 1:03 pm

Thank you.

Try putting a medium-sized capacitor (100uF or so) between the 3V and GND pins on the MAX4466 breakout. That should stabilize power to the mic when the Wifi radio is drawing heavy current.

adafruit_support_mike
 
Posts: 59172
Joined: Thu Feb 11, 2010 2:51 pm

Re: MAX4466 power supply noise

by ogored on Tue Jul 30, 2019 3:25 pm

Okay thank you!

Update:
I tried using a 220uF because it’s what I had on hand and there was no change in behavior.

ogored
 
Posts: 7
Joined: Sun Jul 22, 2018 2:42 pm

Re: MAX4466 power supply noise

by adafruit_support_mike on Wed Jul 31, 2019 12:19 pm

In that case you’re probably seeing EMF interference from the Wifi radio.

The whole point of radio is to generate measurable current in wire with no direct connection to the radio, and Wifi radios are easily strong enough to create noise in wiring near them.

The first and easiest way to reduce EMF interference is to use twisted-pair cable. All the wires in a twisted cable are helices that share the same axis, so they all respond the external EMF the same way. The signals in the wiring are measured relative to each other, so if you add the same offset to all of them, the offsets cancel: (A+x)-(B+x)=A-B for any value of x.

Twister pairs aren’t perfect though, so they may not cancel EMF noise perfectly. If you need more isolation, you’ll need to run your wiring through a conductive shield connected to the GND line.

External EMF creates an electric field that pulls electrons to the edges of a perfect conductor. If you have a block of perfect conductor, the electrons will keep moving as long as they can, meaning they’ll stop in a thin layer along the surface. That layer of displaced electrons creates a new electric field that points the opposite way from the EMF field. Electrons will keep moving toward the surface as long as there’s enough un-cancelled EMF field to pull them, so the field created by the displaced electrons will end up being exactly equal and opposite to the external field. The result is that the EMF field has no effect inside the block of perfect conductor.

The same thing happens in you cut a hole out of the center of the block. The electrons prefer to move along the surface anyway, so you can extend the principle to a thin conductive shell around an empty space. It’s known as a Faraday Shield.

adafruit_support_mike
 
Posts: 59172
Joined: Thu Feb 11, 2010 2:51 pm

Re: MAX4466 power supply noise

by ogored on Thu Aug 01, 2019 5:21 pm

Thank you for your input. I have also tried using a coaxial cable as you mentioned but it was also in vain. I should note for whoever is interested that I switched to using the Electret Microphone Amplifier - MAX9814 with Auto Gain Control breakout and the problem is almost completely eliminated. There’s a little RF noise but the amp has far better power supply noise rejection.

ogored
 
Posts: 7
Joined: Sun Jul 22, 2018 2:42 pm

Please be positive and constructive with your questions and comments.