Signal on Electret-Mic.

Thread Starter

Pepe333

Joined Feb 12, 2018
86
Hi all!,
I'm making a sound trigger, using an electret-mic. When I see the signal from the mic, on an oscilloscope, sometimes the signal startsmoving up and sometimes starts moving down.

I thought that the sound waves, when they hit the microphone, would make the wave always start moving upwards.

What would be the explanation?
 

BillB3857

Joined Feb 28, 2009
2,570
It alsodepends upon the direction of the "hit"whether the movement of the diaphragm moves closer or further from the base element. If the it moves closer when you "hit:" it. If closer, the capacitance decreases and the output voltage will go down, assuming the resistor is tied positive and the microphone is tied to ground. Likewise, if the diaphragm moves further from from the base element, the voltage will go down.
 

Thread Starter

Pepe333

Joined Feb 12, 2018
86
It alsodepends upon the direction of the "hit"whether the movement of the diaphragm moves closer or further from the base element. If the it moves closer when you "hit:" it. If closer, the capacitance decreases and the output voltage will go down, assuming the resistor is tied positive and the microphone is tied to ground. Likewise, if the diaphragm moves further from from the base element, the voltage will go down.
Yes, but when a sound wave hit the diaphram, it will start to move always on same direction, isnt?
I tested than with low frecuencies sounds it start one way, and with high frecuencies it start the other way (not always, but often).
 

Thread Starter

Pepe333

Joined Feb 12, 2018
86
I re-read your original post. A "sound trigger" is a little different than an "impact trigger". Which are you making?
A camera trigger (for take pictures). The question is at first I used a comparator to detect a trigger level, but if the sound starts moving down, I lost a lot of time to reach the postive part of the wave (some ms).
Now I'm doing with 2 comparators in window mode and it works fine and fast, but I dont understand why happens this.
 

Ya’akov

Joined Jan 27, 2019
9,069
Yes, but when a sound wave hit the diaphram, it will start to move always on same direction, isnt?
I tested than with low frecuencies sounds it start one way, and with high frecuencies it start the other way (not always, but often).
Sound is caused by the compression and rarefaction of a medium like air. The phase of the sound that first moves the diaphragm could be either, one would be positive and the other negative.

At lower frequencies, you are probably seeing the direct sound first but at higher frequencies, you may well seeing a reflection first which could be out of phase with the source.
 
Why does it matter? AC couple the signal, buffer-amplify it, rectify it, and integrate it. You'll get a pulse out and it doesn't matter how the mic originally flexed. You're probably going to trigger on a more complex signal anyway (hence the rectify/integration) and not on a single flex of the mic diaphragm.
 

Thread Starter

Pepe333

Joined Feb 12, 2018
86
Why does it matter? AC couple the signal, buffer-amplify it, rectify it, and integrate it. You'll get a pulse out and it doesn't matter how the mic originally flexed. You're probably going to trigger on a more complex signal anyway (hence the rectify/integration) and not on a single flex of the mic diaphragm.
It matters...
When you have an action, like a exploding ballon (between300 and 500 microseconds long) you must be sure to trigger on the first part of the wave than exceed the trigger treshold. If sometimes the wave start rising, and sometimes start falling, and you dont have it in account, then you'll miss a lot of pictures.
 

MisterBill2

Joined Jan 23, 2018
18,167
My experience with electret sensor materials is that they are also a bit temperature sensitive, and at least in some of our prototypes a bright light hitting the diaphram caused an output shift. Almost everything that alters the stress on the element will generate some voltage change. So if you are going to use it for a sound trigger application you will need to set some minimum rate of change in the voltage as a lower trigger threshold . That can be as simple as a series capacitor and a shunt resistor on the amplifier input. I am presuming that you are using some sort of amplification. Then you will also need to set a trigger level so that sounds below some amplitude will not trigger it.
 

TeeKay6

Joined Apr 20, 2019
573
It matters...
When you have an action, like a exploding ballon (between300 and 500 microseconds long) you must be sure to trigger on the first part of the wave than exceed the trigger treshold. If sometimes the wave start rising, and sometimes start falling, and you dont have it in account, then you'll miss a lot of pictures.
@Pepe333
What you see on your scope depends on how you have set up triggering. If the sweep is already in progress when the "event" occurs, you will not miss anything. If the sweep has not yet started when the "event" occurs, then the delays in the scope triggering become important. Thus it matters what the trigger level is set at, what kind of trigger is used (normal, auto, one-shot, etc), and how quickly the leading edge of the "event" rises above the trigger level. If you have an amplifier between the mic and the scope, then the frequency response and delays of the amplifier also become important. There will likely be differing delay (phase) according to the frequencies of the "event" sound.
 

djsfantasi

Joined Apr 11, 2010
9,156
My experience with electret sensor materials is that they are also a bit temperature sensitive, and at least in some of our prototypes a bright light hitting the diaphram caused an output shift. Almost everything that alters the stress on the element will generate some voltage change. So if you are going to use it for a sound trigger application you will need to set some minimum rate of change in the voltage as a lower trigger threshold . That can be as simple as a series capacitor and a shunt resistor on the amplifier input. I am presuming that you are using some sort of amplification. Then you will also need to set a trigger level so that sounds below some amplitude will not trigger it.
A bit off topic, but the effect of an external action on the microphone, such as a “bright light hitting the diaphram causeding an output shift” is the basis for the demonstrated hack against personal assistants, such as “Alexa” or “Google Home”.

Someone external to your property can manipulate these devices with a laser, placing orders and making unauthorized purchases, et.al.
 

Thread Starter

Pepe333

Joined Feb 12, 2018
86
@Pepe333
What you see on your scope depends on how you have set up triggering. If the sweep is already in progress when the "event" occurs, you will not miss anything. If the sweep has not yet started when the "event" occurs, then the delays in the scope triggering become important. Thus it matters what the trigger level is set at, what kind of trigger is used (normal, auto, one-shot, etc), and how quickly the leading edge of the "event" rises above the trigger level. If you have an amplifier between the mic and the scope, then the frequency response and delays of the amplifier also become important. There will likely be differing delay (phase) according to the frequencies of the "event" sound.
Please, see the pictures attached.
 

MisterBill2

Joined Jan 23, 2018
18,167
How important is the delay between the signal crossing a limit and the first interrupt? The effective delay difference is the possible source of a problem, if it is enough to matter. Otherwise the whole exercise is chasing random ghosts.
 

Thread Starter

Pepe333

Joined Feb 12, 2018
86
How important is the delay between the signal crossing a limit and the first interrupt? The effective delay difference is the possible source of a problem, if it is enough to matter. Otherwise the whole exercise is chasing random ghosts.
The delay between the signal crossing and the interrupt launch is just some nanoseconds. But the problem was if the signal starts rising or falling. The solution is work with 2 comparators, for upper and lower limit.
 
Top