[SOLVED] Simple Understanding Problem with Electret Microphone..

[Qube]

hi...i have a simple doubt with Electret mic... i want to make a little microphone amplifier.. but i have a small doubt...

In the below simple circuit,the mic is connected to 9V Dc through a 10K resistor.. so 9/10K=0.0009A..

1)Should the Mic be powered in order to work or not??

2)With 10K resistor and 9V,the mic can only draw 1mA of current...is it enough for it to work??

 

[sky_123]

10k might be a bit low (it is not a critical value, so it will be fine).
Anyway, here is a circuit I created a while back, so I know it works. The values are for 12v supply, but it
will be fine for 9v slightly lower. It has a lot of gain.

 

[Qube]

10k might be a bit low (it is not a critical value, so it will be fine).
Anyway, here is a circuit I created a while back, so I know it works. The values are for 12v supply, but it
will be fine for 9v slightly lower. It has a lot of gain.

I think the Input impedance of that circuit is around 1K ,but the electret mic impedance is around 2K, do you think it will work like that??

 

[FvM]

1)Should the Mic be powered in order to work or not??
2)With 10K resistor and 9V,the mic can only draw 1mA of current...is it enough for it to work??

Both times, clearly yes. 0.5 mA is a typical current consumption.

The mic doesn't draw any current

Obviously, the schematic refers to an electrete microphone with built-in impedance converter, you won't hardly get other types on the market. I does not work without a power supply.

 

[Qube]

My another simple yet frustrating problem of mine,plz help

1)as the circuit show by Sky_123,i think the Output impedance is 2.2K(Rc),can it be connected to a LM386 AMp Ic which has Rin of 50K ?

2)and it has Input impedance of around 1K, can a Electret mic with 2K ohm impedance be connected to this circuit which has 1K input impedance?

3)Should both impedance of the mic and and the RIn of the amplifier should match or Amplifier impedance should be higher or lower than Mic impedance?

 

[sky_123]

Hi,

You'll have no problems connecting it to a 386. It will be fine. It will work with any 2-pin electret element.
It's not a hi-fi circuit (but a 386 really isn't hi-fi either), but certainly noise+distortion is reasonably low with this
preamp - so you could upgrade from the 386 if you really wanted to..
Of course, there are better circuits (and dedicated mic amps, for example with AGC if you needed it), but since you had a
single-transistor design in the first place, it should suit your needs.

 

[Qube]

Hi,

You'll have no problems connecting it to a 386. It will be fine. It will work with any 2-pin electret element.
It's not a hi-fi circuit (but a 386 really isn't hi-fi either), but certainly noise+distortion is reasonably low with this
preamp - so you could upgrade from the 386 if you really wanted to..
Of course, there are better circuits (and dedicated mic amps, for example with AGC if you needed it), but since you had a
single-transistor design in the first place, it should suit your needs.

Thanks for the reply...

i calculated the Input impedance of this circuit it and its has around 900 ohms

Ic=3mA
Rin=4K
Zin=4K||22K||2.7K||2.2K = 900 Ohms

the Electret mic i have has impedance of 2K ohms...

i think the mic amplifier should have higher impedance than the mic.. am i right ??

 

[FvM]

i think the mic amplifier should have higher impedance than the mic.. am i right ??

Not necessarily. But increasing R1, R2 and R6 won't hurt.

 

[sky_123]

I can't recall the reasoning why I picked this impedance, I don't think I even calculated it beyond knowing it
was in the right ballpark, but it worked well for my electret element (I don't have information on it, it was a usual
capsule). However, it should give acceptable results with any electret element. Or increase the impedance as FvM describes.

 

[FvM]

I would primarly consult the electrete micro datasheet. If not available, measure the actual supply current respectively select the load resistor for a few volts across the micro.

 

[Qube]

1)The C1 capacitor forms a Low pass Filter right??

2)And what about C2?

 

[KerimF]

C1 and C2 are here as bypass capacitors (to block DC) though you can see them acting as a low pass filter as well.

Edited:
Sorry... as a high pass filter

 

[Qube]

I think the R1 can be increase to 162K and R2 to 22K... with it the Zin will increase to 1.3K

The Zout is 1.5K when the pot is set to half that is 5K,2.2K||5K = 1.5K

From other source , i came to know this thing about Cin and Cout

For C1, The reactance has to less than the input impedance at the lowest frequency of interest.

For C2, The reactance has to less than the input impedance of the following stage at the lowest frequency of interest.

Why is that the Cin and Cout Reactance of the lowest frequency of interest should be less than the impedance of the Zin of amplifier or the next stage impedance??

 

[KerimF]

First, the input impedance (or resistance in this simple case) is determined mainly by Q1.
Approximately, we can assume the dynamic resistance seen from emitter (for the audio ac signal) :
Re = 26 Ohm / Ie mA
from base:
Rb = Re * hfe

The resistors R1, R2 and R3 determine Ie. Do you know how?
When we will know Rb (for an estimated hfe), the input equivalent resistance would be:
Rb//R1//R2//R6

About the equivalent output resistance:
Ro = R5_lo // (R5_hi + R4)

Second, let us talk about C1 (same for C2).
As you know its ac impedance (absolute value) could be found by:
Z (Ohm) = 1 / wC = 1 / ( 2 * PI * F * C )
PI = 3.14
F in Hz
C in F (Farad)

Z increases as F decreases. Z is close to zero for F very high and close to infinity (open circuit) as DC (F=0).
Since C1 is a bypass element, we like it pass the lowest F_lo of the electret mic.
In the circuit (for ac) C1 is between R6 and Rb//R1//R2
So if we let:
Z = R6 + Rb//R1//R2 = Rsc (in series with C1)
we can find out that the ac signal is attenuated (decreased) by a factor of K=0.707:
K = SQRT { 1 / [ 1 + (Z*Z) / (Rsc*Rsc) ] }

So if we like K be 0.95, what could be Z/Rsc?
In this case Z/Rsc = 0.1 or 1/wC = 0.1*Rsc or
C = 10/(w*Rsc)


(Please let me know if I did any mistake or I missed something you like to know)

 

[Qube]

Also do you think by increasing the Ic,which is now 3mA to 10mA will increase the gain of the amplifier??

 

[KerimF]

Usually the main point in a pre-amplifier design is to get a suitable signal (here as a voltage) that can drive a subsequent stage from a typical known source (here the mic). In other words, a better knowledge (electrically speaking) of the source and the load helps in designing the proper amplifier. For example, having a relatively high gain is as bad as having a low one and if the high/low ratio is large we may need designing a controllable amplifier.

So, in my opinion in the least, the hardest part in any design is to know (by tests and/or datasheets) the 'real' characteristics of what we will use at the input (like a signal source) and at the output (like a load or the input of another stage).

Now, increasing Ic will decrease the input resistance and R4 should also be decreased to keep Vc high enough for the AC signal to swing without clipping (due to Q1 saturation). As you will see there will be a maximum gain that could be attained by changing Ic (between low and high values). Please note that this maximum gain (here voltage gain) may not be desirable for the following amplifier (at its input and when R5 is set at its maximum).

Could we decide on the source and the load?... at least for the followings:

Source:
Its maximum voltage peak (on a typical load suggested by the mic manufacturer perhaps).
Its lowest and highest frequencies that it can generate.
Its output resistance (unless it has an open drain or collector hence acting like a current source).
Its DC output voltage (if not isolated by an internal capacitor).

Load:
Its maximum voltage peak for an acceptable distortion.
Its frequency bandwidth.
Its input resistance.
Its DC input voltage (if not isolated by an internal capacitor)

Fun has just started

Kerim

 

[D.A.(Tony)Stewart]

• Electret mic has a FET inside, so the gain is proportional to the load resistance, which also affects DC offset from V+ {but out want mic out =V+/2}
• Modern sound chips often have 0~30dB of gain and so dc offset in the amp, pop filtering, noise cancelling properties and low frequency rejection are all important.
• Single transistor gain ccts are so old , there are many simpler ways.
• You want to avoid huge electrolytics.

 

[KerimF]

You are right Sunny... On the other hand, learning electronics from discrete components brings more fun and lets choosing the appropriate chips later be more professional.

 

[D.A.(Tony)Stewart]

ALSO electret mics are known to cause hum in laptops since the AC charger is floating with very high common mode high impedance leakage across transformer.

Adding a ground makes a huge difference by absorbing high common mode fields at 50/60 Hz such as external LCD video cable to grounded monitor or hand from mic to a ground plane connected to chassis.

However laptops do not have a ground for safety reasons in case of ground faults and use near plumbing. so, "some" electret mics are worse than others. Reason is lack of balanced signal and high CM fields.. Some use CM choke to reduce this effect. Others improve quality of shield.

 

[sky_123]

Can I ask, what on earth is the purpose of this mic amp that you're making?
If it is for speech, and for connecting to an LM386 as mentioned, then I think it will "work".
I think you're really over-worrying that it may not function adequately enough with any
generic electret capsule.
That circuit has very high gain, so you may want to decrease it, on the other hand if you need the
gain, then the circuit is ok. Freq response of the amp very acceptable for an electret capsule.
I've been running it for a year now, and it works well with a generic electret element that was
then housed to be water resistant, so I needed high gain and relatively low noise, and other factors
were not a concern, so this circuit was very ideal for my needs.

If you want a hi-fi circuit, I suggest you ditch the single transistor idea and look for a
hi-fi circuit with far less distortion+noise than this circuit, pick a ready-constructed
electret (or other) mic with balanced output (it will be powered by 48v if it's electret - but it will
run on less, I use a few PP3 batteries) and pick a way better amplifier than the LM386, I think maybe
that i.c. is 20 years old or more..