Noise after Mic Preamp

[DXNewcastle]

. . . .
I use 3k.
. . . .
I use e-cap 16V and its most positive end is properly connected to the positive end of microphone.
. . . .
Just one thing that I would like to stress out here, why on earth does the preamp only works in the first 6s? Was it just like what you said? Thermal noise?

Yes, its most likely to be thermal noise, and next most likely to be RF. I'm trying to get enough accurate data from you to work out where its coming from, which is why I'm pressing you for precise facts.

You haven't yet reported back the impedance of the microphone and whether a v.smal cap between base and collector reduced the noise.

But even after correcting the transistor data, and still not knowing the impedance of the microphone, there's still a contradiction in your evidence!
If you used a 3k resistor instead of the microphone, then the positive end of C1 should have been at 1.5volts and the negaitive end should have been at 3.6volts. i.e. the positive end should have been more negative than its positive end. This can result in noise after a short period. Would you please re-test that when you have the resistor in place instead of the microphone, and again when you have the microphone in place instead of the resistor and confirm the voltages.

Do we know any more yet about the powering requirements of the microphone? eg how many milli-amps of what voltage and at what impedance?

 

[rosmawati]

impedance at the microphone input without mic in place

10MOhm

Ok, after the noise starts I take off the mic and noise is still there. So I tried hook up 1kOhm between input mic without the microphone and the noise reduce a lil bit. I increased the resistor value to 3kOhm and noise is barely heard. Problems starts when I connect the 5kOhm with microphone where the noise became louder and the voice signal was barely there.

Situation were just the same even after implementing capacitor between B and C of the transistor. I chose 100pF.

If you used a 3k resistor instead of the microphone, then the positive end of C1 should have been at 1.5volts and the negaitive end should have been at 3.6volts. i.e. the positive end should have been more negative than its positive end.

3kOhm resistor in place result:
+ve end of C1 = 1.44V
-ve end of C1 = 3.56V
I captured a highly noised signal along the voltage line. If you said this will create noise after a short period then what should I do?

Microphone in place result:
+ve end of C1 = 5.88V (voltage line has less noise)
-ve end of C1 = 3.54V (voltage line isnt smooth)

 

[DXNewcastle]

Lets look at this latest evidence in 5 parts:

1. Perhaps Capacitor C1 was maybe wrongly polarised. But you've told us that when the mic. is connected, the positive end is indeed positive. So that appears not to be the cause, though your report that with a 3k resistor instead of the microphone which polarises C1 wrongly reduces the noise suggests there is a problem in this area.

2. Perhaps the input and output impedances are wrongly matched. You've told us nothing about the input impedance of the following stage nor the gain of the following stage, so I simply can't help you with the output. You've told us very little about the impedance of the microphone, maybe 18k is appropriate for R3 and maybe not. So I can't help you with the input. (but see below).

3. Perhaps there is too much current flowing into the microphone (through R1). But you've told us the noise is present when the mic. is disconnected.

4. Perhaps there is too much (or too little) current flowing through the Transistor. You've given us 2 DIFFERENT voltages for the base. 3.5volts in post #22. 2.5volts in post #20. (I'm ignoring the error in post #15). Whichever it is, it seems to have 1milli-Amp flowing through its collector-emitter junction, which suggests that the transistor is biased 'on' very slightly. 1milli Amp is not generally enough to generate this noise.

If (and this is a BIG IF) all other components are correctly selected and correctly orientated, then there's only the possibility of a transistor problem. If the transistor was becoming too hot it would be drawing more than 1milli Amp and should have responded to 'Freeze Spray', so that just leaves the bias and the question of R3.

5. "I captured a highly noised signal along the voltage line". Where? On the 9 volt supply? If so, the battery and or Cap C5 are our problem. On the base of TR1 with microphone disconnected? If you can see the noise signal on the base of TR1 then C1 must be faulty.

Have you followed this line of reasoning? If so, you'll realise that there is still more useful evidence which you could have provided us, but with what we have, then either the base voltage is 3.5 volts and the transistor is faulty, or the base voltage is 2.5 volts and so base current can be increased by halving the values of R2 and R3.

If you can supply the information about the microphone and the next stage gain-amplifier. it might be possible to help further.

 

[rosmawati]

If you can supply the information about the microphone and the next stage gain-amplifier. it might be possible to help further.

This microphone is come along with the preamplifier design kit that I have bought. The leaflet says nothing about microphone and its impedance. It only stated the type of microphone.:sad:
For the next stage gain-amplifier, I am trying to reach the information through the GSM module manufacturer itself but until now they do not come up with any reply.

By the way, after re-troubleshooting I think the noise is not initially from the preamplifier. By following your suggestions in the earlier posts, I re-test and trying to re-confirm every results that I have got before.

1. Noise is there at amplifier output without microphone in place. So I disconnect the output amplifier from GSM module MICP1 microphone line input. I also traced noise here around 90~100kHz observing at 250ns.

2. I also observed noise at the same frequency at gsm module power supply line which is in this case 3.7V Li-Ion recharge-able battery.

3. The battery indicated no 100kHz noise signal when it was measured stand-alone ( without hooking up to the gsm module)

Do you have any idea how to reduce the power supply noise? Is it due to the RF signal?

 

[DXNewcastle]

Is there audible noise from 'the GMS module' when the mic-preamplifier is DISconnected, or only when the mic-preamplifier is connected?

Capacitor C5 should assist in keeping the 9 volt power rail stable, though some applications will require an additional small value cap in parallel (say 0.1 uF).

Are you using a 9 volt battery for the preamp (and microphone) and a 3.7 volt battery for 'the GSM module'?

What is the purpose of 'the GSM module' in your application?

 

[rosmawati]

The audible noise is already there when mic-amplifier disconnected.

Are you using a 9 volt battery for the preamp (and microphone) and a 3.7 volt battery for 'the GSM module'?

Yes.

I use gsm module in my application in order to receive and make call which is actually the most basic purpose in gsm. The mic is needed so I can talk for at least 1 metre from the mic and still be can hear at the receiving side.

 

[DXNewcastle]

The audible noise is already there when mic-amplifier disconnected.

Then the cause of the noise cannot be in the microphone preamplifier!
It can only be produced in the circuit that is connected, the GSM module.

 

[rosmawati]

Ok, after I switched from single ended input mic to differential line input, noise from gsm module has gone.

So I connected again the preamplifier circuit to the gsm module. Then noise starts again. As for now, I do not know which module that actually produces noise as they tend to be noiseless when they do not connect to each other. :???:

I guess I still stuck at the preamplifier side's problems.

---------- Post added at 09:18 ---------- Previous post was at 07:41 ----------

Dear DXNewcastle,

I just got a reply from the gsm module manufacturer. They said the module should have :


i ) Differential input MICP1 & MICN1 max voltage should be <1Vpp.
ii ) Input resistance is 50kOhm.

I hope you can help me on this matter.

 

[DXNewcastle]

The 'GSM module' has a microphone input?
Then why use a pre-amplifier? Why not connect the microphone directly to the 'microphone input' on the 'GSM module'? You'll still need to apply the powering voltage to the microphone (which the pre-amplifier supplies through R1 and blocks with C1).

 

[rosmawati]

Reason is I want to capture voice signal from at least 1 metre. If I connect the microphone directly to the microphone input at the gsm module ( with no preamp stage) , it only captures voice signal when we speak as close as we can to the microphone.

By the way, any comment on the next stage amplifier impedance that I have stated in the last post? Does it have anything to do with noise?

 

[DXNewcastle]

The input impedance of 50k will be fine with the 2k or less output impedance of the pre-amplifier.
Please check that there is no DC being provided by the GSM on its 2 input terminals (perhaps to power a microphone?).

Were you able to clarify this contradiction?

You've given us 2 DIFFERENT voltages for the base. 3.5volts in post #22. 2.5volts in post #20.

Can you also clarify this:-

So I disconnect the output amplifier from GSM module MICP1 microphone line input. I also traced noise here around 90~100kHz observing at 250ns.

2. I also observed noise at the same frequency at gsm module power supply line which is in this case 3.7V Li-Ion recharge-able battery.

3. The battery indicated no 100kHz noise signal when it was measured stand-alone ( without hooking up to the gsm module)

Did you see the noise signal on the GSM positive supply while the pre-amp is connected to the 2 microphone terminals? And it stops as soon as you disconnect the pre-amplifier? Is there noise on the GSM negaive supply when its on the positive supply? (or is the negaitive supply grounded to the 'scope?)