Hi again, when I try to change number 1 and 2 in your list @Audioguru, something strange happens. The thing is that what happens is that it sounds like the sound "comes and goes", like it is amplified and then lowered again. Especially if you make a really loud sound, then just after the sound is getting higher.

Now I tried the OP-amp NE5534 instead, because its data sheet says that it should be fine with lower voltage. Would that be better?

The sound is more amplified with OP27 so I would like to use it, but it amplifies the sound so much that I do not have any headroom, it is always very close to -0 dBFS when I measure louder sound, which is why clipping occurs I guess.

If I would voltage divide one more time, so that less voltage reaches the OP-amp, would that decrease the amplification?
I'm just trying to learn.

Thank you!

 

Your original circuit had a mic current so low that the mic barely worked. Your opamp was connected wrong then maybe it also barely worked.
Now that you corrected the wiring as I showed then a loud sound severely overloads the opamp causing its coupling capacitor charge too much and cut gain until the capacitors discharge down to a normal DC voltage. While the capacitors are discharging maybe the output of the opamp has severe distortion that sounds louder than normal.

In a preamp circuit, an NE5534 produces exactly the same voltage gain and output level as an OP27 if they have a suitable power supply voltage. The feedback resistors determine the gain, not the opamp part numbers.

 

okay, what could be a solution to avoid the overload? And which capacitor is the coupling capacitor in this case?

Another thing, when I use OP27 the volume is much higher than when using NE5534, for the same circuit. Can I somehow change the gain for OP27 so that I can get some headroom when I record louder sounds?

And once again, thank you!

 

An automatic gain control (AGC) or clipping circuit can reduce the effects of signal overload.
The input and feedback capacitors can be charged when the opamp is overloaded.

With a 600 ohms load:
Some of the OP27 ICs produce output clipping if the total power supply voltage is only 10V higher than the output voltage swing.
Some of the NE5534 ICs produce output clipping if the total power supply voltage is 6V higher then the output voltage swing.
Then the NE5534 and/or a higher supply voltage produce more headroom.
Maybe your opamps are ebay fakes that do not work properly?

 

What is your circuit now? Did you increase supply voltage from 5V?

 

No, my circuit still has 5 V.

And I do not think they are ebay fakes because I bought them on a reliable site.

 

Neither OP27 nor NE5534 are designed to be operated from a single supply at 5V (too low). You are using them outside their recommended supply voltage range. Select a low voltage, "rail to rail" op-amp. What are your constraints? Supply voltage? Package?

Resistors set amplifier gain in your circuit, not an op-amp. So if the gain changes when the op-amp is changed, something is wrong.

 

okay, that's a pity, because those are the ones I have. But maybe I'll have to buy another. Is that the only solution that you can see? Or can I change the circuit somehow and still use them?
The largest problem right now with OP27 is that is amplifies to much, so I have no headroom, otherwise it works fine.
With NE5534, I can record a 94 dB " 1kHz without problem but instead the amplification is really low.

Supply voltage is 5+V and I would like to be able to record decibel levels around 100 dB without clipping.

 

What will be connected to your output? Is it a home audio amplifier? This information is necessary to determine amplification. For home audio, I would assume 100 dB (sound) -> 316 mV (at the preamplifier output). Once the amplification is set, one can calculate resistor values. Where is your final schematics?

 

Aah, I see. If you can, I would really apprechiate if you helped me with the resistor values then.

To the output a DSP card, ADAU1467, will be connected. I'm able to connect that one to its software Sigma studio. That is where I measure the sound level in dBFS, with a level detector.

My schematics have been changed so much between what I first posted and what Audioguru posted, bur right now you can assume it's very similar to the last schematics that I posted.

Thank you!

I tried to connect with OP-amp NE5534 with 9V power supply now, to see if something changed, but the problem then is that when I record loud sounds the volume first increases a lot to later go back to its regular limit.

 

Does your DSP card have any specifications, i.e. maximum audio signal input level? The clipping may occur because: (a) the preamplifier clips or (b) DSP card audio input clips. Do you know what voltage at the output you need for the 100 dB sound signal?

 

I think the minimum input voltage is 1.71 V and max is 3.63 V.

I'm not sure, but it looks like the sound level is too high when coming into the DSP and that's why the clipping occurs. Because there are not sufficient headroom.

What do you think is the reason for the change in level of sound when exposed to a louder sound?

I do not know that now, but can I calculate it? I'll also have a look in the data sheet.

 

Not knowing the numbers, you may experiment a bit. It seems you get the clipping due to too much amplification. Assuming your circuit is a modified version of the schematics in post #37, increase R2 ten times, i.e. from 100 Ohms to 1kOhm (if the value is still 100 Ohms, after modifications).

 

R3/R2 ratio gives the amplifier gain; increasing R2 10x will lower the amplification ten times. So there should be no clipping now. Is it the case?

 

yes it actually worked now, thanks a lot! I've been trying to change the value of R2 and R3, without succeeding, but now the gain is much smaller and that makes a difference, maybe I'll even try to increase R2 even more to get some more headroom. Now I measure it to -7 dB for a 94dB signal at 1kHz.

Thank you very much!

 

In post #39 I listed errors in your schematic in post #37 and I do not know if they all were done.
Please post your final schematic. We do not want it amplifying its DC input offset voltage as I said before.

Since your circuit is using a very low supply voltage then you need a clipping detector circuit and do not allow any loud sounds.