Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Pre + filter + Amp - assorted questions

Status
Not open for further replies.

atferrari

Well-Known Member
Most Helpful Member
Directional microphone (output to earphones) to listen birds (no bats) in quiet areas.

Today I built and tested a preamp for an electret microphone followed by a LM386 based amplifier. Both worked fairly well.

Next step is to include in between, a tunable (variable-state) filter with Gain, Q and filter mode selection available.

My questions:

1) The three stages have gain control by design. I intend to fix the filter Gain =1.

Could / should I , also fix the gain of the the preamp? If so, to what level at the output?

2) I could see in the scope that, by yelling, the electret output can reach up to 250 mV or more. While I do not expect anyone yelling around, shouldn't I consider eventual high levels as that, that could hurt my brother's ears? This is my gift to him if I manage to build something reasonable.

Is here where compression enters the game?

3) The filter requires +/-9V supply. Does it make any sense to use it also with the preamp, taking away both resistors to polarize the noninverting input when using a single supply?

4) A preamp should it be loaded with a resistor? If so what is the value?

5) The DC blocking capacitors at the preamp and filter's output, what value should they have?

Replies and also comments, are welcome. Gracias.
 
Last edited:
Compression needed may be possible by using no more than a suitably positioned diode.
Can you post a schematic of your preamp, Agustin?
 
Compression needed may be possible by using no more than a suitably positioned diode.
Can you post a schematic of your preamp, Agustin?

It is closer than you could suspect, Alec. :) Here you have it. (Post # 2)
 
Schematic added
 

Attachments

  • Directional microphone.pdf
    24.7 KB · Views: 150
i feel myself so dump reading your guestion :D...
 
1) The three stages have gain control by design. I intend to fix the filter Gain =1.
Could / should I , also fix the gain of the the preamp? If so, to what level at the output?
Your preamp opamp will not work because its + input has no DC reference voltage. With a dual-polarity supply the DC reference voltage should be 0V. For a single positive supply the DC reference voltage is half the supply voltage.
The gain control needs to be adjusted so that loud sounds do not cause the preamp opamp to produce clipping.

2) I could see in the scope that, by yelling, the electret output can reach up to 250 mV or more. While I do not expect anyone yelling around, shouldn't I consider eventual high levels as that, that could hurt my brother's ears?
Yes, loud sounds cause hearing damage. Maybe two diodes back-to-back can clamp the level.

Is here where compression enters the game?
A compressor is used when there is nobody to turn down the volume control with music and speech. It is probably not needed here.

3) The filter requires +/-9V supply. Does it make any sense to use it also with the preamp, taking away both resistors to polarize the noninverting input when using a single supply?
ANY opamp circuit can have simple changes to its biasing so it can use only a positive supply.

4) A preamp should it be loaded with a resistor?
No, it is not needed and is not wanted.

5) The DC blocking capacitors at the preamp and filter's output, what value should they have?
The calculation for a capacitor value to pass a certain low frequency into a certain resistance is simple.
What is the lowest frequency?
 
I forgot.
It looks like you have the earphones connected in series and are probably out-of-phase instead of connected in parallel and in phase. Out-of-phase sounds weird.
 
Your preamp opamp will not work because its + input has no DC reference voltage. With a dual-polarity supply the DC reference voltage should be 0V. For a single positive supply the DC reference voltage is half the supply voltage.

A resistor from de + input to ground then, but I do not know how to calculate its value.

The gain control needs to be adjusted so that loud sounds do not cause the preamp opamp to produce clipping.
Yes, loud sounds cause hearing damage. Maybe two diodes back-to-back can clamp the level.

Could you show exactly what to try and where, AG?

ANY opamp circuit can have simple changes to its biasing so it can use only a positive supply.

It would be good to use just +9V but the filter, as is, it requires +/- 9V.

I am not sure how to polarize properly its four op amps. Maybe two 100K resistors in series, between +9 and ground and their junction to the + input?

If so, may I use just one pair of 100K resistors as above and polarize all the op amps in my circuit?


The calculation for a capacitor value to pass a certain low frequency into a certain resistance is simple.
What is the lowest frequency?

I am not sure, maybe 30 Hz? I have no experience in audio circuits, AG.

It looks like you have the earphones connected in series and are probably out-of-phase instead of connected in parallel and in phase. Out-of-phase sounds weird.

Yes, they will go in parallel. In my case, no matter what, I have just one ear in sound condition (pun intended). The left one was damaged by an infection in a voyage from Canada to Italy. In less than 48 hours I started to perceive just noise. When it is strong, it means PAIN to me. Anyway, brother goes OK.

Gracias for your help AG!
 
Last edited:
A resistor from de + input to ground then, but I do not know how to calculate its value.
Since the datasheet for the TL07x opamp shows that the Jfets on its inputs produce an input resistance of trillions of ohms then ANY resistor value will be good. If the input coupling capacitor is 100nF (0.1uF film type) then a 100k resistor will be fine for all audio frequencies.

Could you show exactly what to try and where to add clamping diodes?
See my attachment.

It would be good to use just +9V but the filter, as is, it requires +/- 9V.
No it does not. Like any other opamp circuit, it is easy to be powered from a single positive supply.
Make a voltage divider with two 22k resistors for half the positive supply (+4.5V) and filter it with a 100uF capacitor to ground. Then everything in the filter that is now grounded connects to this new +4.5V bias voltage.

I am not sure how to polarize properly its four op amps. Maybe two 100K resistors in series, between +9 and ground and their junction to the + input?
Since the input current is extremely low then the bias circuit in the previous paragraph is fine all the opamps.

I am not sure, maybe 30 Hz?
This circuit might be used for picking up distant voices so 30Hz or 40Hz is fine. There are four coupling capacitors so for the entire circuit to pass 40Hz, each one must be calculated for 10Hz.
The calculation is 1 divided by (2 x pi x R x C) where R is Ohms and C is Farads.

Yes, they will go in parallel. In my case, no matter what, I have just one ear in sound condition (pun intended). The left one was damaged by an infection in a voyage from Canada to Italy. In less than 48 hours I started to perceive just noise. When it is strong, it means PAIN to me. Anyway, brother goes OK.
You know how earphones have a 3-connections plug? If each ear is in series (without using the common ground between them) then they will be out-of-phase.
Como Estas.
 

Attachments

  • preamp.png
    preamp.png
    8.9 KB · Views: 153
I agree with AG's suggestions. The diodes begin to have a noticeable limiting effect when the opamp output amplitude is ~0.5V.
 
R17 (220 Ohms) will EASILY Protect against Too Loud of Sounds in the headphones.
 
R17 (220 Ohms) will EASILY Protect against Too Loud of Sounds in the headphones.
My headphones were 8 ohms about 52 years ago. Then R17 reduced the sound level.
But not if the headphones are 600 ohms.
 
R17 (220 Ohms) will EASILY Protect against Too Loud of Sounds in the headphones.

And will also reduce maybe too much the weaker ones. Not a complete solution I think.
 
Let me tell that had I read in due time what is posted in this thread I would probably have avoided most of my mistakes.

Reworked circuit is attached. I concluded that I can go away without a DC blocking cap between the preamp and the filter because the 5 op amps are polarized at +4,5 V.

I am starting to re build the preamp tomorrow just in case someone chimes in with additional comments.

Gracias for the replies.
 

Attachments

  • Directional microphone.pdf
    23 KB · Views: 125
Last edited by a moderator:
You removed the trimpot that adjusted the gain of the preamp. The trimpot must be set so that the loudest wanted sounds do not cause distortion from the diodes.
 
And will also reduce maybe too much the weaker ones. Not a complete solution I think.

Actually it works very good with 8 or 16 Ohm Headphones.
No real loss on Quieter sounds. (At least No Difference than compared to Turning the volume down.)

AudioGuru, Yes if 600 Ohm headphone, that resistance will need to be Increased a lot. (5K or 10K)
Actually Why would you even use an LM386 with those.
Almost any Op-amp would drive them to good Volume.
 
You removed the trimpot that adjusted the gain of the preamp. The trimpot must be set so that the loudest wanted sounds do not cause distortion from the diodes.

OK, I understand the point. :facepalm: :facepalm: Trimpot then!
 
First step done. The diodes are limiting signal at around 0,5V pk-pk. Good. Have to check but the actual Rf value seems to be is quite less than 470K.

I expect to be testing the filter during the weekend.
 
First step done. The diodes are limiting signal at around 0,5V pk-pk. Good. Have to check but the actual Rf value seems to be is quite less than 470K.
Some people shout and others whisper. Some people kiss the mic and others stay away from it.
Our hearing's sensitivity to loudness is logarithmic so reducing Rf to half (235k) reduces the loudness only a little. 47k will sound half as loud.
 
Some people shout and others whisper. Some people kiss the mic and others stay away from it.
Our hearing's sensitivity to loudness is logarithmic so reducing Rf to half (235k) reduces the loudness only a little. 47k will sound half as loud.

Hola AG,

Thanks for replying.

In line with that, I've been coinsidering to ensure that the preamp could be adjusted in situ. Myself yelling or whispering at my bench makes no sense vis a vis the actual final use. Could it be a pot then?
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top