Need Help! LM386 with Automatic gain from 20 to 200

Audioguru

Joined Dec 20, 2007
11,249
The LMV358 is the noisiest opamp I have ever seen ("heard").
I am glad that they removed the crossover distortion from the original LM358.
Because its supply voltage is so low then its slew rate allows output up to 80kHz. The original LM358 had trouble above only 2kHz.
 

rogs

Joined Aug 28, 2009
279
Sorry.
I have never used an audio circuit with a supply as low as 3.3V.
The only one I have used at 3.3V - as an simple audio pre-amp for an electret mic - is the LMV358, which you have advised this chap is not suitable.

Although it is clearly not a good choice for a high quality low noise mic preamp - especially for use with dynamic mics - I felt it might well suffice here?

You disagree, and all we are doing is confusing this chap. So I'll step aside, and leave further advice in your more experienced hands.

No point in confusing the OP unnecessarily!....:)
 

rogs

Joined Aug 28, 2009
279
Certainly give it a try.

I suspect Audioguru was looking at the data sheet for the Texas Instruments version of the LMV358, rather than the much better version made by Fairchild Semiconductor. (National)....
 

Audioguru

Joined Dec 20, 2007
11,249
I suspect Audioguru was looking at the data sheet for the Texas Instruments version of the LMV358, rather than the much better version made by Fairchild Semiconductor. (National)
National was bought by TI recently.
The spec's should all be the same except Fairchild got a nOOb kid to wrongly make their datasheet graph of the noise voltage.
 

Attachments

rogs

Joined Aug 28, 2009
279
As I said before, I defer to your greater experience....I simply do not have the expertise to know which manufacturers data sheets are right, and which are wrong.

I just use the devices, and comment on how I find them for a particular task, in practice.

Using a 3.3V supply for an audio project is something I've only done once before....

The LMV796 that the OP suggested earlier in the thread looks to be a lot better, noise wise, and retains the low voltage and rail to rail output capability. But I've never used one... so whether it has any other 'quirks'. I simply don't know....
 
Last edited:

Audioguru

Joined Dec 20, 2007
11,249
Read my first paragraph in my post #8 about the problems with that horrible mic preamp circuit.
An LMV796 opamp in that circuit will do almost the same.

The mic preamp opamp should be non-inverting so it has a fairly high input impedance that does not attenuate the signal from the electret mic.
 

Audioguru

Joined Dec 20, 2007
11,249
Sorry but it seems that you do not know the basics of electronics.
I will help you if you tried but you do not know enough to even begin to try.
Maybe there is a teacher here who can help you.
 

tshuck

Joined Oct 18, 2012
3,534
Audioguru

If you can't help then don't.And don't post in my theads anymore.
And you haven't help me anyway,So i don;t need you to help me.
All you do is dis poeple who need help.What help have you given me.
There more poeple on here who have help me then you ever will.

<SNIP>

I may be only 13 and starting out in this field of electronics. I do know some basics. but not in audio.

You come on here and dis everyone. So for the second time KISS MY ASS>
I haven't been a part of this thread yet, but this really angered me.

Please, please do not drag this respectable forum through the mud with your slander!

I don't care if your are 13 or 113, that is no excuse to act like a spoiled brat. People here are trying to help you and THAT is all you have to say.

If you are so aggravated by the lack of progress, perhaps you should learn these concepts on your own instead of belaying such a small attitude.

I haven't offered help because I work in digital electronics, however, Audioguru's name does him justice, he knows his stuff, so show a little respect.
 
Last edited by a moderator:

Georacer

Joined Nov 25, 2009
5,182
This is a borderline ban situation.

@Etronic
Please understand that this is a quite large forum and many different people come and go. If you see something you don't like, try to ignore it. If you see something that offends you, please report it via the
mark at the top of the post.

However, there are some members that daily spend a large amount of time trying to help other members and are backed up by knowledge, experience and consistency. These members form the core of AAC and as a result have our respect.

In your case, Audioguru might have dropped in with no help, but you answered with swears and profanity. This, automatically, puts you in the wrong side, despite of any arguments of yours.
Firstly we establish a civil environment where we can discuss and then we will hear any arguments.

This is the second time you curse other members. On the third, a ban will be issued. Be warned.
 
Last edited:

Audioguru

Joined Dec 20, 2007
11,249
The datasheet for the opamp you are using shows inportant parts that are missing in your circuit:
1) A voltage divider using RB1 and RB2 to bias the input of the opamp at half the supply voltage.
2) An input capacitor to prevent power to the electret mic from messing up the bias voltage.
3) A capacitor in series with the feedback resistor to ground to prevent the opamp from amplifying the DC bias voltage.
 

Attachments

Audioguru

Joined Dec 20, 2007
11,249
You must use a little arithmatic to select correct values of resistors and capacitors.
Your circuit has a gain of only 1 at high audio frequencies and it attenuates low audio frequencies.
 

Attachments

Audioguru

Joined Dec 20, 2007
11,249
Is 200K enough for rb1 and rb2 for the the biasing of the opamp?
I can't remember which opamp you are using. Its datasheet says what is its maximum input bias current. Use 10 times the current in RB1 and RB2.

I changed some values on the opamp.
The low frequency cutoff is created by the 1k resistor in series with the 1nF capacitor which is 160,000Hz. Then 800kHz and higher will have a maximum gain of 21 and the maximum gain at 160,000Hz is 14.9. All audio frequencies will have a maximum gain of only 1.

I showed you the simple arithmatic to calculate it.
If the value of your 1nF capacitor is 1uF then 160Hz will have a maximum gain of 14.9 and only some DTMF frequencies will not be detected.

The lowest DTMF frequency is 697Hz. Then the cutoff frequency of your opamp should be 697/5= 139.4Hz or lower. If your 1nF capacitor is 2.2uF then the cutoff frequency will be 73Hz which is fine for DTMF.
 

Audioguru

Joined Dec 20, 2007
11,249
I know? But i can only use what i have in parts draw.
These are the only values i have.
.001,3.3uF,4.7uF,2pF,10uF Ceramic.
10uF,100uF,2.2uF Aluminum electrolytic. But? I heard these are to
noisy for audio. Is this true?
I said that the circuit will work fine if you replace your 1nF capacitor with 2.2uF. You have 2.2uF so use it.
Capacitors do not produce noise.

How did you come up with the number five divide by the low group 697Hz.
The level is reduced to 0.707 times at the cutoff frequency (but then DTMF might not work) but the level is almost flat at 697Hz if the cutoff frequency is 5 times less.

Was this base on the decoders output?
No, it is the required frequency response to the input of the decoder.
DTMF is two frequencies (a low frequency and a high frequency) and both must have almost the same level. The lowest frequency is 697Hz and the capacitor must have a high enough value so it is not attenuated compared to the level of the higher frequency.

There is no information in the LMV796 datasheet to calculate for 1RB and 2RB for biasing.The max input current is 100 fempto amps or 50pA at 1 volt. At 125degrees celius.Ambient Temperature.
That is what you needed to know. The maximum input bias current of only 50pA is extremely small so the value of RB1 and RB2 can be very high. 100k is fine.
 

Ron H

Joined Apr 14, 2005
7,014
Change the coupling cap between the op amp output and the decoder input to 1uF (or 2.2uF). You should also add a 100nF cap between the vcc and ground pins on the op amp, with leads as short as possible, to prevent oscillation. Otherwise, the circuit looks OK, assuming an overall gain <210 will be adequate.
 

Ron H

Joined Apr 14, 2005
7,014
Are you sure your microphone will work with such low voltage and current? Do you have a datasheet for it?
Why do you have 3 separate symbols for ground? Are they all connected to each other?
Do you have the negative terminal of your power supply connected to ground?
 
Top