Maker Pro
Maker Pro

Impedance matching using lm324

scarebyte

Aug 22, 2012
15
Joined
Aug 22, 2012
Messages
15
Hi,

I have an intercommunicator, that have an speaker/mic. I want to connect it in parallel and using an comutator, to a mic and speaker. For impedance fix I'm using an op-amp(lm324) in buffer configuration. I get notorious distortion, so I'm using a 6.8kohm connected to ground and out, that is ok?. The mic have 500ohm rate impedance output. The connections in the figure are ok?, how can improve?

Thanks.
Regards.
 

Attachments

  • opam.jpg
    opam.jpg
    81.5 KB · Views: 459

john monks

Mar 9, 2012
693
Joined
Mar 9, 2012
Messages
693
Normally you an electrolytic capacitor between and operational amplifier and a speaker.
An lm324 only puts out about 30 milliamperes which is next to nothing.
For this you might consider an LM380 power amplifier. This is designed for driving speakers.
Your other operational amplifier is connected backwards. Your microphone should be driving either the + or - input.

Please redraw your schematic and we will go from there.
 

scarebyte

Aug 22, 2012
15
Joined
Aug 22, 2012
Messages
15
Normally you an electrolytic capacitor between and operational amplifier and a speaker.
An lm324 only puts out about 30 milliamperes which is next to nothing.
For this you might consider an LM380 power amplifier. This is designed for driving speakers.
Your other operational amplifier is connected backwards. Your microphone should be driving either the + or - input.

Please redraw your schematic and we will go from there.

Ok, but my supply voltage is 32VDC, what opamp I can use, maybe lm384?. Why the capacitor between speaker and opamp, and what value?
Updated schematic attached.

Thanks
 

Attachments

  • 001[1].jpg
    001[1].jpg
    67.5 KB · Views: 230

john monks

Mar 9, 2012
693
Joined
Mar 9, 2012
Messages
693
You might try using a LM384 with a 10 volt zener in the power supply line.
The capacitor between the speaker and opamp is to keep the DC (direct current) off the speaker. The value should be about 100uF. You probably should put a capacitor between the microphone and the input of the opamp for the same reason. That value probably should be around 1uF.
Your next step is to put in the feedback resistors and build one.
 

scarebyte

Aug 22, 2012
15
Joined
Aug 22, 2012
Messages
15
You might try using a LM384 with a 10 volt zener in the power supply line.
The capacitor between the speaker and opamp is to keep the DC (direct current) off the speaker. The value should be about 100uF. You probably should put a capacitor between the microphone and the input of the opamp for the same reason. That value probably should be around 1uF.
Your next step is to put in the feedback resistors and build one.

Feedback resistors? But a buffer have non feedback resistors. I want a buffer to avoid the problem of non equal impedances.

Regards
 

john monks

Mar 9, 2012
693
Joined
Mar 9, 2012
Messages
693
First try without resistors. I think you will find the gain a little low but it should work.
Feedback resistors are no big deal. If this becomes an issue let us know. We can deal with it.
 

scarebyte

Aug 22, 2012
15
Joined
Aug 22, 2012
Messages
15
First try without resistors. I think you will find the gain a little low but it should work.
Feedback resistors are no big deal. If this becomes an issue let us know. We can deal with it.

Ok, I tried with the lm380 and a zener of 10v. The connection is in the attached pic. The problem is that when I connect the speaker the voltage of the zenner drops from 10V to 1.3V. The same drop happen if I use a voltage divisor.
Why happen this?

Thanks.
 

Attachments

  • foto.jpg
    foto.jpg
    72.5 KB · Views: 458
Last edited:

john monks

Mar 9, 2012
693
Joined
Mar 9, 2012
Messages
693
Two things. You might try changing R1 to 570 ohms 1/2 watt to keep your power supply from sagging.
2. You may need to place a 2.7 ohm resistor is series with a 0.1uF capacitor from you output pin to ground. Your LM380 might be oscillating.
You should disconnect the minus input pin from the output pin and connect the input pin to ground.

I hope you didn't do any damage but I just noticed that the pin out on you schematic might be wrong.
If your chip is an LM380N the pin-out is as follows:
Pin 2 is your positive input.
Pin 6 is your negative input.
Pin 7 is the power supply return line, should be connected to ground.
Pin 8 is your output pin.
Pin 14 is the power supply.

So pin 2 should be connected
 
Last edited:

scarebyte

Aug 22, 2012
15
Joined
Aug 22, 2012
Messages
15
Two things. You might try changing R1 to 570 ohms 1/2 watt to keep your power supply from sagging.
2. You may need to place a 2.7 ohm resistor is series with a 0.1uF capacitor from you output pin to ground. Your LM380 might be oscillating.
You should disconnect the minus input pin from the output pin and connect the input pin to ground.

Ok, but if I connect the input to ground, how the audio signal can pass to the output?

Regards.
 

john monks

Mar 9, 2012
693
Joined
Mar 9, 2012
Messages
693
I'm sorry but I think our messages crossed. But anyway try not connecting that pin to ground first. This will give you and inverting amplifier and this is probably what you want.

As I was looking at your schematic I noticed that your ion-out may be wrong so you probably should verify this with the datasheet. There are two versions on the datasheet I am looking at, LM380N and LM380N-8.
 

scarebyte

Aug 22, 2012
15
Joined
Aug 22, 2012
Messages
15
I'm sorry but I think our messages crossed. But anyway try not connecting that pin to ground first. This will give you and inverting amplifier and this is probably what you want.

As I was looking at your schematic I noticed that your ion-out may be wrong so you probably should verify this with the datasheet. There are two versions on the datasheet I am looking at, LM380N and LM380N-8.

Yes, im using the pins you said, no problem with that. The problem is that when i put a 330ohm resistor in series with the zener diode, it burned xD. I will try with 570 and more watt.
I have here also tl072cn and mic 386, maybe using one of them i can avoid the voltage drop in the zener?

2. You may need to place a 2.7 ohm resistor is series with a 0.1uF capacitor from you output pin to ground.

why?

Thanks
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
Your diagram shows the signal input going to - input which is connected to the output of the op-amp. There will be some fighting going on here. It is no wonder the op-amp takes excessive current.

The input should go to the + terminal and this should be biased to about half the supply voltage. Use a capacitor to pass the signal.
 

john monks

Mar 9, 2012
693
Joined
Mar 9, 2012
Messages
693
Biassing the LM380 is not the same for a usual op-amp.
I think scarebyte got his biasing right. However he might consider connecting the + terminal to ground.
Scarebyte, what type of microphone are you using?
 
Last edited:

scarebyte

Aug 22, 2012
15
Joined
Aug 22, 2012
Messages
15
Biassing the LM380 is not the same for a usual op-amp.
I think scarebyte got his biasing right. However he might consider connecting the + terminal to ground.
Scarebyte, what type of microphone are you using?

Now im testing with a pc mic, but it will be connected to this mic:
http://norsap.com/integrated communications/6487.pdf
The current opam is taking is like 4.4mA with the 6.8kohm res. I think with the 570ohm will me much more current.

Regards
 

john monks

Mar 9, 2012
693
Joined
Mar 9, 2012
Messages
693
This looks like a go.
The idilling current for the LM380 is quite low. But when you drive a speaker with some volume the current will become much greater. In fact you should place about a 100uF capacitor in parallel with the Zener to keep the voltage up during audio peaks.

Whatever you do make sure not much voltage gets on the microphone. They're rather sensitive.
 

scarebyte

Aug 22, 2012
15
Joined
Aug 22, 2012
Messages
15
This looks like a go.
The idilling current for the LM380 is quite low. But when you drive a speaker with some volume the current will become much greater. In fact you should place about a 100uF capacitor in parallel with the Zener to keep the voltage up during audio peaks.

Whatever you do make sure not much voltage gets on the microphone. They're rather sensitive.

Ok, thanks.
I simulate the circuit with proteus, and is posible achieve 13v only with 22ohm resistor, but the current is 354mA :S, is not much? . Is normal that behavior?.

Regards.
 

Attachments

  • foto.jpg
    foto.jpg
    111.8 KB · Views: 715

john monks

Mar 9, 2012
693
Joined
Mar 9, 2012
Messages
693
The problem with a 22 ohm resistor is that it will dissipate about 26 watts.
As long as you are doing a computer simulation try replacing the 22 ohm resistor with 15 volt zener and removing the 10 volt zener.
Your simulation must have a lot of audio going through the LM380.

Just noticed that you still have the - input and output connected together.
They should not be.
 
Last edited:

scarebyte

Aug 22, 2012
15
Joined
Aug 22, 2012
Messages
15
The problem with a 22 ohm resistor is that it will dissipate about 26 watts.
As long as you are doing a computer simulation try replacing the 22 ohm resistor with 15 volt zener and removing the 10 volt zener.
Your simulation must have a lot of audio going through the LM380.

Just noticed that you still have the - input and output connected together.
They should not be.

If I put only a 15v zener, is like 800mA that go through the opam xD.
So, how i can have 10V but with low current going thrugh the opam?
If I broke the wire connecting input and output the current is still high.

Thanks
 
Last edited:

john monks

Mar 9, 2012
693
Joined
Mar 9, 2012
Messages
693
I'm mystified. Can you put a separate current meter going into each LM380?
Something is drawing current and i'm trying to figure out what it is.
Can you put a voltmeter on the output of each LM380?
I would like to find out if they are being biased right.
What is the voltage across the 10 volt Zener?
 
Top