Balanced Audio Preamp Circuit

Thread Starter

leroyyyzz

Joined May 10, 2021
5
I tried to make a preamp circuit but I'm not getting any output at all. The simulation on LTSpice seems to work well.

The Op Amp I'm using is LMV821 datasheet
I'm using a balanced dynamic microphone
Zener Diode is 5V1.
I don't mind if its noisy as long as I get a some sort of amplification from the mic to hopefully somewhere around line level.

1630943121525.png

1630943398901.png

I'm fairly new to circuit design so any help would be appreciated.
 

AnalogKid

Joined Aug 1, 2013
10,986
I think you are starving the opamp of power. Decrease R1 to 2K, and delete R7. Also, add a 0.1 uF ceramic cap across C1, and locate it as close as possible to the U2 power and GND pins.

ak
 

Thread Starter

leroyyyzz

Joined May 10, 2021
5
Thank you for the replies, everyone.

I think you are starving the opamp of power. Decrease R1 to 2K, and delete R7. Also, add a 0.1 uF ceramic cap across C1, and locate it as close as possible to the U2 power and GND pins.

ak
I tried your suggestion, but I'm still not getting anything, but waves in the simulation on LTspice do look better when I zoom in.

hi lero,
Post your LTS asc file.
E
That's my bad, I've attached it to this reply post.

If it's a balanced microphone, why is one side of it connected to ground?
Sorry if I'm asking dumb questions, but I'm still learning about all this. I have the negative output also connected to ground as well. Would it be better to connect the negative mic input straight to the negative output without the use of ground?
 

Attachments

MrSalts

Joined Apr 2, 2020
2,767
Why not use a differential input amplifier if you have a balanced mic? Then you get a ground-referenced output. From your LM308 datasheet.

C6E63CD9-51FB-4EE4-96F1-A431ED976EC4.jpeg


Also, you have a 5.1V power supply. The datasheet says to use a minimum of +/-14v (28v) supply.

FF21EFE5-C022-4446-BD0F-4EDBAEA9B31E.jpeg
 

AnalogKid

Joined Aug 1, 2013
10,986
Why not use a differential input amplifier if you have a balanced mic? Then you get a ground-referenced output. From your LM308 datasheet.
Nice capture.
Also, you have a 5.1V power supply. The datasheet says to use a minimum of +/-14v (28v) supply.
Per post #1, he is not using a 308. My guess is that he grabbed it from the LTS library because LTS doesn't have TI parts.

ak
 

Thread Starter

leroyyyzz

Joined May 10, 2021
5
Why not use a differential input amplifier if you have a balanced mic? Then you get a ground-referenced output. From your LM308 datasheet.
Also, you have a 5.1V power supply. The datasheet says to use a minimum of +/-14v (28v) supply.
Thanks for the reply, I'll definitely do that next time. It'll be preferable if I could just edit the the schematic now as I already have the board made.

Yup, AK's got it right. He replied while I was typing this up, but I'm actually using the LMV821 op amp datasheet. I couldn't find the op amp I was using in LTspice and was using a placeholder one. I've updated the ltspice model to the correct one now and the simulation is working.
1630968232360.png1630968291342.png
 

Thread Starter

leroyyyzz

Joined May 10, 2021
5
Just some more information about the issue I'm having. I tested my circuit again and I believe my output is saturated. The output of my opamp is almost the same voltage as my Vs and giving me a DC signal.

I got these results by measuring with respect to ground using an oscilloscope.
Vs = 4.88V DC
Net (U1 to C6) = 4.84V DC
Out = 0V DC
Test = Very noisy 2.82V DC

Does anyone have any idea on why this could be happening or how to fix it? I've written the voltages that I'm getting in the circuit diagram. Thanks
1630979634835.png
 

Attachments

ericgibbs

Joined Jan 29, 2010
18,766
hi lero,
I see as you are using a single supply, you have two 10K's as divider, centre tap connected to +IN of the OPA, in order to set a mid voltage supply value, at the OPA output.

The downside is that the input impedance for most of the frequency range is only 5K, much too low for a decent pre amp.
You don't say what type 'mic' is the driving input

Also the AC analysis shows a very poor overall frequency response for the preamp.

You need to start again with a new design.

E
ESP_ 749 Sep. 07 08.23.pngESP_ 748 Sep. 07 08.22.png
 

AnalogKid

Joined Aug 1, 2013
10,986
The input impedance affects the mechanical damping of a dynamic microphone's diaphragm, which affects how well the element captures sharp acoustic transients. For some microphones, an impedance of 200 ohms is recommended.

ak
 

Ian0

Joined Aug 7, 2020
9,667
Here's choice of simple but not cheap, or cheap but not simple.
THAT1510 http://thatcorp.com/1510-1512_Audio_Preamplifier_ICs.shtml
SSM2019 https://www.analog.com/media/en/technical-documentation/data-sheets/SSM2019.pdf
INA217, https://www.ti.com/lit/ds/symlink/ina217.pdf?ts=1631024513920&ref_url=https%3A%2F%2Fwww.google.com%2F
will do the job with three resistors, but they're not cheap.

On the other hand, a couple of cheap transistors and some resistors will make a rather ordinary op-amp into rather a good low-noise preamp. It will even work with a 741 (though perhaps, an LM358 would be pushing things a little too far)
BB38CF4E-5D59-4DBF-BE58-6175CAE0DE51.jpegthere is a misprint: R8/C5 should go to 0V not V-.
BC847c will be good enough for most purposes. 2SA1312 will get very good noise performance, or if you want exceptionally low noise for low impedance microphones, 2N4403 which has a larger junction so lower base resistance will improve things even further.

One snag: all of the above need positive/negative supplies
 
Last edited:

Thread Starter

leroyyyzz

Joined May 10, 2021
5
Thank you everyone that replied to this post. I finally got it working, it was just a fried op amp. The amount of help here in this community has been incredible.

Why is Vtest 2.82 V?

ak
I went back to check and the op amp I was using must have been damaged, I must have shorted something while I was testing it. I'm ecstatic that it's finally working! Thank you for all the help, ak! It has been very helpful.

hi lero,
I see as you are using a single supply, you have two 10K's as divider, centre tap connected to +IN of the OPA, in order to set a mid voltage supply value, at the OPA output.

The downside is that the input impedance for most of the frequency range is only 5K, much too low for a decent pre amp.
You don't say what type 'mic' is the driving input

Also the AC analysis shows a very poor overall frequency response for the preamp.

You need to start again with a new design.

E
Thank you for all the information! I'll do an AC analysis in my future projects. I ended up getting it working, but yep you're right. It's pretty shoddy, but fits what I wanted to make for now. I definitely try and use a different design the next time.

Just for anyone looking at these forum posts in the future, the microphone I was using had these specs
Element: Moving coil dynamic
Polar pattern: Omni-directional
Frequency response: 300Hz to 6KHz
Sensitivity: -70dB ± 3dB (0dB = 1V/microbar at 1KHz)
Impedance: 500Ω ± 30% @ 1KHz

Here's choice of simple but not cheap, or cheap but not simple.
THAT1510 http://thatcorp.com/1510-1512_Audio_Preamplifier_ICs.shtml
SSM2019 https://www.analog.com/media/en/technical-documentation/data-sheets/SSM2019.pdf
INA217, https://www.ti.com/lit/ds/symlink/ina217.pdf?ts=1631024513920&ref_url=https%3A%2F%2Fwww.google.com%2F
will do the job with three resistors, but they're not cheap.

On the other hand, a couple of cheap transistors and some resistors will make a rather ordinary op-amp into rather a good low-noise preamp. It will even work with a 741 (though perhaps, an LM358 would be pushing things a little too far)
View attachment 247472there is a misprint: R8/C5 should go to 0V not V-.
BC847c will be good enough for most purposes. 2SA1312 will get very good noise performance, or if you want exceptionally low noise for low impedance microphones, 2N4403 which has a larger junction so lower base resistance will improve things even further.

One snag: all of the above need positive/negative supplies
Thank you, Ian. Those resources are very helpful. I'll have a look through all the options and see what I'll use in the future depending on what I have on hand.
 

Ian0

Joined Aug 7, 2020
9,667
The simple hybrid preamp should also have some electrolytic capacitors as drawn, about 10uF.AA3A784C-D059-4C04-AC47-0DB2658AF34B.jpeg
The circuit is from Self Small Signal Audio Design, so a black mark to his proof readers! I suppose that if Q1 and Q2 were perfectly matched they would be unnecessary. You can get matched pairs of BC857, called BCM857.
This circuit was in most mixing desks in the 1970s and 1980s, so if you listen to music from that era, it may well have come via this preamp.
 

AnalogKid

Joined Aug 1, 2013
10,986
The simple hybrid preamp should also have some electrolytic capacitors as drawn, about 10uF.
Not if you apply the correction in your own post #14.

The circuit is a variation of one from a National Semiconductor app note, datasheet, and series of magazine articles back in the 70's. At that time, TI was not a major player in analog circuit design, while National Semi was the absolute king of the hill.

ak
 

Ian0

Joined Aug 7, 2020
9,667
I suspect that this might have been the time TI was trying to get themselves a share of National’s market - hence the TL071. At the time there were also Analog Devices, Burr-Brown and PMI all too expensive for mere mortals!

wherever I’ve seen the circuit used it has had the electrolytics. With matched transistors and a gain of only 4 on the output stage, probably just get away without them.
 

Veracohr

Joined Jan 3, 2011
772
If you want to use lower supply current, use a larger resistor and buffer the zener with a transistor. If you use a 6.2V zener and a darlington buffer, you'll get about 5V:

Screen Shot 2021-09-12 at 10.51.16 AM.png


Also, you have a 5.1V power supply. The datasheet says to use a minimum of +/-14v (28v) supply.

View attachment 247399
That's not what that spec means. It's saying that when the supply voltage is ±15V, the input voltage range is AT LEAST ±14V. Doesn't have anything to do with the OP's question.
 
Top