diyAudio (
-   Analog Line Level (
-   -   Transistor Preamp Design for ADC (

ksubieta 2nd October 2012 04:46 PM

Transistor Preamp Design for ADC
I am designing a preamp to input into an ADC with a reference voltage of 2.5V (0 - 5V allowable swing) with these design parameters:

Design parameters:

12V power supply
2.76V/V gain, 8.87 dB.
20 Hz = High pass 3 dB roll off frequency
20k Hz = Low pass 3 dB roll off frequency
DC bias at 2.5V
Max input voltage: 900 mV (1.8 Vpp)
Min input voltage: 20 mV
Max output voltage: 2.5 V (5 Vpp voltage swing)

You can find the schematics here, it has all the different schematics I have tried, with the most recent one at the bottom of the page:

If the above link does not work, copy and paste the URL:

Here are my questions:

I am trying to get a 2.5V DC bias. I have done it using a voltage divider (see most recent schematic). Works fine in simulation, but is unstable when trying to hook this up to the ADC/DAC circuit. Using an adjustable voltage regulator (LM317) in its place has led to too much biasing, and it starting to affect other parts of the circuit causing clipping at 1.4V, and a decrease in the overall gain. Also, there is not enough current for the output to drive a 32 ohm speaker, so I am using an OPamp 5523N on the output stage, wired up as a voltage follower with a gain of 1, and that is being used to drive the speaker. But I am trying to make this out of transistors, so using am op amp feels like cheating, and I want to know how to make a preamp that can drive the speaker using only transistors.

1. Is there a better more stable way to DC bias the output to 2.5V then putting DC on the output right after C3 (see most recent schematic)?

2. How can I increase the current of the output so it can drive a 32 ohm speaker?

davada 18th October 2012 03:54 AM

Hi Krystina,

I take it you are being marked on this and the design parameter where handed to you.

Is this correct?

I not a big fan of capacitive coupling in amplifier stages. It seem like a waste.
I would direct couple and bias the second stage with the output of the first.

I could work this all out with pen and paper but it would be a lot easier and faster just to do it in spice. Are you using LT?

Maybe you could post the spice files.


davada 18th October 2012 03:59 AM

Hi Again,

A 32 ohm load is bit to much for such a small transistor as a 2n2222.
I think you need to go to medium power transistor at the output.
A push pull topology would be even better. Is there a reason you're using a single ended power supply?
You might also consider using a darlington configuration at the output to get the current drive up.

ksubieta 18th October 2012 05:25 AM

2 Attachment(s)
Since you are curious, I am doing this for my senior design project, and the design parameters are specified due to the ADCs capabilities.

These design parameters were decided upon because the ADC I am inputting into has a span of 0 - 5V, and the max signal that my electric guitar is outputting on the oscope is 0.9 mV. I am using a single ended power supply because I have a wall wart which supplies 12V DC, and I am not quite sure how to make a -12V supply. I have uploaded the LT spice files here, 6 is the most recent.

"I would direct couple and bias the second stage with the output of the first."

Not sure what you mean by this, unless you are just saying to remove the 10 uF cap in between the first and second stage.

I wonder if there is a better way to input into an ADC. I have been reading that some ADCs just need a cap on the input, and they bias the signal internally. Do you know anything about this?

Thanks for the reply,

ksubieta 18th October 2012 05:43 AM

Haha ok i think i know how to make a negative supply rail now: [URL=""]

davada 20th October 2012 06:11 AM

Hi Krystina,

When you joined the forum you told the moderators this was a digital project and so they directed you here. But this is really an amplifier project and there is a better place for this thread over in the solid state analog line level. Ask a moderator to move your thread over there. I think you will get a lot more attention if your in the right place.

In the mean time I think you should study this.

for the OPA1632

davada 21st October 2012 01:26 AM

1 Attachment(s)
Hi Krystina,

Have a look at this.

All times are GMT. The time now is 10:03 AM.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright 1999-2017 diyAudio

Content Relevant URLs by vBSEO 3.3.2