diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Solid State (http://www.diyaudio.com/forums/solid-state/)
-   -   Preamp concept (http://www.diyaudio.com/forums/solid-state/64843-preamp-concept.html)

PMA 24th September 2005 10:11 AM

Preamp concept
 
2 Attachment(s)
I am thinking about building a new preamp. The concept is shown in the attached image.

bocka 24th September 2005 10:46 AM

Hi PMA,

have you thought once about a folded cascode? Your preamp can modified simply: Connect the base of your third bjt to a constant voltage. Remove the resistor from the collector of the second transitor to ground and connect the collector directly to the negative supply.

The benefits of a folded cascode are:

1. Only one amplifying stage
2. Very high cutoff frequency

I like the output stage you made. But it can be made more simple: Remove the output transistors at the right side and connect two resistors (lets say about 470 ohms) from the drivers to the output. I made something similar and found the results very promising.

darkfenriz 24th September 2005 10:52 AM

Hi Pavel
Before I can comment anything I have some questions:
1. What's the RC near diamond buffer? Does it make anthing useful?
2. How do you set low offset voltage?
3. I assume that you are aware that you need very clean rails for your LTP-VAS arrangement and CCSs too, right?

cheers and regards

PMA 24th September 2005 11:00 AM

Hi bocka and darkfenriz,

B1) The folded cascode is great, I am thinking about comparing more solutions, cutoff frequency of the circuit shown is about 6MHz
B2) I want to keep the lowest output impedance, as I have excellent sound results with it (even if compared to 50 Ohm output impedance)

D1) The RC is a kind of frequency compensation
D2) I do not take into account low DC output offset in this step, this can be solved by standard way (BTW output offset depends on currents through differential pair)
D3) Yes, I am aware of it, again it can be solved by several ways, this is just a concept, which is just able to work

darkfenriz 24th September 2005 11:28 AM

Quote:

The RC is a kind of frequency compensation
Applied along the section of voltage gain of 1....?
Is it class AB output?
On this RC there is a buffer error voltage drop, dividing by RC impedance you get parasitic error current drawn from VAS.
I think you are trying to be mysterious about the point of this arrangement... This seems to generate more distortion at higher frequencies. Or was it just a cure for non-stability on 6Mhz when strange things do happen?

I think I don't have to say that the rest of the circuit is perfect?
:D
regards

P.S. For some extend your VAS is also folded cascode.

anli 24th September 2005 01:32 PM

I'd connect pre-out stage transistors collectors to out stage transistors emiters. Such step will make output buffer input impedance significantly more linear.

darkfenriz 24th September 2005 01:47 PM

Quote:

Originally posted by anli
I'd connect pre-out stage transistors collectors to out stage transistors emiters. Such step will make output buffer input impedance significantly more linear.

isn't it simply an effect of higher bias?

anli 24th September 2005 01:53 PM

It is effect of constant Uce for pre-out transistors.

anli 24th September 2005 01:55 PM

BTW, Pavel, where are emiter resistors for pre-out pair? :)

PMA 24th September 2005 03:33 PM

Quote:

Originally posted by darkfenriz

Is it class AB output?

No, it is class A.


Quote:

Originally posted by anli
It is effect of constant Uce for pre-out transistors.
Yes, this is greatly desirable.


Quote:

Originally posted by anli
BTW, Pavel, where are emiter resistors for pre-out pair? :)
Andrej, where would you like the output transistors to travel? ;)


All times are GMT. The time now is 06:35 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2