diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Pass Labs (http://www.diyaudio.com/forums/pass-labs/)
-   -   An idea for X-amp absolute DC offset (http://www.diyaudio.com/forums/pass-labs/37503-idea-x-amp-absolute-dc-offset.html)

lumanauw 9th July 2004 04:07 PM

An idea for X-amp absolute DC offset
 
2 Attachment(s)
Usually X-amp uses not-usual system for making steady DC offset, because both base of differential is used for input and feedback. Mr Pass sugested to use low value R (like 75ohm) at both outputs. It is R1 and R2 in my drawings.
In X-amp there is absolute DC offset and relative DC offset (that the speaker see)

Some designs uses another R in the bases of differential (R3 and R4 in my drawing) but I think this is not too effective. They will form voltage divider more than adjusting DC offset.

What do you think about using servo like in my drawing for adjusting absolute DC in X-amp? This way we dont have to use R1,R2,R3,R4.

Relative/speaker DC offset is still determined by the matching of differential transistors.

The idea is like this. The junction of emitors of differential (emitor of Q1 and Q2) can be "sliding" forced by emitor of Q3. The base of Q3 is driven by opamp servo. Q3 will slide until the servo see small DC (compared to ground) at both outputs.
This way, in the eyes of differential, Q3 is a common base transistor, that will not disturb the function of differential.

Will this can adjust DC offset in X-amp without R1,R2,R3,R4?

Nelson Pass 9th July 2004 07:17 PM

A servo is fine, but I haven't found it necessary. By loading the
output with 100 ohms or so each side to ground, we get good
absolute DC stability, and it is often supplemented with feedback
resistance to the Sources of the diff pair. In addition, more
stability is achieved by loading the + and - input (which are virtual
grounds) to ground. This reduces the tendency toward absolute
DC drift, and also helps insure absolute stability when there is
no source impedance. :cool:

lumanauw 10th July 2004 12:45 AM

Hi, Mr. Pass,
I can understand about the use of small value resistors at the output (R1 and R2 in my drawing). But I dont understand R3 and R4 at the bases of differential inputs.
If it works like you said, they works as a virtual ground (=junction of input and feedback), and can be loaded to ground to give reference when there is no signal, how much low can this R3-R4 be? Will the differential works the same if R3-R4 is 100k and if they are only 1k? They wont perform voltage divider that weakens the signal entering the bases?

arthur 10th July 2004 02:23 AM

hifizen's post have the dc servo idea:
http://www.diyaudio.com/forums/showt...520#post378520


All times are GMT. The time now is 10:24 PM.


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