pass先生设计的关于 GC的7个基本电原理图

flyingx

pass先生设计的关于 GC的7个基本电原理图

7 Easy Pieces  Post #1  
Just when you thought it was safe to go back to the work bench....

Here's 7 easy pieces for chip amps.

<img src="attachments/dvbbs/20048196182966430.gif" border="0" onload="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onmouseover="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.style.cursor=\'hand\'; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onclick="if(!this.resized) {return true;} else {window.open(\'attachments/dvbbs/20048196182966430.gif\');}" onmousewheel="return imgzoom(this);" alt="" />
www.passlabs.com/np/GC-VAR-INV-FDBK-1a.pdf

You might find this a trivial circuit, but for those who
don't know it, it can be very important. Chip amplifiers
have enormous amounts of open loop gain (120 dB = 1,000,000)
which can create stability problems when the circuit is set
for low gain, say only 10 or 20 dB. This circuit places a
resistor to ground at the negative input, which throws away
some of the open loop gain, resulting in less feedback. In
fact you can put a variable resistor here, and tune the
amount of feedback to taste.


<img src="attachments/dvbbs/20048196212693200.gif" border="0" onload="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onmouseover="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.style.cursor=\'hand\'; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onclick="if(!this.resized) {return true;} else {window.open(\'attachments/dvbbs/20048196212693200.gif\');}" onmousewheel="return imgzoom(this);" alt="" />
www.passlabs.com/np/GC-VAR-BAL-FDBK-1a.pdf

Does the same thing for a balanced input circuit. Generally
you want the pots at equal value, but for the most perfect
common mode input rejection, separate pots allow trimming.


<img src="attachments/dvbbs/2004819623135748.gif" border="0" onload="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onmouseover="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.style.cursor=\'hand\'; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onclick="if(!this.resized) {return true;} else {window.open(\'attachments/dvbbs/2004819623135748.gif\');}" onmousewheel="return imgzoom(this);" alt="" />
www.passlabs.com/np/GC-OUTPUT-BIAS-1a.pdf

You say your chip amp needs more output bias? Here's an old
trick.


<img src="attachments/dvbbs/20048196274568699.gif" border="0" onload="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onmouseover="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.style.cursor=\'hand\'; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onclick="if(!this.resized) {return true;} else {window.open(\'attachments/dvbbs/20048196274568699.gif\');}" onmousewheel="return imgzoom(this);" alt="" />
www.passlabs.com/np/GC-OUTPUT-BIAS-2a.pdf

Here we see the same thing as GC-SS-3a, but without the
SuperSymmetry jazz, just a pair of chip amps. The voltage
sources will be low values, as will the output resistors,
and additional output stage is provided by the DC difference
in the output voltages across those output resistors. The
load output sees the split value, so there's no DC seen there.
You also get the advantage of delivering about twice the
current as a single chip amp.


<img src="attachments/dvbbs/20048196543728688.gif" border="0" onload="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onmouseover="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.style.cursor=\'hand\'; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onclick="if(!this.resized) {return true;} else {window.open(\'attachments/dvbbs/20048196543728688.gif\');}" onmousewheel="return imgzoom(this);" alt="" />
www.passlabs.com/np/GC-BOOTSTRAP-1a.pdf

What? You want your chip amp to behave as if it's seeing a
higher impedance and also deliver more current? You can
parallel chip amps equally, or you can set one up as a
current bootstrap, relieving the load on the first amp,
but also leaving it in control of the signal. Sort of
like power steering. Watch how you set the current gain
of the bootstrap. For this circuit, 50% is a good number.


<img src="attachments/dvbbs/20048196302178143.gif" border="0" onload="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onmouseover="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.style.cursor=\'hand\'; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onclick="if(!this.resized) {return true;} else {window.open(\'attachments/dvbbs/20048196302178143.gif\');}" onmousewheel="return imgzoom(this);" alt="" />
www.passlabs.com/np/GC-BOOTSTRAP-2a.pdf

And of course you can go crazy and use a lot of them. In
this case, the current gain would probably want to be set
to divide the current equally between all the amps, including
the first one. In the case of 4 bootstrap amps, each would
probably want to be set at delivering 20% of the output
current.


<img src="attachments/dvbbs/20048196372748727.gif" border="0" onload="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onmouseover="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.style.cursor=\'hand\'; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onclick="if(!this.resized) {return true;} else {window.open(\'attachments/dvbbs/20048196372748727.gif\');}" onmousewheel="return imgzoom(this);" alt="" />
www.passlabs.com/np/GC-ZEN-ALEPH-1a.pdf

How can we resist not making a Zen amp with such a current
source? Here we see that the Aleph current source is easily
duplicated by a chip amp.


原文出自 http://www.diyaudio.com/forums/showthread.php?s=&threadid=36585

flyingx

This circuit is a work in progress towards building a Bootstrapped SuSy GainClone using LM3875 op-amps, based on Nelson's circuit suggestion earlier in this thread. I have simulated it in Spice but I have not built a protoype yet. The sumulations look good, but as we all know, simulation is not the real world. Before I move into building a test circuit, does anyone have any comments on improvements?

Cautionary Note: Although I have given this circuit a good deal of thought, I am not an EE. Use with caution and at your own risk!

Terry

<img src="attachments/dvbbs/2004820343419376.gif" border="0" onload="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onmouseover="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.style.cursor=\'hand\'; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onclick="if(!this.resized) {return true;} else {window.open(\'attachments/dvbbs/2004820343419376.gif\');}" onmousewheel="return imgzoom(this);" alt="" />

flyingx

Post #25  
quote:
Originally posted by lumanauw
How does GC current bootstrap works?  


The bootstrap circuit is a lot easier to analyze if you simplify it a little. Pretend you drive it from a single ended source (so GND the -in). Then the first opamp looks just like a non-inverting opamp. The 2nd opamp is a little harder to analyze. Firstly there's the output resistor from the first opamp going into the load, that's probably a current sharing resistor so think something low value. 2ndly you see that the RF2 for the bootstrap opamp ties into the same thing. I suspect that RF1+RF2 of the 2nd opamp dictate the current sharing with Rout of the first opamp. You'd want their values to be the same yet small. Next you see that there's a positve feedback mechanism on the 2nd opamp but it's through a resistor so it sort of "defers" the + input voltage to the first opamp. Both of these facts lead it to the notion that the 2nd opamp is more like a comparator, all it wants to do is drive HARD towards one rail or the other. Then if you think about what happens if the 2nd opamp tries to drive too hard, the NFB mechanism of the first opamp sees this and tries to correct for it which will affect the way the 2nd opamp behaves.
So in short, what you have is that the 2nd opamp is a "current buffer" for the first opamp but is rigged to be slightly out of control so that the first opamps NFB loop is the dominant factor in the action.

Here's a simplified version... take this and add some PFB around the buffer and change the driver to a diff input and you'll see a similar result. Look at the OPA541 datasheet and you should see a similar circuit (although the figures are mislabeled in it).
--
Danny
<img src="attachments/dvbbs/2004820340515069.jpg" border="0" onload="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onmouseover="if(this.width>screen.width*0.7) {this.resized=true; this.width=screen.width*0.7; this.style.cursor=\'hand\'; this.alt=\'Click here to open new window\nCTRL+Mouse wheel to zoom in/out\';}" onclick="if(!this.resized) {return true;} else {window.open(\'attachments/dvbbs/2004820340515069.jpg\');}" onmousewheel="return imgzoom(this);" alt="" />

小鬼头

YINAN

O底下加一横的元件是什么啊????????

小鬼头

天涯羽飞

唉，大师的白纸一张，我们就得黄物一堆！