Bob Cordell Interview: Negative Feedback

Jump to Latest
#2,221
Hi syn08,
I think dimitri is questioning the quality of most consumer gear. The "high end" units seem to have less respect for standards than the semi - pro market even.

You are right though. It really doesn't take much of a mismatch to throw the CMRR completely out the window.

Consider too, a balanced out is only an adapter stuck on to what is a unbalanced signal while it exists "inside the box". Then it depends more on whether you have a problem to solve or not before opting for a balanced connection.

My personal view is that often people are far better served with the standard unbalanced -10 dBv signal. Small studios as well. this after years of observation.

-Chris
 
#2,222
Bob Cordell said:



Have you considered using the THAT Corp InGenius 1200 Balanced Line Receiver? I think it does it all in one 8-pin DIP. Their full-complementary dielectrically isolated dirty sand might be just as good if not better than National's.

Cheers,
Bob
Click to expand...

No, but I played with the (on top of my head) TI's INA134. Perfect in terms of CMRR, acceptable in terms of distortion, not really there in terms of noise. But INA134 would fit most of the less demanding amp requirements. If the target is 100dB THD and 100dB S/N then its perfect.

I'm looking for some time now into the THAT parts, I'll give them a shot ASAP.
 
#2,223
My personal view is that often people are far better served with the standard unbalanced -10 dBv signal. Small studios as well.
Click to expand...

agreed, unbalanced is much better (and cheaper) than unproperly executed "balanced"

Perfect in terms of CMRR, acceptable in terms of distortion
Click to expand...

90db CMRR is with zero source impedance, but with rejection of common mode signal IC solution will do nothing with large ground loop via ground/shield/interwinding capacitance. Think how to break the loop 😉

Sorry for off-topic (Ovidiu, we can continue by PM or start a new ropic)
 
#2,224
dimitri said:

agreed, unbalanced is much better (and cheaper) than unproperly executed "balanced"

90db CMRR is with zero source impedance, but with rejection of common mode signal IC solution will do nothing with large ground loop via ground/shield/interwinding capacitance. Think how to break the loop 😉
Click to expand...

Agreed, but I don't see why the amp should limit in any way shape or form the CMMR.

Balanced is there, just in case. The amp is switchable to single ended or balanced via a backpannel slide switch.

BTW, it was the INA103 that I was testing, not INA134.
 
#2,225
Hi syn08,
Agreed, but I don't see why the amp should limit in any way shape or form the CMMR.
Click to expand...
Neither do I, a reality check now and again is helpful. It does no good to have a perfect interface in a very imperfect world. Once you get to a certain level of CMRR, then you change gears to optimize sound quality. This even if you detract from CMRR a little bit.

-Chris
 
#2,230
Re: Re: Re: Re: Re: Re: Re: PMP amp

Terry Demol said:


Just out of interest, what is the price tag on one of those (909)?

T
Click to expand...

[OT]
It was actually a joke - last time I've checked (a few years ago) the 909 was "only" about 40k euros (including VAT and shipping to Europe). The Wavac SH-833 has a price tag of $350,000. A nice price tag for listening to 10% distortions:

http://www.stereophile.com/amplificationreviews/704wavac/index5.html

But of course, it *must* sound great, just read through the review...
[/OT]
 
#2,234
You will note that both the JC-1 and the JC-2 have mostly 2'nd harmonic, by far. You have to push the heck out of these designs to get significant 3'rd harmonic or more. Since my designs are perfectly symmetrical, it is surprising where the distortion originates. I suspect with the JC-1, it is the P and N channel mismatch on the VAS stage (2nd stage), but it is still a mystery to me were it is coming from with the JC-2 as measured by 'Stereophile'. It is probably a product of the circuit feedback topology.
PS I am also told that you can get a gold plated version of the Burmester for $1000 or so. I am thinking of offering the same thing with Vendetta.
 
#2,237
Hi, Mr. Curl,

I suspect with the JC-1, it is the P and N channel mismatch on the VAS stage (2nd stage), but it is still a mystery to me were it is coming from with the JC-2 as measured by 'Stereophile'. It is probably a product of the circuit feedback topology.
Click to expand...

Could this a possible cause? I saw a design, where the upper and lower common source mosfet (VAS position) has different G-S resistor value. I ask myself, why the designer do that.
When I look at datasheets, N and Pch has different current gain for a certain Vgs voltage, say 5V for N and -5V for P. The problem is that a differential pair, the left and right leg have the same current balance (when the right -1mA, the right has to be+1mA). If we put the same G-S resistor value, it will need different left and right leg current of differential to make the same push and pull current of the VAS or vice versa. We know that unbalance in left and right current of differential pair legs produces harmonic.
 
#2,240
Re: Re: CFB clamp

Bob Cordell said:
[snip]
I’ve used a variation of this in some of my amplifiers that I refer to as a “Flying Baker Clamp”. Normally a Baker Clamp diode sees the full p-p signal swing and its capacitance variation can cause some high-frequency nonlinearity at the output node of the VAS. What we would like is a clamp diode arrangement wherein under normal signal conditions both sides of the diode are moving equally with the signal, so that a high voltage diode is not needed and junction capacitance variation is minimized.
[snip]

Cheers,
Bob
Click to expand...

Hi Bob,

This seems a very clever idea, but as long as the VAS output is also loaded with the nonlinear base-collector capacitance of the pre-drivers, it is not a big deal. :sad:

Cheers.
 
You must log in or register to reply here.
Top Bottom