diyAB Amp - The "Honey Badger"

[AudioSan]

Your welcome.

[sandrohv]

Originally Posted by keantoken

The ratio of re' to RE doesn't matter because it is the Ic swing that sets the harmonic generation across Re', and the Ic swing is the same regardless of RE.

If our VAS was voltage driven and the feedback loop did not keep the IC swing the same in all conditions in order to meet the output demands, then RE would decrease the harmonic generation of re' by reducing Ic swing (at the same time as reducing gain). But that is not the circuit that is on our bench.

The resistor does however increase the conversion of OPS drive current to voltage at the VAS input which CAN increase distortion, but luckily again our VAS is current driven so the effect is not very strong.

[...]

I have not once, ever, found the idea of "local feedback linearizing the VAS due to Miller capacitance or RE" to be helpful, and in fact every time I've seen it brought up it leads in totally the wrong direction. It's like a computer virus which hides itself by causing the CPU to skip over countless important error checks.

Hi Keane, this is probably one of the best posts I have seen in DIYAudio... not only because it is true, but it de-mystifies a lot of none-sense floating around.

So just to add my 2 cents here:

RE in the second stage (VAS)

As Keane correctly points out, for the second stage (VAS), all that matters for practical terms is the I-V transfer function, since the stage is configured as an integrator, driven by a high impedance (output of the Input stage). Also, distortion is purely proportional to the error current being drawn from the input stage output. So anyone who says that RE is needed to linearize the second stage voltage GM, does not know what he/she is talking about. I believe this wrong school of thought came from Cordell's book.

If fact, Sam Groner's paper as well as Keane's note above points out that having RE is detrimental to distortion. The mechanism is as follows:
- The RE resistor increases the conversion of OPS drive current to voltage at the second stage input which is non-linear (Delta VBE + RE x Non-linear OPS input current)
- The second stage input voltage, which is non-linear, creates a feedforward current through the Miller cap (CM)
- This feedforward current is non-linear and has to come from the input stage
- This generates a non-linear voltage across the LTP's input terminals which gets multiplied to the output by the amplifiers gain.

Since the feedforward current increases as the CM impedance decreases over frequency, this distortion mechanism increases over frequency. Therefore, it will be hardly present at 1K, but if you have a low distortion amp, it will manifest at 20K.


So why do reason most amps have an RE in the second stage? This is two fold:
- It makes it easy to implement second stage current limiting
- It stabilizes the second stage minor loop formed by the Miller cap.

... and well, people don't know the effect mentioned above

2. Local feedback linearizing the VAS due to Miller capacitance or RE
Yes and no.

The Miller capacitance linearizes the second stage by forcing its I-V transfer to be 1/sCM, but does linearize what actually matters for distortion. Therefore, explaining distortion from the point of view of the Miller cap linearizing the second stage is fairly useless. This poor explanation of second stage distortion comes from Doug Self.

Recognizing that what matters for distortion is the error current drawn from the input stage output, the linearizing feedback from CM does not address the following errors:
a.- IB from the buffer transistor which in turn has the following sub-errors:
i.- Output stage input bias current current or load current on the second stage divided by beta^2 (assumes a 2 transistor/darlington circuit)
ii.- the IB of the CE transistor in the second stage which is due to Delta VCE, Delta IC and the above
b.- The input voltage of the second stage being modulated... see [1]
c.- Baker clamp, where in non-clipping conditions, its non-linear capacitance adds to CM which is the 'feedback path' of the second stage loop.

Again, awesome post Keane, please keep it up!


P.S. re + RE terminology Stuart uses cames from me... that is typical Op-amp design terminology.

[thompsontechs]

Sounds pretty damned good on the lab speaker. I'm looking forward to getting the parts to get the other channel up tomorrow. All goes well, I can put a bow on this one in a couple days.


Next~!