Bob Cordell's Power amplifier book

[keantoken]

Oops, in the previous post I got the orange and red traces confused.

And here is the FFT and OLG plots for with TPC set at 2.5% overshoot. There is practically no difference in THD; shunt compensation has 3db less (I improved the compensation slightly since the last post). So much for local feedback. And this is with a buffered CE TIS.

[keantoken]

My compensation scheme is like the schematic on page 85 of Cordell's book, except R7 is in series with C1 instead of C2.

[Bob Cordell]

Originally Posted by magnoman

Glad to see this thread alive again with some technical discussion, thanks.
As is often the case for me, I'm not 100% sure I am visualizing the intended circuit or implementation, specifically the above mentioned RC shunt compensation.

Since this is Bob's thread would it be possible to reference a specific figure(s) in his book? I would think that Bob would encourage and benefit from such a discussion.

Keantoken, your electronics insight is amazing for someone any age, but at your age its truly a gift.

Thanks
-Antonio

Hi Antonio,

I'm also glad to see this great discussion going on. I've been a bit out of the loop with a vacation and now a bad cold. The fuzzy head due to the cold medication makes it hard to make sensible comments.

Your observation about Keantoken is right on target.

Cheers,
Bob

[davada]

Originally Posted by dimitri

For settling please find this book

HI Dimitri,

The page link says it's unavailable.

Cheers,

[Struth]

Hi Guys

pg.85 has fig.4-5 showing lag compensation.

A series RC hangs from the VAS collector to ground. The zero in this network is supposed to cancel the pole for the input to the VAS. A second lag network is simply a cap to ground from the VAS base, which is also the output of the diff-amp.

Slew rate is low and asymmetrical with this approach, as Bob states. Basic miller compensation works better in his examples anyway.

Have fun
Kevin O'Connor

[keantoken]

The RC is across the TIS input, and at the TIS output is a cap to ground, which is reversed from Bob's example. This configuration gives lower distortion and can have higher slew rate.

I noticed in Bob's example the TIS input cap to ground is totally wrong, this will inject supply noise right into the output, and the current loop for this cap has to go through all the decoupling caps to get back to the TIS emitter. The cap needs to be across the TIS input. Otherwise no wonder it doesn't work very well.

PS. This is a nice discussion Bob but I think I may have killed it!!!

[Struth]

Hi Guys

In the sequence of comp scheme drawings in Bob's book, the lag method of this figure (4-5) is the first, and presumably the worst yet simplest to understand. He describes it as "simplest" and "often not the best".

I think that because he seems so invested in the compensation scheme used in his mosfet amp (because it works so well!), there can be some "short hands" taken with other schemes that are not preferred. Hanging a cap to ground from the signal path is something one can think of loosely, and then the AC ground that is the reference for the VAS emitter becomes one and the same as real ground. However, in the real world these are quite different, as Keantoken pointed out. I think Bob just made a "colloquial equivalence" oversight there (proving that Bob is human).

Keantoken, when you described "across the TIS input" I envisioned it as you meant it as that is in line with my own preferred compensation schemes. It should also be intuitive for anyone to visualise who can picture schematics easily.

Have fun
Kevin O'Connor

[keantoken]

The schematic on page 242 has something close enough to what I'm talking about, although there isn't a section on this alternate form of shunt compensation.

[Struth]

Hi Guys

Fig 11-17 (pg242) combines transitional comp with the favoured RC across the diff output, and lag RC from the VAS output to ground. The opne-loop circuits combine various compensation methods, as in fig.25-15. Lots of ways to achieve stability.

Have fun
Kevin O'Connor

[LineSource]

"The Devil is in the details" is an often used English idiom meaning that small things in plans and schemes that are often overlooked can cause serious problems later on.

Some audio amplifier designers believe that "The Devil is is in the ground" and favor using compensation schemes which minimize ground connections. There are white papers for fully balanced audio amplifiers claiming that only power supplies and cable shielding should be connected to ground. A ground noise source is sometimes added to Spice simulations in these papers.


Example complementary differential amp design uses current mirrors, Hawksford VAS with TMC compensation, and triple-T output. LTSpice THD for 300W @8ohms is under 1ppm.

The first attached FFT uses a R+C across each input LTP to control high frequency oscillation. Note that the 2nd harmonic distortion is lower than the 3rd harmonic.

This complementary differential amplifier simulates with lower THD when, instead of the LTP compensation, an R+C is connected to each VAS collector and ground to control high frequency oscillation. On the second attached FFT, note that the second harmonic is larger than the third harmonic. No noise source is added to ground.



I have not found a complementary differential power amp schematic with a VAS solution for Kevin O'Connor's 4T Wilson CM. "I think one of the simplest solutions is to use the Wilson CM with fourth BJT added, as this sets both diff collectors at the same potential - or at least the collector in question has a set voltage."