Using a CCS on SE Output Stage

[eeyore]

Just wondering what the benefits and problems with using a CCS with a SE output stage. Providing it can sink enough power (I am thinking of it in a low power <3W SE stage), can it provide sonic advantage, or does it provide more problems then it solves?

[kmtang]

You would need approx 700V B+ and the heat dissipation from CCS is about 10 watts rms.

Johnny

[eeyore]

Okay, so more voltage is needed, or enough to accommodate the full range of voltage swing for the tube. I assume that the CCS is installed at the cathode, and not between the PSU and the output transformer?

What are the benefits? More linear operation of tube?

[DHT112A]

Like this????

Active loaded all DHT 300B

It extremely inefficient.

[danzup]

Originally Posted by eeyore

Just wondering what the benefits and problems with using a CCS with a SE output stage. Providing it can sink enough power (I am thinking of it in a low power <3W SE stage), can it provide sonic advantage, or does it provide more problems then it solves?

It is really the worst idea i have seen from long time .
Nothing to improve on classic SE !
The least number of components , the better !

[Globulator]

You have a concept error - you want to be pumping current through the OPT primary, not voltage - so a CCS is a waste of time and energy, and works against you - you want the current to change!!

An SE output tube is a perfect voltage -> current amplifier.

[kevinkr]

You could use a CCS in a parafeed arrangement with the OPT, note however that you would need ~twice the supply voltage to achieve the same plate voltage swing as you would get with a plate choke.

A CCS can also be used in the cathode circuit, but must be bypassed by a large, high quality capacitor in order to get the required gain, rp, and ac plate voltage swing. (AC current across the OPT primary will vary as a function of signal level, DC current would be fixed) Some people do just this, but I can see little benefit in an SE amplifier. There is one specific case where this may make sense - a PP amp where toroidal OPTs are being used and where dc current balance is extremely critical in order to prevent core saturation that quickly occurs with small dc imbalances. (Another way would be fixed bias with a cathode current sensing servo which would be much more complex.)

An SE output tube while definitely my preferred output device is not really a perfect voltage > current amplifier, one only has to look at the plate curves to realize that. Good yes, perfect no...

[eeyore]

So another words, it is not a great idea for the output stage...

[Globulator]

Originally Posted by eeyore

So another words, it is not a great idea for the output stage...

The way I see it is that the output transformer (OPT) is a current driven device rather than a voltage driven one. This is a debatable point but that is my view.

Therefore you want the tube (a voltage controlled current source) to just be a tube. Adding a CCS turns it into a voltage source to drive the OPT with which I feel is rather a waste of time.

The best idea is to choose a nice linear tube and bias it so that at the operating voltage (most of the B+ voltage in fact) gives you a nice straight input voltage/output current character. Beyond that you have to bridge the outputs and use a phase splitter - i.e. use a class A push-pull topology.