Electron tube pre output stages shootout

[Walter66]

Just the cost of iron makes me shiver...
That would make me opt for a mu-follower.
BTW, the standard 1668 is 100H, it's the 1667 that is 270H.

Good catch, corrected the choke type. Mu- follower wasn't the standard of german studio gear, they all used output transformer and choke for a parafeed output. Thats because they wanted to achieve a defined and constand low impedance output. This is a cost- no-object project, so don't worry about the price. This is just about the best technical solution for a line tube pre.

 

[Walter66]

How much gain do you want from it? You really do not want the feedback loop to load the output stage so you might typically ensure it is 10 times the nominal load on the output. With the 12K:600 transformer shown on your schematic you would aim for the NFB to load it by no more than 120K. With a 430R cathode resistor in the first stage this implies a closed loop gain of about 49dB - probably not what you want.

Cheers

Ian

The circuit aims for 20dB of gain. I see, its much too high. Could the feedback loop be applied anywhere else, more smart? Or by reducing the gain of first stage with adding an extra neg. current feedback loop?

 

[ruffrecords]

The circuit aims for 20dB of gain. I see, its much too high. Could the feedback loop be applied anywhere else, more smart? Or by reducing the gain of first stage with adding an extra neg. current feedback loop?

Is that 20dB including the 13dB loss in the output transformer or excluding it?

Cheers

ian

 

[abbey road d enfer]

The circuit aims for 20dB of gain. I see, its much too high. Could the feedback loop be applied anywhere else, more smart? Or by reducing the gain of first stage with adding an extra neg. current feedback loop?

You may start with increasing the cathode resistors.

 

[Walter66]

Is that 20dB including the 13dB loss in the output transformer or excluding it?

Total amount including OPT. So the two active stages should have 33dB amplification minus 13 dB loss due to transformer.

 

[ruffrecords]

Total amount including OPT. So the two active stages should have 33dB amplification minus 13 dB loss due to transformer.

You will probably be OK with the design as it stands. For that amount of gain the feedback resistor needs to be about 20K. That in parallel with 12 K is 7K5 which should be no problem with 15mA of quiescent current.

Edit: SInce you have it wired as a triode I think its plate resitance will be just over 2K based on gm = 9mA/V and u = 21 which is fine for driving a 7K5 load.

Cheers

Ian

 

[Walter66]

Thank you very much, Ian. Your calculations helps a lot.
Best regards, Walter

 

[bockaudio]

Because a cath-follower has low drive impedance, many think it's a good reason for using a lesser step-down ratio, e.g. 4:1 instead of 10:1, but it increases the tube distortion by twisting the load line.

FINALLY someone else besides a dead author says it.

 

[Walter66]

After testing the circuit, it can be certified that this kind of output works like a charm. Added an 1K in the anode, so b+ becomes 295V with 200V for the CV391.
It has a relative high gain, which have been reduced by inserting a neg. feedback from anode CV tube to the cathode of EF800, I think its 150K with 1uf. This gave excellent sonic results. Output cap has been changed to 2uf, that heavily depends on the output trans.
The 1.5M feedback loop taken from the output in this schemo has been eliminated.
Thanks all for your input, appreciated it very much.