Yes, but the output stage still needs the same peak-peak grid voltages to drive the output load.
With a low impedance node (due to the shunt feedback effect) the pentode driver stage is forced to supply larger than normal currents to meet the output grid drive requirements. This makes for larger excursions of the pentode driver grid voltages, causing more gm variation (from the large current variation) in the driver stage.
Obviously, the Tabor's current source like tail, on the 6AU6 differential drivers, helps to fix that some. Only 10 Ohms for cathode degeneration looks a bit sub-optimal though.
-----------------------------------
On the other hand, taking the (output stage plate, or UL tap) feedbacks all the way back to the driver cathodes puts more gain in the "local" loop (assuming a 3rd front end stage yet to make up for the gain loss in the drivers now) without causing any stages to have to transition thru large current variation, due to low Z loading. Tubes in general perform better with high Z loads.
The issues now become one of "over effective" Neg-Fdbk causing most of the low order distortion to dissappear (beside the nasty high order stuff). No low order distortion spectrum "tail" leaves no "detail" in the sound effects. But this is fixable by either reducing the local feedbacks (higher Rs, this does not put any 2nd harmonic back in though) or by loading the added input stage a little more than usual to give it some SE sound effect.
Or by sending a global feedback to the input stage's grid 2, to leave in a residual triode sound (from the "triode" mismatch in the g1 and g2 transconductance characteristics, using a normal high Z load on the input stage plates to minimize current variation in this case).
With a low impedance node (due to the shunt feedback effect) the pentode driver stage is forced to supply larger than normal currents to meet the output grid drive requirements. This makes for larger excursions of the pentode driver grid voltages, causing more gm variation (from the large current variation) in the driver stage.
Obviously, the Tabor's current source like tail, on the 6AU6 differential drivers, helps to fix that some. Only 10 Ohms for cathode degeneration looks a bit sub-optimal though.
-----------------------------------
On the other hand, taking the (output stage plate, or UL tap) feedbacks all the way back to the driver cathodes puts more gain in the "local" loop (assuming a 3rd front end stage yet to make up for the gain loss in the drivers now) without causing any stages to have to transition thru large current variation, due to low Z loading. Tubes in general perform better with high Z loads.
The issues now become one of "over effective" Neg-Fdbk causing most of the low order distortion to dissappear (beside the nasty high order stuff). No low order distortion spectrum "tail" leaves no "detail" in the sound effects. But this is fixable by either reducing the local feedbacks (higher Rs, this does not put any 2nd harmonic back in though) or by loading the added input stage a little more than usual to give it some SE sound effect.
Or by sending a global feedback to the input stage's grid 2, to leave in a residual triode sound (from the "triode" mismatch in the g1 and g2 transconductance characteristics, using a normal high Z load on the input stage plates to minimize current variation in this case).
Last edited:
Post #140 for Zout measurements.
http://www.diyaudio.com/forums/tubes-valves/72536-el84-amp-baby-huey-14.html
I never did any THD measurements but it is not a low distortion design.
Yves showed some test results for his ECL86 amp on which the Baby Huey is based.
http://www.dissident-audio.com/PP_ECL86/DA36.pdf
Cheers,
Ian
Ah, that's not me, that's Wawebourn.
And yes, it is also the current path for the drivers (PCL200's I think)
- Status
- Not open for further replies.