PP First stage, triode or pentode?

[jcalvarez]

Hello!
I have a question, what would be the preferred option for the first stage in a typical 6L6/EL34/KT88 push pull amp (with gNF) using a voltage amplification stage, LTP and output tubes? Triode or pentode? Advantages/disadvantages? I tend to favour triodes, but there are plenty of successful amplifiers using pentodes for the first stage, hence the question.
Thank you!
Jose

 

[Merlinb]

Pentode is kinda old school now. They were popular when signal sources were weak so you needed a lot of gain, but these days the opposite is true. Most people seem to convert their Mullard 5-20 front end to triode these days, for example.

 

[kmtang]

I would prefer using the triode in the first stage. Even the pentode would have higher gain than the triode, it is very sensitive to both bias and screen voltages to get minimal distortion. Low distortion can easily get from triode simply by finding the right amount of bias.

 

[rayma]

The optimal screen resistor value is sensitive to the individual tube.
No screen resistor, no problem.

 

[jcalvarez]

Thank you for your comments. Looks like the consensus is triode. It is definitely simpler than a pentode, lower ra, and generally triodes are pretty linear.

 

[6A3sUMMER]

Even with CD players that have Full Scale DAC voltage (about 2.1 Vrms; 3V peak) . . .
Then if you are designing a 3 stage push pull 845 amplifier, with input tube, phase splitter/driver pair, and 845 tubes with both grids biased at -225 Volts, Then you might consider using a Pentode input tube. You will need the gain, or with all triodes, you might need 4 stages to get the amplifier to work.

Oh . . . 845 tubes are old school.

Systems are not all the same. Sometimes they require different tubes, or more of them.

$0.03

 

[Bandersnatch]

How ever you slice the first stage, the one with a 6SN7 second stage is going to have lower output Z to the finals.

Getting into some more detail, there is no reason to deliver such low voltage to the first stage. Providing higher voltage will allow an increase in current; and neither is at risk for overheat due to high plate dissipation... 🙂

Douglas

 

[jcalvarez]

Getting into some more detail, there is no reason to deliver such low voltage to the first stage.

I'm not sure whether I understand what you are saying. In this topology, the first stage is typically DC-coupled with the LTP. Increasing the plate voltage for the first stage will reduce the plate to cathode voltage of the LTP, resulting in decreased LTP bias, thus there is less swing to play with until you get a positive grid.

 

[Merlinb]

I'm not sure whether I understand what you are saying. In this topology, the first stage is typically DC-coupled with the LTP.

He's talking about the enormous 270k dropping resistor in the power supply. On the whole it would have been better to re-think the coupling between the stages, rather than to starve the pentode of HT voltage.

 

[jcalvarez]

He's talking about the enormous 270k dropping resistor in the power supply. On the whole it would have been better to re-think the coupling between the stages, rather than to starve the pentode of HT voltage.

Understood. That's the consequence of using an ECC83 dc-coupled to the first stage. They need very low anode voltage in the pentode.

 

[grovergardner]

Just to add my own prejudice...The first schematic is a typical Mullard-styler topology, like an Eico HF-60. The second is probably the Eico HF-87, which used the same Mullard topology but with a triode at the input. Other considerations (like power output and THD) aside, I've always though the HF-87 is a lovely amp, better-sounding than the HF-60, because of the cathode bias and that triode at the input.

 

[jcalvarez]

Correct. Mullard 5-20 on the left, HF-87 on the right.