• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

6AS7GA as push pull output?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I assume LTSpice isn't developed to handle larger voltages.

I have done simulations to 650 volts of B+ with corresponding plate voltage swings beyond 1KV. Then I built the resulting amp and it was reasonably close to the sim. These simulations involve TV sweep tubes. I have never simulated anything with a 6AS7.

Maybe one of the models you are using can't handle the voltage.
 
So just looking at the load lines for a 6SA7 in push pull AB1 for a headphone amp.

The question I have is - with push pull and AB1, can you push the dissipation line up due to less work by the tube - thus having class A below the 110mA & 13W curve, but class B part of AB1 beyond but below say 180% (ie 200% - the % duty cycle overlap).

Screenshot 2020-11-21 at 11.43.02.png

There's not much in terms of triode push pull AB1 load line examples rather than pentode, so I'm trying to get my head around the mapping.
 
Last edited:
egellings,

6AS7 high miller effect capacitance?

u = 2; Cgp = 11pF;
(1+2) x 11pF = 33pF
Cgk 8pF
33pF + 8pF = 41pF

A 300B with Cgp = 15pF and u = 3.85, and an actual gain under load of 3.0 . . .
(1 + 3) 15pF = 60pF, more than that because of Cgk.

A 300B has more capacitance than a 6AS7.
 
Last edited:
So just looking at the load lines for a 6SA7 in push pull AB1 for a headphone amp.

The question I have is - with push pull and AB1, can you push the dissipation line up due to less work by the tube - thus having class A below the 110mA & 13W curve, but class B part of AB1 beyond but below say 180% (ie 200% - the % duty cycle overlap).

View attachment 895347

There's not much in terms of triode push pull AB1 load line examples rather than pentode, so I'm trying to get my head around the mapping.

Figured this out :)
 
The VA/PI/Driver in my 6P36S "Black Edition" amp will easily drive 6AS7G to full output with room to spare for a gNFB loop.

I have just made the VA/PI and Driver sections of this into separate circuit boards. Let me know if you're interested in a set :)
 

Attachments

  • 6F12P-6N8S-6P36S-Black-edition..png
    6F12P-6N8S-6P36S-Black-edition..png
    66.2 KB · Views: 128
Complete PP 6AS7 Amp with Brute Force Driver

The VA/PI/Driver in my 6P36S "Black Edition" amp will easily drive 6AS7G to full output with room to spare for a gNFB loop.

I have just made the VA/PI and Driver sections of this into separate circuit boards. Let me know if you're interested in a set :)


This copied from RCA Receiving Tube Manual RC16 published in 1950. Not much is new in audio under the Sun or anywhere else.
Bootstrapping avoids the HV PS.:D
 

Attachments

  • 6AS7 from RC16.jpg
    6AS7 from RC16.jpg
    421.2 KB · Views: 148
The VA/PI/Driver in my 6P36S "Black Edition" amp will easily drive 6AS7G to full output with room to spare for a gNFB loop.

I have just made the VA/PI and Driver sections of this into separate circuit boards. Let me know if you're interested in a set :)

Wow, the Super Williamson Driver (sorta). Interesting what you did with the 6SN7 (6N8S) differential driver stage. Don't the unbypassed cathode resistors introduce some degenerative feedback, raising the output impedance from that stage? Although I'm sure it also reduces distortion there, which is a good thing.

For an amp that big, it makes sense to have a separate plate supply transformer for the output tubes. The driver stages don't need that much of a transformer for their plate supplies. I did it that way once, to get a 440V B+ to a driver stage for a pair of 300Bs with 360V B+. I think the separate power transformers helped the sound quality (but I can't be sure of that).
--
 
Last edited:
They raise the Zout, but the grid leak for the following stage is 330k so it doesn't matter, and there's plenty of gain already :) Depending on the amp, I either use DC boost for the front end, or a quadrupler off of the doubler as in post 252. One older design used a choke input for the output stage, and cap input for the front end giving 360V and 560V respectively from an 800VCT transformer.
 
Last edited:
They raise the Zout, but the grid leak for the following stage is 330k so it doesn't matter, and there's plenty of gain already

You know, that's one of the problems with the Williamson topology.

If using triode output tubes like 2A3 or triode-wired EL34 in Class A, I find the sensitivity is a bit too high, like only 200mV to full power, when I'd prefer something like 1V to full power. Reducing the gain at the diff driver stage could work nicely for that.

Also, I've noticed that the diff driver stage is where the highest signal levels are, and consequently where distortion levels are highest. Reducing the distortion there should make a pretty big difference in the distortion performance of the whole amp.

Nice trick, I'm gonna steal it from you! :)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.