This amp has never quite lived up to expectations. It currently lives on a PCB and I'm thinking of rebuilding it point-to-point. Before doing so, I thought I'd solicit any opinions as to design changes which might improve sonic performance. Thanks in advance.
You need more gain out of the first stage- you're in a situation where you've applied very little feedback, which is worse than lots and worse than zero.
If the screens aren't regulated, that could be a major improvement. I assume you actually have B+ on the plates...
If the screens aren't regulated, that could be a major improvement. I assume you actually have B+ on the plates...
At the link below you have un highly improved design. It's a Rene Cariou design (one of the famoust french tube amp designer)
The power supply has been little modified by Pascal Tajan (one other french tube fan)
Marc
The power supply has been little modified by Pascal Tajan (one other french tube fan)
An externally hosted image should be here but it was not working when we last tested it.
Marc
G2 :grid 2 from 6L6.
It's rated at 317 V.And Anode is rated at 310V.
Do you have any schem with 6L6 about 30W??
It's rated at 317 V.And Anode is rated at 310V.
Do you have any schem with 6L6 about 30W??
Is what you have a 6L6G or GC? If it's the latter, run your screens at 315V, anode at 500V, bias them AB2 and enjoy 50 watts of pentode power 😎
For best sound, run ultralinear, AB1 and 420 to 450V. Po is ~26 to 30W.
For best sound, run ultralinear, AB1 and 420 to 450V. Po is ~26 to 30W.
question!
@Idefixes
Nice schem!
@ Geek
Can I do this without any changes to the schem?
I think NO!
Can you tell me the changes???
Nice schem!
@ Geek
Can I do this without any changes to the schem?
I think NO!
Can you tell me the changes???
I don't understand what do you mean.Can you make it clearly for me?
If I go with 420 to 450V Va there will be changes to anode voltages at the first stage and the phase splitter.I thing that this will effect the charectistics of these stages.What do you thing?
Sorry for the newbie questions.But I'd like to learn.
Thanks
No problem! We all began somewhere 🙂
Pa is "Anode dissipation" Pa(idle) means "Anode dissipation with zero signal input". This is the vacuum tube equivalent to "Quiescent" current or power in solid-state amps (Iq, Pq).
Biasing your amps for a certain power just means multiplying your anode current by the anode voltage. So, for 25% Pa @ 450V, you get:
(25/4)/450V = ~14mA per tube at idle.
However, do experiment with values. The figure I gave you is a guesstimate for your setup and is book-standard for AB1. Try biasing the tubes as high as 60% Pa (which is almost pure class-A) and see what sound you like.
As for your driver section, yes, it'll slightly change the operating conditions, hence sound. Your 10K resistor decoupling the section is a good place to start, but try higher values for the sound you want. Even though the 12AT7 is rated 330V maximum Va, several friends and I have run them up to +520V with only a 20% degredation in tube life.
Here's also a short reference for you as to audio classes:
Class-A: Tube conducts 100% of cycle, plate current does not vary from minimum to maximum power. Pa(idle) is usually run at 90% of maximum. Suitable for single-ended as well as push-pull amps.
Class-AB: Tube conducts less than 100% of cycle, but more than 50%. Plate current varies widely from minimum to maximum power. Pa(idle) is usually run anywhere from 25 to 50% of maximum. Suitable for push-pull amps, or single-ended units with creative NFB techniques.
Class-B: Tube conducts 50% of cycle. Plate current varies widely from minimum to maximum power. Pa(idle) is usually run at less than 15% of maximum (depends on tube type). Suitable for push-pull amps only.
There are also numbers that can appear after the classes, 1 or 2. 1 means no grid current is drawn and no power is needed from the driver. 2 means grid current is drawn and power is needed from the driver. While grid current adds distortion, it does increase the efficiency of an amplifier and NFB usually cancells out the distortion.
Hope this helps 🙂
Pa is "Anode dissipation" Pa(idle) means "Anode dissipation with zero signal input". This is the vacuum tube equivalent to "Quiescent" current or power in solid-state amps (Iq, Pq).
Biasing your amps for a certain power just means multiplying your anode current by the anode voltage. So, for 25% Pa @ 450V, you get:
(25/4)/450V = ~14mA per tube at idle.
However, do experiment with values. The figure I gave you is a guesstimate for your setup and is book-standard for AB1. Try biasing the tubes as high as 60% Pa (which is almost pure class-A) and see what sound you like.
As for your driver section, yes, it'll slightly change the operating conditions, hence sound. Your 10K resistor decoupling the section is a good place to start, but try higher values for the sound you want. Even though the 12AT7 is rated 330V maximum Va, several friends and I have run them up to +520V with only a 20% degredation in tube life.
Here's also a short reference for you as to audio classes:
Class-A: Tube conducts 100% of cycle, plate current does not vary from minimum to maximum power. Pa(idle) is usually run at 90% of maximum. Suitable for single-ended as well as push-pull amps.
Class-AB: Tube conducts less than 100% of cycle, but more than 50%. Plate current varies widely from minimum to maximum power. Pa(idle) is usually run anywhere from 25 to 50% of maximum. Suitable for push-pull amps, or single-ended units with creative NFB techniques.
Class-B: Tube conducts 50% of cycle. Plate current varies widely from minimum to maximum power. Pa(idle) is usually run at less than 15% of maximum (depends on tube type). Suitable for push-pull amps only.
There are also numbers that can appear after the classes, 1 or 2. 1 means no grid current is drawn and no power is needed from the driver. 2 means grid current is drawn and power is needed from the driver. While grid current adds distortion, it does increase the efficiency of an amplifier and NFB usually cancells out the distortion.
Hope this helps 🙂
@Geek
Thank you for the detailed explanation!
I don’t know if I’ll use 6L6,yet ( What do you say about EL34? I don’t know which tube to choose,EL34 or 6L6).
If I choose 6L6 I’ll try ultra linear AB1 at 450V. But I’d like to ask you one more question. Go with self bias or fixed bias? And why?
What about NFB? How much NFB to try?
Thank you for the detailed explanation!
I don’t know if I’ll use 6L6,yet ( What do you say about EL34? I don’t know which tube to choose,EL34 or 6L6).
If I choose 6L6 I’ll try ultra linear AB1 at 450V. But I’d like to ask you one more question. Go with self bias or fixed bias? And why?
What about NFB? How much NFB to try?
resident said:
Yer welcome! That's what we're here for 😉
Both are great sounding tubes. Some like EL34 better, others 6L6. I like both.
I like fixed bias for PP because of in operation flexibility and you don't need perfectly matched tubes. You also can squeak a little more power out of the tubes too.
Self-bias is also great, especially for SE amps. The circuit simplicity also lends itself for more reliability.
Audiophoolery metaphysics aside, try your amp with no NFB. If it's stable (no oscillations, especially ultrasonic ones) and you like the sound, go for it.
I like a little NFB myself to "mellow" out the superfast transient responses. Rule-of-thumb is maximum 10 to 12 dB of NFB. I say use what % needed for you to enjoy the sound, since it's you that'll be listening to it all day 🙂
- Status
- Not open for further replies.