Hello. I am trying to build a hifi amp from scratch, and when I studied the tube hifi schematics, I found most, if not all, use a large dropping resistor (B+) right before the 1st input gainstage. The values are about 47k to 68k, while closer to the transformer it is more between 4k to 10k.
Thank you.
Thank you.
Normally the voltage for a ouput stage is higher than the dirver stage.
The driver stage has a low current so the use of a high value resistor has two purpose, one to drop the voltage to the value from project, then create a R-C filter taht kill the ripple that is an enemy for the gain stage.
The phase splitter normaly use a higher voltage and more curren than input stage , with a medium value and a cap you reach the right value and create a R_C filter.
So the input stage normally has two R-C filter, one from phase splitter and one it is own filter.
In some cases there is a secondary winding of the supply trafo dedicated to the gain stage and phase splitter stage but the concept is the same.
Walter
The driver stage has a low current so the use of a high value resistor has two purpose, one to drop the voltage to the value from project, then create a R-C filter taht kill the ripple that is an enemy for the gain stage.
The phase splitter normaly use a higher voltage and more curren than input stage , with a medium value and a cap you reach the right value and create a R_C filter.
So the input stage normally has two R-C filter, one from phase splitter and one it is own filter.
In some cases there is a secondary winding of the supply trafo dedicated to the gain stage and phase splitter stage but the concept is the same.
Walter
Member
Joined 2009
Paid Member
I do it on SS amps too.
I had an old Maplin design which hummed a bit.
So used RC decoupling of front end and hum went away.
Valves are sensitive to noise on power rails, especially if front end has a lot of gain. I tend to use CRC power supplies with power stage off first C and front end off second C.
Thank you I understand better, I am putting all the pieces togheter.
I have one more question, do you think I should also use a large value plate resistor in the input stage ? Will it be more beneficial in this stage at removing noise, vs adding noise from the high value resistance.
I have one more question, do you think I should also use a large value plate resistor in the input stage ? Will it be more beneficial in this stage at removing noise, vs adding noise from the high value resistance.
Last edited:
Early gain stages are class A.
The current demand does not depend on signal level.
>It does influence the charging of the capacitor making it slower, but will this add noise if the cap cant recover fast enough from transient ? or will it not happen in practice ?
In certain tube guitar amps, the whole B+ network would collapse in a pleasant way, with the amp being driven hard by the player. This resulted in an organic form of compression, which the player could make use of musically.
Now it's an art to design all the time constants in the B+ RC chain just right to get that effect. Even better to make it "in / out" switchable. I'm sure patents have been issued around this. I think there''s one where they derive an envelope signal from whatever is coming into the amp - and use that to modulate whatever operating point they want within the amp. Turn up your guitar - only the tone of the amp changes; loudness stays the same.
In certain tube guitar amps, the whole B+ network would collapse in a pleasant way, with the amp being driven hard by the player. This resulted in an organic form of compression, which the player could make use of musically.
Now it's an art to design all the time constants in the B+ RC chain just right to get that effect. Even better to make it "in / out" switchable. I'm sure patents have been issued around this. I think there''s one where they derive an envelope signal from whatever is coming into the amp - and use that to modulate whatever operating point they want within the amp. Turn up your guitar - only the tone of the amp changes; loudness stays the same.
If you look at the tube datasheets, power tubes (output) are specified at 350 volts up plate voltage, and signal tubes are specified at 250 volts down plate voltage. So you need to drop the input plate voltage, besides the hum reduction of the dropping resistor. Most don't have a capacitor on the input plate voltage; the plate voltage has to swing to produce AC into the next stage. The swing of the plate voltage is the amplified signal of a vacuum tube, unless they couple off a resistor holding the cathode up off the negative supply.
Frederico,
I enjoy your enthusiasm and welcome you to this hobby, but there is more to amplifier design “than using a large plate resistor to remove noise” etc.
Based on the questions you ask I want to suggest that you find a good, proven schematic of an existing design, rather than to put together your own. As you build it you will learn a lot and will end up with a well-performing amplifier.
Also would be a good idea for them to read some books on amplifier design and operation. Not that this is a bad question for them to have asked , but it shows me they don't really understand how tube amplification works.
I concur.
@Frederico - it's taken me the best part of 2-3 years to get to the point of understand enough to be dangerous
but recognise there's a whole lot I don't understand.I would say building a kit is level 1, modifying the kit is level 2, building a non-kit from an existing level design level 3, designing an amp based on a design is level 4 and a complete new design is level 5+. Each jump is a leap. and progressively gets more complex.
I went from a known design (6sn7+6as7) to thinking it would be easy designing my own from scratch. Through that pain, maths and slow understanding I've switched a little - "my" design has taken an existing design (atmasphere m60) and adapted it to use different tubes (ie level 4) - the result means you still need to go through the design process, the maths - including the power supply work which is a considerable amount of work.
You will need to think about what you're driving, what your input is, gain, power, class, impedance, frequency response, distortion, feedback, architectures, failure modes, variation of supply with mains regulation, .... then how to get a good supply, heaters, elevations if required etc etc etc... this all requires understanding of filters, inrush prevention etc.
Not that I've not enjoyed the journey and I've learnt a lot. However I think I would have learnt more, faster, from a simple kit to start with and playing with it and watching the outcomes before moving up.
Oh, and start with a transformer design... down the path of OTLs there be dragons (and bigger dragons hiding in the power supplies).
Last edited:
Absolutely! Stephe was modest not to mention her videos of amplifier building. Excellent resource for any level amplifier bulder. You could see her doing her latest build on youTube videos:
Here is the whole playlist
6SQ7-EL34 Tube Audio Amplifier - YouTube
And more info on her website
Skunkie Designs EL34 6SQ7
The first tube stage is always run in class A which means the current drawn by the stage is essentially constant. In any case, the turnover frequency of the RC filter is usually well below the lowest audio frequency of the audio band (else it would not do much in the way of ripple reduction) so it looks like a short circuit to audio signals.
Cheers
Ian