design help needed.

[neazoi]

hi I designed a simple SE triode preamplifier that can be seen in the picture. the preamp plays nicely and there is a vast ammount of bass compared on my previous preamp but I need a little bit of help on how can I bias it into class-A operation. I know what I have to do to bias it but I do not know the values of the components.
Should the mA at the anode be the maximum allowed by the tube datasheet?
This tube draws about 12.5mA at max and I only can make it draw 8.2mA, so this is not class-A right?
I need some info please.
cheers!

[kevinkr]

No, it is operating in class A for the amplitude levels you require otherwise distortion would be severe. (Other classes of operation imply conduction over less than 360 degrees.)

The component values you have chosen look reasonable for the tube type and voltages you have chosen.

To increase plate current reduce the cathode resistor value, note that as current increases the plate voltage will drop which will have the effect of reducing the change in plate current that you would expect from a reduction in cathode resistor value. Values between 47 and 100 ohms are reasonable, try a couple of values in this range and determine which you like best. For lowest output impedance a high quality bypass cap is recommended (given the several hundred uF required it should be made up of several good caps in parallel.) Alternately you could use something like 3 red leds in series in place of the cathode resistor.

[neazoi]

I used a 470mf bypass cap in order to extend a bit the range in the bass section. The leds idea is great indeed.
Also I have adapted a pontentiometer as cathode resistor and I change values easily. Although I have never reached more then 8.2mA at the anode even if I set the cathode resistor in small or large values... Maby this means that the tube is highly used? I do not know...
The sound is very acceptable though...
I was wondering of how could I push the tube to its limits to achieve a classA device but as you say it has to operate in classA even if the plate current is much lower...?
Thank you very much!!

[Cycline3]

This is a class A circuit. The grid resistors should probably be made smaller - on the input and next stage as shown in your pic. 1M is maximum permissible according to tube sheet. While they allow smaller bypass caps, I think a circuit sounds better with lower value grid resistors. If you drive the tube hard - a high value here can actually change the bias voltage causing you to lose linearity. Try 220k, 330k and check your plate current and sound again.

If you aren't using a negative feedback loop, im not sure 100 ohms in the cathode is enough negative feedback to keep distortion down. You say this sounds OK?

[neazoi]

Ok to be honest this is my second successfull attempt to design a preamplifier from scratch using an RF tube.
I thought so about the grid resistors, It seemed too big to me too...
The strange thing to my point is that yes, the preamplifier sound good to almost any value of cathode resistor with only small changes mainly in the bass section.
What value would you propose for the cathode? smaller or greater? the datasheed sayr recommended 100ohms that is why I chosed so.
The high range transparency is very good when I change the MKT output cap with a wima MKP and I guess it will be nicer with more expensive caps.
I think the greater the value of this cap the more extend in low frequencies?
For the input cap I choosed to use the output cap of the previous stage (the one inside a cd player, etc) and I prefered not to add another passive component there...
Finaly I believe a second identical stage will boost up the total gain, but what the value of the capacitor between two stages can be? The same as the output cap or smaller?

Wour help is very important to me as I am newbie in designing, and I can say I enjoy it!
Thank you so much for the info!

[kevinkr]

Originally posted by Cycline3
This is a class A circuit. The grid resistors should probably be made smaller - on the input and next stage as shown in your pic. 1M is maximum permissible according to tube sheet. While they allow smaller bypass caps, I think a circuit sounds better with lower value grid resistors. If you drive the tube hard - a high value here can actually change the bias voltage causing you to lose linearity. Try 220k, 330k and check your plate current and sound again.

If you aren't using a negative feedback loop, im not sure 100 ohms in the cathode is enough negative feedback to keep distortion down. You say this sounds OK?

A good ECC88/PC88 without local feedback should still be capable of rather respectable thd% at a couple of Vrms. See Morgan Jones for more on this. FWIW I no longer use any feedback, local or otherwise in most of my designs and achieve linearity everywhere except an SE output stage that would be considered more than acceptable even by SS design standards. SY will probably back me up on this one..

[kevinkr]

Originally posted by neazoi
Ok to be honest this is my second successfull attempt to design a preamplifier from scratch using an RF tube.
I thought so about the grid resistors, It seemed too big to me too...
The strange thing to my point is that yes, the preamplifier sound good to almost any value of cathode resistor with only small changes mainly in the bass section.
What value would you propose for the cathode? smaller or greater? the datasheed sayr recommended 100ohms that is why I chosed so.
The high range transparency is very good when I change the MKT output cap with a wima MKP and I guess it will be nicer with more expensive caps.
I think the greater the value of this cap the more extend in low frequencies?
For the input cap I choosed to use the output cap of the previous stage (the one inside a cd player, etc) and I prefered not to add another passive component there...
Finaly I believe a second identical stage will boost up the total gain, but what the value of the capacitor between two stages can be? The same as the output cap or smaller?

Wour help is very important to me as I am newbie in designing, and I can say I enjoy it!
Thank you so much for the info!

Why do you need more gain? If you really do, choose a tube with a higher mu like the 5842, Russian 6S4P-EV (both 40 - 50) or D3A (70) triode connected. Minimizing the number of stages is always good for transparency. I'm very minimalist in audio path design, and rather the reverse in psu design.

[Cycline3]

Originally posted by kevinkr
A good ECC88/PC88 without local feedback should still be capable of rather respectable thd% at a couple of Vrms. See Morgan Jones for more on this. FWIW I no longer use any feedback, local or otherwise in most of my designs and achieve linearity everywhere except an SE output stage that would be considered more than acceptable even by SS design standards. SY will probably back me up on this one..

OK sounds great.. have an schematics of what you mean? Even a generalized one? Just to show how you are laying out a stage? Cause it sounds like you mean running full tilt on gain, no FB from cathode resistor, no global feedback and from reading and experience, that usually is not a good combo for hi fidelity. But I'm open to the idea for sure - and interested to see what you mean. Thanks, -Sean

[Cycline3]

Originally posted by neazoi
I thought so about the grid resistors, It seemed too big to me too...
What value would you propose for the cathode? smaller or greater? the datasheed sayr recommended 100ohms that is why I chosed so.
I think the greater the value of this cap the more extend in low frequencies?
For the input cap I choosed to use the output cap of the previous stage (the one inside a cd player, etc) and I prefered not to add another passive component there...
Finaly I believe a second identical stage will boost up the total gain, but what the value of the capacitor between two stages can be? The same as the output cap or smaller?

First and foremost this is a general explanation to your questions.

I would drop the grid resistors for linearity and lower noise. As for the cathode R, there isn't a hard and fast rule - it depends on the circuit and the voltage of your power supply. When you choose plate and cathode Rs, you are defining the operating voltage of the plate, the bias voltage on the tube and the current going through it. A safe and typical design is to run the tube at about 60% of it's plate dissipation. In this case just under 1watt (I believe this is a 1.8 watt tube). So you choose your plate and cathode resistors so that you have that .95 watt dissipation, and the voltage on the plate is say 60% of B+ so you have lots of headroom to swing AC voltage (your music signal) without distortion. Make sense? So you choose resistors that make this happen - it's not a rule where you always use 100ohms. MAke sense?

[kevinkr]

Originally posted by Cycline3


OK sounds great.. have an schematics of what you mean? Even a generalized one? Just to show how you are laying out a stage? Cause it sounds like you mean running full tilt on gain, no FB from cathode resistor, no global feedback and from reading and experience, that usually is not a good combo for hi fidelity. But I'm open to the idea for sure - and interested to see what you mean. Thanks, -Sean

The key is I choose tubes with the appropriate mu for the application I am going to use them in. Global feedback is not required in most applications, except arguably in a power amplifier where it provides reduced output impedance and lower distortion, but you can design for better linearity in the first place, and if you can design a speaker system to work with source impedances typical of SE amplifiers then you don't even need the global feedback.

Take a look at my website and look for the dht transformer coupled linestage. (Its also on the positive feedback online website.)