6N3/6N6 headphone amp using PCB.

[JoeAlders]

Its this one,ecxept the cathode resistors are now 182 ohms.

Is that first cathode resistor 182 Ohm? Because that first stage is then on 1.7 mA only.

 

[Ryssen]

Yes

 

[JoeAlders]

When I am running your diagram in LTspice and doing an analysis I get DC voltage at the common point cathode top tube and plate bottom tube of the 6N6 triodes around 59.5V which I find low. Isn't it better to get there around halve the supply voltage. (=280V)?
But perhaps I do not have the correct 6N6 models.

 

[JoeAlders]

What also puzzle me is that you are using 6N6 triodes, which can handle at least 4 W plate dissipation. So what is in fact the problem with your original design with 280V supply?

 

[Ryssen]

But perhaps I do not have the correct 6N6 models.

Do you have a model for ECC99 wich I use?

 

[JoeAlders]

Yes I do. I will run your diagram using the ECC99 in the output stage (White Cathode Follower) and look at the results. Will let you know tomorrow.
Joe.

 

[Ryssen]

I think I founde the error...
R12 is connekted at the wrong side of c4,so the grid gets no ground potential....
I will fix it tomorrow.

 

[kodabmx]

Thought so.

 

[Ryssen]

It´s working now.no I am under 5W/tube.
But now I am beginning to understand the importance of using matched tubes....

 

[baudouin0]

IF you change R1 from 82R to 100R that will reduce the current. The tubes don't need matching for WCF. I would be surprised if the ECC99 and 6N6P should different with the linearity of the circuit - tubes don't really have a sound.

 

[Ryssen]

I use 182 ohm for R1 now. Well sommething changes when I change the tubes. (sound)

 

[JoeAlders]

I use 182 ohm for R1 now. Well sommething changes when I change the tubes. (sound)

Well I am now a little bit puzzled here about the actual resistor values 182 Ohm, 82 Ohm 100 Ohm?
Is it possible to post the actual resistor values of all resistors which are now "in the hardware"?

And perhaps someone can shed some light here about that global feedback. It is connected from the output to the grid at the input tube, which has a series resistor of 100k.
Then in the simulation there is that voltage source, but in practice it will be an input volume control resistor.
Isn't then the amount of global feedback depending on that volume control actual resistor value to earth?

 

[Ryssen]

Is it possible to post the actual resistor values of all resistors which are now "in the hardware"?

I am using this schema except the cathode resistor 182 ohm.And Ecc99.... And then 6N2P as input tube
And I use a 1ooK potentiometer.

 

[JoeAlders]

When I am running a DC analysis my DC values differ from yours significantly.
For instance the plate voltage of U3 = 83 V in my analysis and around 122V in yours. Same for the output voltage:
mine: 89V, yours: 124V.
I am using 5751 and 6N6P both Ayumi Nakabayashi models.
plate current U3=1.64mA and plate current U1=17.7 mA.
both in my LTspice circuit.
Circuit components are now the same as yours.

 

[Ryssen]

Been listening for a while now,very well defind bas,whery transparant.
Listening to Zakk Wyld f.e.x.......

 

[baudouin0]

I would stick to 82 - 100R as it sets the current in the output and hence what you can drive. If you make R1 182 you should also make the final plate WCF resistor higher too. Yes the input pot will have some effect on the feedback that's why the input resistor is big.

 

[Ryssen]

I would stick to 82 - 100R as it sets the current in the output and hence what you can drive.

Yes I will make it 100 ohms.

 

[baudouin0]

A 12ax7 is fine for the first stage.

 

[JoeAlders]

A 12ax7 is fine for the first stage.

I use the ECC99 as first stage. A very fine tube.

 

[baudouin0]

Actually just about anything will work! The more open loop gain you have there the lower the distortion should be but the distortion is quite low already.