[Biff]

A fellow inmate on the vinyl asylum posted these comments about the design of the Cornet2. I have no way to "argue" because I am not an electronics guy. I'd be curious to hear what Jim has to say about this. All I know is I love my Cornet2 and the way it sounds. Isn't that what really matters?

Thanks.

(comments from inmate below)

Would anyone load a 300b into an impedance lower than it's own plate resistance?

I think most understand that would lead to a high percentage of harmonic distortion.

Rule of thumb would be a ratio of plate resistance to plate load of 1:3 for an output stage and 1:10 for a voltage gain stage.

I'm sure there is a method to the madness, I just don't see it.

Numbers, The plate resistance of a 12ax7 with a 910 ohm, unbypassed cathode resistor is somewhere north of 150k.

That plate resistance is in parallel with the 150k plate resistor so we have an output impedance of 75k driving the RIAA filter that is in parallel with the following stage's grid resistor.

At the knee of the filter the reactance of C202 is equal to the value of R205. 6800 + 6800 in parallel with the grid resistor 150k = 12637 ohms with a 10K in series. So we have a source impedance of 75k driving 22637 ohms.

That creates a very vertical load line for the 12ax7 resulting in high harmonic distortion. Maybe the voltage swing is so low that that doesn't matter. My question is why load the 12ax7 into such a low impedance?

What's the up side?

[poty]

In my opinion, several sentences from the comments are doubtful. It's difficult to answer "with numbers" as soon as I don't have any simulation software on hand and "manual counting" would be rather time-consuming, but...
Lets look into 12AX7 datasheet and specifically at "Class A resistance-coupled amplifier" section. Here we can find the standard usage with the following params:
Ub=300V (Cornet 2 Ua=300V)
Rk=900Ohm (Cornet 2 Rk=910Ohm) - the only difference is bypassing the Rk in the datasheet
Rp=100kOhm (Cornet 2 Rp=150kOhm)
Rg=100kOhm (Cornet 2 Rg=47kOhm in the 1st stage, 150k in the 2-nd stage)
Rload=240kOhm (Cornet 2 Rload= from 110kOhm at 20Hz to 16,5kOhm at 20kHz in the 1st stage, from 8 MOhm at 20Hz to 11,5 kOhm at 20kHz in the 2nd stage)

Counted load from datasheet: around 70 kOhm; from Cornet 2:  1st stage 63kOhm to 15kOhm, 2-nd stage 147kOhm to 10,5kOhm.

It seems there is no such catastrophic differences from the datasheet usage in design. The only question left is high-frequency loads. But at the output there is only small amount (-40dB at 20kHz comparing to 20Hz) of high frequencies. Input voltage is 5mV gives us only around 200-250mV (1st stage) of the anode voltage swing total, so linearity is clearly not damaged.

Maybe I'm wrong, but my impression that the output impedance of the stages on AX7 in the design is about 170kOhm. About 2 years ago I counted the value somehow and proved by recounting the RIAA curve.

P.S. By the way - Plate Resistance from the datasheet for General Electric 12AX7 is 80k for 100V Plate Voltage and 62,5k for 250V Plate Voltage (Cornet 2 - around 150V). Unbypassed Rk makes Plate Resistance even lower.

[hagtech]

I suppose this is one of the reasons I gave up posting on AA and other boards...

All I can say is that my choices for biasing and loading were not a mistake - they were very much intentional.  There are a number of reasons the circuit is the way it is - and I don't want to give all my secrets away.  As a designer I am not beholden to rules of thumb that came out of a 1925 tube book.  I make my own way and think outside the box trying many new things.  Most don't work, but sometimes you stumble upon something really creative and useful. 

jh

[jtwrace]

I suppose this is one of the reasons I gave up posting on AA and other boards...

All I can say is that my choices for biasing and loading were not a mistake - they were very much intentional.  There are a number of reasons the circuit is the way it is - and I don't want to give all my secrets away.  As a designer I am not beholden to rules of thumb that came out of a 1925 tube book.  I make my own way and think outside the box trying many new things.  Most don't work, but sometimes you stumble upon something really creative and useful. 

jh

And that's what makes a designer!   

[galyons]

I suppose this is one of the reasons I gave up posting on AA and other boards...

There are too many "Google-sperts"  that post on many of the forums.  They post as if they were "knowledgeable" , but really just mostly post nonsense or regurgitate out of context.  Then they frantically "Google away" to save face. 

AC is not immune in most of the circles.  Fortunately we have been blessed with them mostly staying out of here!!!

Cheers,
Geary

[Biff]

I suppose this is one of the reasons I gave up posting on AA and other boards...

All I can say is that my choices for biasing and loading were not a mistake - they were very much intentional.  There are a number of reasons the circuit is the way it is - and I don't want to give all my secrets away.  As a designer I am not beholden to rules of thumb that came out of a 1925 tube book.  I make my own way and think outside the box trying many new things.  Most don't work, but sometimes you stumble upon something really creative and useful. 

jh

Jim - I hesitated to post those comments, but I wanted to respond to the OP with some intelligence. As it turns out, my first response to him was probably the best - simply put, "Have you ever heard one?" (and he has not).

The proof of the pudding is in the eating as they say. I love my Cornet2 and don't see a need to change anything.

I think you do great work.
Perry

[mjock3]

There are too many "Google-sperts"  that post on many of the forums.  They post as if they were "knowledgeable" , but really just mostly post nonsense or regurgitate out of context.  Then they frantically "Google away" to save face. 

AC is not immune in most of the circles.  Fortunately we have been blessed with them mostly staying out of here!!!

Cheers,
Geary

I once heard it put this way. "The internet is full of reporters, but no editors."