Express Stage 3 (V2) Clipper Rk: 10K > 8.2K

[PCollen]

Am I correct in thinking that altering the value of the third triode Rk from 10K to, say 8.2K or 6.8K would reduce the onset of clipping/distortion and therefore increase the headroom of the amp ? Has anyone experimented with this, or has it been discussed before (need I ask ) ?

[Cliff Schecht]

You are backwards. Increasing the gain will make the amplifier clip earlier, not later. Reducing the cathode resistance reduces the negative feedback on the cathode (cathode degeneration is a form of NF) and increases the gain of the stage. Capacitor bypassing also makes this resistance look like a low impedance and increases gain. You want to make this value larger or, better yet, reduce the plate resistor value.

[PCollen]

You are backwards. Increasing the gain will make the amplifier clip earlier, not later. Reducing the cathode resistance reduces the negative feedback on the cathode (cathode degeneration is a form of NF) and increases the gain of the stage. Capacitor bypassing also makes this resistance look like a low impedance and increases gain. You want to make this value larger or, better yet, reduce the plate resistor value.

I'm more intested in considering ONLY the effects on third stage (V2)performance, and not the overall amplifier.

My understanding was that RK = 10K biased this triode very much into the right, non-linear region on the DC load line where distortion of the input waveform would occur at the triode output (plate). Therefore, a lower value of Rk would bias the triode more towards the left on the DC load line, in a somewhat more linear area and thus lead to less distortion of the input at the triode output. Taking it to the extreme, consider replacing the 10K with a 1.5K..lots of signal level out of the triode and into the PI, but a very clean, linear waveform. Back to the 10K...much less signal level to the PI, and non-linear.or even clipped. So, would 6.8K give a greater, cleaner signal level to the PI vs. 10K ? I guess I'll get around to trying it soon enough and find out.

[jjman]

I wish I fully understood load lines and such.

My 3rd stage eventually clips on the bottom of the current wave, when driven hard, and not on the top. I've read the phrase "cold clipper" used to describe this biasing approach.

If an Express truly works this way (mine's not using Express voltages) I would expect a reduction in the 10k to increase the clean output potential of that stage.

However, the PI and the output is already clipping by the time the 3rd preamp stage reaches clipping. So the result would be clipping the PI harder/sooner "instead of" adding stage 3 clipping. I like the way the preamp clipping comes in at the end of the volume control.

[azatplayer]

I think if you are trying to clean things up, lowering the 150K clipping R is the place to look.

[PCollen]

I think if you are trying to clean things up, lowering the 150K clipping R is the place to look.

I don't understand how lowering the 150K grid-reference resistor on the 3rd stage of the Express would make that stage run cleaner, although I do understand how it would effect the frequency roll-off into that stage....can you be more specific ?

[fishy]

Reducing the value of that resistor will reduce the signal level into the stage. Having used a 150 down to a 33k in a Liverpool and a 150k and 68k in an Express, I think it is more a change of feel in the amp. There is still a lot of gain in the amp with a 33k.

[PCollen]

Reducing the value of that resistor will reduce the signal level into the stage. Having used a 150 down to a 33k in a Liverpool and a 150k and 68k in an Express, I think it is more a change of feel in the amp. There is still a lot of gain in the amp with a 33k.

I'll check it out....lowering that Rg on the 3rd stage is raising the -3dB roll off point of the high pass filter created by that Rg and the preceeding coupling cap (.0022): f(-3dB) = .159 / RC.

[fishy]

Reducing the value of that resistor will reduce the signal level into the stage. Having used a 150 down to a 33k in a Liverpool and a 150k and 68k in an Express, I think it is more a change of feel in the amp. There is still a lot of gain in the amp with a 33k.

I'll check it out....lowering that Rg on the 3rd stage is raising the -3dB roll off point of the high pass filter created by that Rg and the preceeding coupling cap (.0022): f(-3dB) = .159 / RC.

That RC network is regarded as the primary place to tune the amp. You can adjust both to maintain a similar freq or either component to change things a little.
It is worth the effort to try a few values / combinations even if you end up back where you started.

[guitarmike2107]

lowering the cathode resistor will reduce the sensitivity of that gain stage, i.e. the likely hood of overdriving. but it will also increase the gain of the stage causing the PI and power amp stage to overdrive faster.. therby actually giving you less headroom overall.

Check out this link by John Polstra
http://home.polstra.com/amps/wreck1/scope/

What it shows is that that gain stage (3rd) only really starts distorting once the power amp and PI are at full clipping.

I like the Komet style split anode tap trick to reduce the overall gain upfront, it doesnt affect the tone of the amp as much, but gives more clean range on the volume control.

Mike

[PCollen]

lowering the cathode resistor will reduce the sensitivity of that gain stage, i.e. the likely hood of overdriving. but it will also increase the gain of the stage causing the PI and power amp stage to overdrive faster.. therby actually giving you less headroom overall.

(snip)...

Mike

altering the value of the stage 3 Rg (150K) does not increase the stage gain ( http://amps.zugster.net/tools/triode-calculator), but would effect the strength of the various frequencies into the grid due to the changes in roll-off and therefore the voicing of the amp. Low-mids and upper-lows at a higher dB level at the grid due to a higher Rg could be percieved as increased stage gain, but may be increased signal level input for those frequencies into the tube. Example: stock, the F(-3dB) = .159 / (C x R) = 482 Hz . With Rg = 220K, F(-3dB) = 328 Hz, and with Rg = 100K, F(-3dB) = 722 Hz. And the roll-off STARTS higher.

[guitarmike2107]

lowering the cathode resistor will reduce the sensitivity of that gain stage, i.e. the likely hood of overdriving. but it will also increase the gain of the stage causing the PI and power amp stage to overdrive faster.. therby actually giving you less headroom overall.

(snip)...

Mike

altering the value of the stage 3 Rg (150K) does not increase the stage gain ( http://amps.zugster.net/tools/triode-calculator), but would effect the strength of the various frequencies into the grid due to the changes in roll-off and therefore the voicing of the amp. Low-mids and upper-lows at a higher dB level at the grid due to a higher Rg could be percieved as increased stage gain, but may be increased signal level input for those frequencies into the tube. Example: stock, the F(-3dB) = .159 / (C x R) = 482 Hz . With Rg = 220K, F(-3dB) = 328 Hz, and with Rg = 100K, F(-3dB) = 722 Hz. And the roll-off STARTS higher.

I said the cathode resistor Rk, not the grid leak resistor Rg.

Ta

[PCollen]

lowering the cathode resistor will reduce the sensitivity of that gain stage, i.e. the likely hood of overdriving. but it will also increase the gain of the stage causing the PI and power amp stage to overdrive faster.. therby actually giving you less headroom overall.

(snip)...

Mike

altering the value of the stage 3 Rg (150K) does not increase the stage gain ( http://amps.zugster.net/tools/triode-calculator), but would effect the strength of the various frequencies into the grid due to the changes in roll-off and therefore the voicing of the amp. Low-mids and upper-lows at a higher dB level at the grid due to a higher Rg could be percieved as increased stage gain, but may be increased signal level input for those frequencies into the tube. Example: stock, the F(-3dB) = .159 / (C x R) = 482 Hz . With Rg = 220K, F(-3dB) = 328 Hz, and with Rg = 100K, F(-3dB) = 722 Hz. And the roll-off STARTS higher.

I said the cathode resistor Rk, not the grid leak resistor Rg.

Ta

OOps.. that you did. Sorry for the mis-understanding.