design principle?

[stevlech]

I have a question about treble peakers: a few amps I've seen have a peaker between the plate load resistor and the coupling cap. Is there a specific purpose for this? Why would a designer choose to do so instead of placing it after the coupler e.g.- JCM 800?

[Reeltarded]

Between? You mean across?

Over the plate it cuts highs,

over the signal side of the divider it's a bypass to retain signal in a range, <peaker>

over the grnd leg of a divider it is cutting the same frequencies that it would retain across the signal side. <cutter>

There is math. Do you like math? I don't.

The above divider is equal both sides for this example.

[stevlech]

Thanks for the reply. Actually, I mean coming off the plate where a coupling cap is often located. The Naylor Superdrive comes to mind.

[Reeltarded]

Yep. That is a split load plate with a supression capacitor over the signal active side.

Couldn't have said that a year ago!

So, we want the tone of a 220k plate with only the signal of a 100k, so we take the signal from the junction of a 120k and a 100k. Split plate load.

To the right in your image there is the split with a cap for cut over it, a coupler, and a resistor before the next stage.

[stevlech]

Are you saying that it's equivalent to this circuit?
In this example, I see how the two resistors form a voltage divider with the resistor on the B+ side being grounded as far as AC is concerned. The circuit in my first post doesn't form a divider.

Maybe I'm not looking at it right, but I don't see it. I just see a peaker with DC on it and THEN a cap to block the DC from continuing down the signal path.

Gonna have to ponder this some more.

Thanks again for answering.

[Reeltarded]

Aha.. i do see what you are saying and now I am confused as well.

I needed more context. I looked up and found a Superdrive 60. Looks like there is a pass filter made after the plate to the coupler from a 100k with a switchable 220p coming off but this graphic I am looking at is hard to read here.. And.. The values at the head look to be insane. What is the plate limit there? Holy pig doodie. This thing has to be wrong..

No way. We are both talking about copies of the same insane schematic? That can't be right. 560k plate with a 470p over it on the plate end of the circuit opposite a 220k feeding it? Math is one thing, arithmetic another. No way.

That corcuit is missing a big cap on the grid to keep you from feeling the force. Wowch... unless that 220k mistake is that it's a divider..

[katopan]

The treble peaker and coupling cap are in series so electrically it doesn't matter about the order. It's just much more common to have the coupling cap first.

[Reeltarded]

Never even once considered that. Even knowing it can be that way. Target lock!

[vibratoking]

I hope this helps you. The schematic shows the approximate configuration. The placement of C1 and R6 is what you are asking about I believe. R3 and R7 form a divider that represents the pot set at it's mid resistance point. The schematic shows C1=220p, but this was overwritten with a spice directive during the sims.

OuputMagParVsSer.png shows two frequency response plots. The green plot is for the Naylor configuration, C1=470pF and R6 = 560k, while the blue plot is for the other type, C1=470pF and R6=500Meg. The 500Meg essentially creates an open from on side of the resistor to the other. It is essentially removed.

OutputMagPar470pVs100uF.png show 2 plots in the Naylor configuration with R6=560k. The blue plot is for C1=100uF while the green plot is for C1=470pF. You can see the 470pF cap creates a large reduction in bass frequencies and reduces the falloff in the 1 to 4 kHz range.

The plots show some of the differences that you asked about. It does make a difference where that cap goes and if it is in parallel with the resistor because it determines the pole and zero locations. The mathematical reason for the differences involves pole-zero analysis which is not really in the scope of this forum, I don't believe. Perhaps someone with a stomach for teaching that will chime in. Here is a wiki link that discusses poles and zeros and has some Bode illustrations. It may help.

[stevlech]

Wow. Thank you vibratoking. Your answer was far more thorough than I anticipated. I believe katopan answered my question: according to vibratoking's schem, C2 is after the C1/R6 filter whereas most circuits place C2 first and I was wondering why a designer would elect to position it there. I apologize for any inconvenience.

Thanks again to all of you.

[surfsup]

vibrato,

OuputMagParVsSer.png shows two frequency response plots. The green plot is for the Naylor configuration, C1=470pF and R6 = 560k, while the blue plot is for the other type, C1=470pF and R6=500Meg. The 500Meg essentially creates an open from on side of the resistor to the other. It is essentially removed.

OutputMagPar470pVs100uF.png show 2 plots in the Naylor configuration with R6=560k. The green plot is for C1=100uF while the blue plot is for C1=470pF. You can see the 470pF cap creates a large reduction in bass frequencies and reduces the falloff in the 1 to 4 kHz range.

I think you mixed it up. Look at the plots, the Green plot matches the Green plot, both greens are 470p starting at 25dB. I believe the 100u allows more highs to pass, effectively limiting lows (the blue dip in lower frequencies)?

The Blue higher output at lower frequencies on the first graph is puzzling. This should probably be done without the C3/R5 components...

[Reeltarded]

Please keep discussing so one day I can be a bad ass too... maybe not as bad, but at least I'll have the ass part down.

[vibratoking]

I think you mixed it up. Look at the plots, the Green plot matches the Green plot, both greens are 470p starting at 25dB. I believe the 100u allows more highs to pass, effectively limiting lows (the blue dip in lower frequencies)?

The Blue higher output at lower frequencies on the first graph is puzzling. This should probably be done without the C3/R5 components...

Thanks for the catching my mistake. I did have blue and green swapped for the OutputMagPar470pVs100uF.png. I have corrected it above.

I believe the higher output that you mentioned is due to the location of the zero. With 470pF//560k the zero is at about 600Hz, while for 470pF//500Meg the zero is at about 0Hz.

I don't know why I would do the sim without C3/R5? They impact the pole location.