Reducing bypass cap boost

[sand]

At the bottom of page 28 in Merlin's book (http://www.freewebs.com/valvewizard1...Gain_Stage.pdf) there is a section where he describes how to lower the degree of boost one gets from the bypass cap. It's basically an additional resistor in series with the bypass cap. The formulas there show that the gain without that additional resistor is 38, and with it it's 47.

So...using that method, the boost at higher frequencies is lowered, but the overall gain is higher. Is this correct? If yes, what to do in order to reduce the degree of boost while maintaining the same gain level? Double the two 1k resistors? Thanks.

[Wilder Amplification]

At the bottom of page 28 in Merlin's book (http://www.freewebs.com/valvewizard1...Gain_Stage.pdf) there is a section where he describes how to lower the degree of boost one gets from the bypass cap. It's basically an additional resistor in series with the bypass cap. The formulas there show that the gain without that additional resistor is 38, and with it it's 47.

So...using that method, the boost at higher frequencies is lowered, but the overall gain is higher. Is this correct? If yes, what to do in order to reduce the degree of boost while maintaining the same gain level? Double the two 1k resistors? Thanks.

Since boost IS gain, how do you expect to "reduce boost yet maintain the same gain"? Or are you trying to change the boost/gain but maintain the same output signal level of the stage?

[diagrammatiks]

It creates a tuned filter that's frequency dependent.

If you look at the graphs, the unbypassed cathode resistor has a single sloped line.

the bypassed cathode resistor has a 2 slope line.

the additional cap with a resistor adds another slope to the top of the line.

[Merlinb]

Sand, if you use a BIG capacitor then the gain of the stage will be flat at all frequencies. Adding the resistor just reduces everything equally.

If you use a SMALL capacitor then only mids/treble will be boosted. Adding the resistor just lowers the amount of mid/treble boost. If you kept increasing it then eventually the cap would have no effect at all, and you would be back to a flat frequency response at whatever the minimum possible gain of the valve already is.

[sand]

If you use a SMALL capacitor then only mids/treble will be boosted. Adding the resistor just lowers the amount of mid/treble boost.

That's the part I get.
What i don't get is this: the formulas indicate that there is an increase in overall gain with the resistor, from 38 to 47, right?

So...does it mean that with the added resistor the mids/treble boost was reduced, but the frequencies below that were boosted?

[riz]

What he's saying is, that at LOW frequencies, it's as if there IS NO BYPASS CAP (since the value of the cap in the diagram he's referring to is small--he's talking about PARTIAL bypass), so minimum gain value of the tube/no boost. That's represented by the first equation =38. With the resistor in series with the cap, you get SOME bypassing at HIGHER FREQUENCIES, hence the higher gain figure=47. LOL-- now what I don't understand is the next sentence, why both 1k resistors are in SERIES with the cap, to AC?! Looking into that...

[sand]

What he's saying is, that at LOW frequencies, it's as if there IS NO BYPASS CAP (since the value of the cap in the diagram he's referring to is small--he's talking about PARTIAL bypass), so minimum gain value of the tube/no boost. That's represented by the first equation =38.

Then I don't get this, from the book:

At low frequencies the cathode is completely unbypassed, so the stage will have minimum gain.

Does that mean that the stage will have minimum gain at low frequencies only, not minimum gain overall (since there is a small value bypass cap) on it? So, the 38 would be representative of the stage gain only at low frequencies? The way I read it is that the stage (one resistor + bypass cap) has a gain of 38.

[Merlinb]

Does that mean that the stage will have minimum gain at low frequencies only, not minimum gain overall (since there is a small value bypass cap) on it? So, the 38 would be representative of the stage gain only at low frequencies?

Yes, that is what is meant. Mid/treble will be boosted above that minimum gain of 38, while the low freq's are unaffected and get a gain of 38.

[sand]

Yes, that is what is meant.

And thus the lights went on Thank you. Great books, BTW, very helpful. What's the 3rd one going to be about

[sand]

LOL-- now what I don't understand is the next sentence, why both 1k resistors are in SERIES with the cap, to AC?! Looking into that...

Let me know when you have an answer. I can't figure that one out either

[Merlinb]

Let me know when you have an answer. I can't figure that one out either

For an AC equivalent circuit you look at one end of the cap and then look at all the ways you can get to the other end of the cap. In this case you would have to go through both resistors in series. This is related to something called "finding time constants by by inspection".

[riz]

I've been running into equivalent circuits more and more, and I guess it's time I studied that concept. Thank you for the pointer!