Resistor in series with cathode bypass cap?

[ShoemanGB]

I've seen a few designs that have a switchable resistor in series with the bypass caps on a preamp tube.  Can someone please clue me in to why that might be done?  I'm guessing that adding some resistance to ground for the cap may form a filter of sorts?

[pdf64]

eg? Often it's a high value (100k) to avoid the switch popping.

[Latole]

I see resistor in parallel with bypass cap, never in series.

Can you show circuits you see that ? 

[SILVERGUN]

I've seen a few designs that have a switchable resistor in series with the bypass caps on a preamp tube.  Can someone please clue me in to why that might be done?  I'm guessing that adding some resistance to ground for the cap may form a filter of sorts?

This can be called a "boost limiting" resistor. The bypass cap provides boost. Putting a resistor in series with it limits that boost.

[d95err]

What type of resistance value is referring to?

If it’s dignificantly bigger than the cathode resistor, it’s just for preventing pops and effectively takes the cap out of the circuit.

If it’s about the same as or lower than the cathode resistor, it could be a way to shape frequencies by reducing the effect of the cap.

Don’t think I’ve ever seen the latter in a tube amp though.

[SILVERGUN]

Ref: "Designing tube Preamps for Guitar and Bass" - Merlin - 1st edition
Page 29


Or by replacing the resistor with a 100K pot for use as a "gain" control
Page 63

[sluckey]

Here's another example of a gain control...

     http://sluckeyamps.com/smoky/smoky.pdf
 

[eurekaiv]

I saw this first in a Divided by 13 CJ11 schematic I found online. I've built two amps now using this boost setup and I find it works very well. You can think of it as limiting the amount of effect the bypass capacitor has on the cathode bias by adding resistance in series with the cap to ground. I need to try the gain control in that Smokey schematic, that looks interesting as well.

https://el34world.com/Forum/index.php?topic=26307.msg287893#msg287893

[Latole]

I've seen a few designs that have a switchable resistor in series with the bypass caps on a preamp tube.  Can someone please clue me in to why that might be done?  I'm guessing that adding some resistance to ground for the cap may form a filter of sorts?

This can be called a "boost limiting" resistor. The bypass cap provides boost. Putting a resistor in series with it limits that boost.

I alway learn something here, Thank you all

[AmberB]

New information for me as well.  That 100K pot in series with the bypass cap does sound interesting...

[ShoemanGB]

Thanks for chiming in everybody.  I tried to find some of the examples but my search-fu skills failed me.  I did find an image I had saved from this thread that may have been what got me looking into the idea, see the layout image for the Allen Classic 10 in the first post:   https://el34world.com/Forum/index.php?topic=21870.0   and note the Boost switch.
  My Merlin book is the 2nd edition, so the page references Silvergun lists don't match but I'll do some digging there and see if he mentions the mods still.
  And of course in the end the best way to find out is to actually wire it and see. 

[2deaf]

You can think of it as limiting the amount of effect the bypass capacitor has on the cathode bias by adding resistance in series with the cap to ground.

The bypass capacitor doesn't have any effect on the cathode bias voltage.  The cathode bias voltage is a DC parameter and the bypass capacitor works by passing AC.

A large bypass capacitor eliminates the negative feedback that was created by the AC voltage across the cathode resistor.  A large bypass capacitor appears to be a straight piece of wire as far as AC is concerned.  A potentiometer in series with the large capacitor then appears to be a variable resistor in parallel with the cathode resistor for AC purposes.  The larger the variable resistance, the larger the negative feedback from the voltage drop across the effective cathode resistance.  When the variable resistance is set to zero, all negative feedback from the cathode is gone and the stage has its maximum gain.

One of the features of parallel resistances is that the resulting resistance changes exponentially with a linear change in one of the resistors.  Consequently, the gain of the stage in question will vary exponentially with a linear change in the pot that is in series with the bypass capacitor.  Also, most of the gain variation is going to happen when the pot goes from zero to 5K when the cathode resistor is 1K5.     

[2deaf]

Thanks for chiming in everybody.  I tried to find some of the examples but my search-fu skills failed me.  I did find an image I had saved from this thread that may have been what got me looking into the idea, see the layout image for the Allen Classic 10 in the first post:   https://el34world.com/Forum/index.php?topic=21870.0   and note the Boost switch.
  My Merlin book is the 2nd edition, so the page references Silvergun lists don't match but I'll do some digging there and see if he mentions the mods still.
  And of course in the end the best way to find out is to actually wire it and see.

The Allen Classic 10 gets its boost by eliminating the global NFB from the speaker voltage.  It shunts the NFB voltage to ground via a capacitor when the switch is closed.  The global NFB isn't replaced by the cathode NFB when the switch is closed because the cathode resistor is bypassed by the same capacitor. 

[eurekaiv]

You can think of it as limiting the amount of effect the bypass capacitor has on the cathode bias by adding resistance in series with the cap to ground.

The bypass capacitor doesn't have any effect on the cathode bias voltage.  The cathode bias voltage is a DC parameter and the bypass capacitor works by passing AC.

I meant to type bypass twice but the words are similar and my fingers didn't do what my brain wanted this time I guess. Thanks for pointing that out.

[ShoemanGB]

Thanks for chiming in everybody.  I tried to find some of the examples but my search-fu skills failed me.  I did find an image I had saved from this thread that may have been what got me looking into the idea, see the layout image for the Allen Classic 10 in the first post:   https://el34world.com/Forum/index.php?topic=21870.0   and note the Boost switch.
  My Merlin book is the 2nd edition, so the page references Silvergun lists don't match but I'll do some digging there and see if he mentions the mods still.
  And of course in the end the best way to find out is to actually wire it and see.

The Allen Classic 10 gets its boost by eliminating the global NFB from the speaker voltage.  It shunts the NFB voltage to ground via a capacitor when the switch is closed.  The global NFB isn't replaced by the cathode NFB when the switch is closed because the cathode resistor is bypassed by the same capacitor.

Maybe now I'm in over my head.  So the 22K resistor that the boost switch appears to take in or out of the caps connection to ground is not actually adding what thought it was?  I see a 22k NFB resistor in series with a feed from the 8 ohm output jack to the cathode upstream of the cathode resistor and it's 2.2uf bypass cap.  Is that 22k at the switch also forming part of the NFB circuit?    I'm confused.  Which is often normal for me. 

[2deaf]

So the 22K resistor that the boost switch appears to take in or out of the caps connection to ground is not actually adding what thought it was?  I see a 22k NFB resistor in series with a feed from the 8 ohm output jack to the cathode upstream of the cathode resistor and it's 2.2uf bypass cap.  Is that 22k at the switch also forming part of the NFB circuit?

The global NFB circuit is composed of a voltage divider with the 22K from the speaker working against a parallel combination that changes when the switch is opened and closed.  When the switch is open, the 22K from the speaker works against the parallel combination of the 1K5 cathode resistor, the internal cathode resistance, and the switchable 22K resistor.  The internal cathode resistance is about 1K3, so the whole parallel combination is 696 ohms.  If the switchable 22K never existed, the parallel combination would be 675 ohms.  So to answer your question, the switchable 22K does form part of the NFB circuit but it sure isn't a very big part. 

[2deaf]

The Allen Classic 10 was probably not the best example of switching a bypass capacitor.  The Mesa Dual Rectifier has a 47K resistor in series with a bypass capacitor.  When the 47K is shorted by an LDR, the cathode becomes bypassed.  When the 47K isn't shorted, its resistance is high enough that the bypass capacitor has no discernible effect.

If you just have a switch and a bypass capacitor, the switch side of the bypass capacitor will become charged when the switch is open.  The capacitor will suddenly discharge when the switch is closed creating a pop as pdf64 mentioned.  The 47K resistor keeps the switch side of the bypass capacitor referenced to ground so that there is no discharge when the switch closes.





     

[2deaf]

As noted, the fixed resistor in series with the bypass capacitor can be replaced with a potentiometer configured as a variable resistor to form a gain control.  Due to the exponential nature of this gain control, most of the usable range is going to be from zero to 5K on the variable resistor and 10K is nearly the same as infinity.  A 10K reverse audio pot works well for this type of gain control.

There is the opportunity to alter the frequency response as you alter the gain by using a bypass capacitor that only partially bypasses the cathode.  The Mesa example is doing this by using a 1uF bypass capacitor instead of the usual 22uF.

Attached is another example of using a potentiometer with bypass capacitors.  This one can do gain only or gain with frequency response change.

       

[boyfarrell]

Was curious. Can really see from the plots why the log pot would work well here.

Interesting gain control, which will also darken the tone at the same time it reduces gain.