Adding a Boost Function instead of a MV pot?

[shaun]

It's been a while. Always good to come see what's happening in EL34 World.

I've been asked to build a PP tube amp for a local guitarist. Lucky me. He wants a lot, including a foot-switched overdrive channel, and a "boost" channel of about 3-6dB so he can go from clean rhythm into clean solo mode without any overdrive characteristics.

I have no trouble with the OD because I've used a Marshall cold-clipper foot-switched channel in designs before. I used the Hoffman relay power rectifier from this website - great stuff.

My question concerns the boost. He wants this foot-switchable also.

He does not want a Master vol, so I thought maybe I could install two sizes of grid leak resistor right before the cathodyne phase inverter. This is usually a 1M in amps like the PR. I figured if I had the signal path connected to, say, a 750K grid leak, then with the footswitch it changed to the 1M grid leak, I'd get increased impedance for the signal and thus a boost effect. (750k is just a guess. I'll have to experiment to get what he wants, probably.)

Of course, the amp would be running at a slightly lower volume when the 750K res was engaged because it would allow more signal to bleed to ground, and that's a shame, but this seems like the simplest arrangement I can think of - just like two settings on a master vol pot, in a sense. But I'm probably missing something, like maybe it'll go "thunk" real loud when the footswitch is activated.

I'm wondering whether anyone here has tried this?

The next thought I had was that if this was possible, maybe I could pull the same trick with a long-tailed pair, although it would probably mean switching the two 1Megs to 750Ks simultaneously. Plus the LTP is much more nuanced and might not behave as predictably as a cathodyne PI. Might not be a happy arrangement.

Any thoughts/suggestions? Thanks as always.

[mountainhick]

Someone correct me if this is wrong:

You would need to drop half the signal voltage i.e. 50% to drop 6db. So maybe shoot for a bit smaller drop.

Without adding a gain stage to actually boost, it would be a matter of subtraction. Need a voltage divider.  Output impedance of a 12ax7 grain stage is about 40K. to drop 50% of signal you'd need a 40K load resistor between signal and ground.

I would add a resistor parallel to and before the 1M grid leak that can be switched in/out to ground.  Maybe start with 100K. Parallel with 1M is about 90K, and 40K/90K voltage divider gives you about 30% attenuation, 70% signal which is down 3db.

https://www.digikey.com/en/resources/conversion-calculators/conversion-calculator-parallel-and-series-resistor

https://ohmslawcalculator.com/voltage-divider-calculator

https://sengpielaudio.com/calculator-gainloss.htm

I don't think you'd have a large pop if after a coupling cap, decoupled from DC, but you could use a snubber across the switch if needed.


Other things that can work:

Cathode bypass cap switch.

Tone stack lift.

[AlNewman]

I have a mod in my silverface super reverb between 1M and 300K grid leak at the PI.  It's a different sound and feel, and does hit the plates of the output a bit harder, but I wouldn't call it a boost.
I would look towards a switchable cathode bypass cap. 

[kagliostro]

At the moment I don't remember the model but there is an Hiwatt that has someting that can be considered interesting

MV ... years ago out friend Da Geezer planned someting that may be considered interesting for the pourpouse of this thread

Later I'll try to find the documentation about what I say in my database and I'll publish it here

Franco

[Jennings]

You could also look at maybe the approach Sound City and others took with their "sensitivity" switch.  Basically a switched preamp gain stage cathode resistance.  Same sort of gain boost/cut effect by varying the stage's bias.  You can also make the resistance variable with a trim pot and dial in the level.  Have a look at the attached schematic...you're looking at the area marked "Gain".

[shaun]

Some excellent suggestions, gents - thanks! I like mountainhick's idea of a parallel resistor to ground before the 1M leak resistor - makes perfect sense. I am not good at following the math, mountainhick, but I get what you're saying and will attempt to do some math. I might include some presets in the design so the end user can set the boost level where ever they want for that stage.

I was really pleased to get a look at the Sound City schematic. Thanks Jennings. I initially considered switching the cathode bias in a preamp tube, and I have used that system in the past with pentode preamp tubes and a manual switch on the faceplate, but in this new design I was aiming for the most simple version for foot-switching that I could come up with. But the SC schematic reminded me that I might have a spare 1/2 tube in the design I'm developing, so I could dedicate that triode as a boost stage.

AlNewman answered a question that worried me - whether tonal characteristics would change, and it looks like they could if I'm not careful.

But also, here's my thinking: an MV can be set at 1M (if it's a 1M pot), or at any other res value on the wafer. I'm thinking of a post PI MV in an SE amp, for example. All I want to do is have two resistors instead of the MV pot, then make it switchable. It seems so simple that I'm assuming there's a problem with the idea. For popping, I can do as mountainhick suggests and add a snubber. I can't think of any other problems, but maybe I'll have to try it to find out.

[pdf64]

... I've been asked to build a PP tube amp for a local guitarist. Lucky me. He wants a lot, including a foot-switched overdrive channel, and a "boost" channel of about 3-6dB so he can go from clean rhythm into clean solo mode without any overdrive characteristics. ...

+6dB requires 4x the power output.
I can't envisage how a non-overdriven 6dB boost would be feasible unless it's a massively powerful amp, or the clean rhythm sound is set really quiet.

[kagliostro]

The Hiwatt schematics are here

for the Da Geezer scheme I must go on with the search

When you use the footswitch you short the 1M resistor and connect the 47K resitor to ground, part of the signal is shorted to ground

For your pourpouse you must realize a High signal level and short a part to ground, when you want your OD the 1M resistor must not be shorted

Franco

[shaun]

Thanks Kagliastro. Much appreciated.

I'm clearly way off when I mentioned a 3-6dB boost. Or at least, my client is. All he really wants is to be able to go from a rhythm volume to a slightly louder clean lead setting, and he's heavily influenced by Jazz, so he's not planning on doing much shredding.

[kagliostro]

I've found this demo for the SAP

http://www.youtube.com/watch?v=3JVGjonKABQ#

Description

Hiwatt SAP Balance control Demo DR 201 1969 KT88

No Pedals used. www.myspace.com/allrightnowband My 1969 hiwatt being shown for those interested in the Balance cicuit.

A touch of reverb added at mixdown.

For those who cannot see the Balance control and Switch setting too clearly:

We start with the Balance control full up and the switch disengaged.
 
At 00.31 I turn the Balance to slightly less than half way.

00.43 sees the Balance at about 9 o'clock.

00.54 sees the Switch engaged, note the Balance control is now ineffective.

01.19 disengages the Switch.

01.25 Balance back up to full.

01.39 sees the Switch engaged again.

02.09 sees the Switch disengaged, Balance control is now operative again.

02.22 sees the Switch engaged again.

02.50 Switch disengaged.

03.05 Balance taken down to half.

03.08 Balance taken to 10 O'clock.

03.14 Balance back up to full.

03.21 Switch engaged again.

----

And here the Da Geezer Circuit


Read the original thread for explanations

http://www.el34world.com/Forum/index.php?topic=10933.0

Franco

[stratomaster]

My inclination would be to nerf the normal mode.  Add a bypass cap to an unbypassed stage or maybe create a voltage divider right before the phase inverter input.  Use a switch to short the top resistor of the voltage divider. 

Ever notice how ECO mode vanished from cars right around the same time Sport mode showed up?  Same psychology.

[mountainhick]

create a voltage divider right before the phase inverter input.  Use a switch to short the top resistor of the voltage divider. 

I like this bypass idea. Needs a relay switch for foot switching.

[shaun]

... maybe create a voltage divider right before the phase inverter input.  Use a switch to short the top resistor of the voltage divider. 

Yep, I like it too. And I was planning on installing a relay. Seems simple enough, and shouldn't impact tone too drastically. I was thinking of switching the grid leak values using a relay, but stratomaster's idea is better because it keeps a stable grid leak resistor as a constant, and just switches out the top res. to engage or disengage the voltage divider. Poifect!

[jbrrrrr]

Some of these suggestions are reminding me of the Inst/Line level input switch on the Mesa 2:90 power amp - it's a very unscientific way of throttling (or nerfing) the Line input mode, where the Inst. setting looks more like a traditional guitar amp input stage.

Here, the input has a 1k5 resistor to ground for the grid leak, and a series 47k resistor is shorted for the line level state.

The only thing I might consider is that it might be better to do this later in the preamp stage arrangement, because at the input or first stage, it might have too much influence on how the core tone is developed.  At the PI, the "reverse-boost" might not be "clean" enough - maybe 2nd stage is better?  IDK.  Just saw this post and was reminded of this method of controlling input level/gain.

[kagliostro]

Instead to use a voltage divider I'll consider to use a Split Plate Load Resistor with it's advantages
The signal at the junction of the plate resistors will be attenuated, if you connect to the plate the signal will be at it's max
You can mebage the % of signal that is feed to the following stage using different resistor values (the sum must be the original value of the plate resistor)

This without alter the gain stage tone and avoiding the noise of voltage divider added resistors

Franco


p.s.: Note that the schematic is only an example and the second stage is not mandatory to be a CF, it can be whatever you want

[pdf64]

... This without alter the gain stage tone and avoiding the noise of voltage divider added resistors
 ...

Although in circuit analysis we assume the HT supply node to be 0V AC, in reality, due to imperfect decoupling, there may be a 'small but non zero' degree of HT ripple and audio artifacts there.
So it might not be preferable to use a split load with a lot of signal reduction, because then ratio of the unwanted stuff on the HT node will be larger in comparison to the audio signal.

[mountainhick]

Instead to use a voltage divider I'll consider to use a Split Plate Load Resistor

Higher voltage there, be sure to use a relay that can handle it.

Prefer these ideas:
-Bypass switch post coupling cap on a voltage divider on the signal path,
-Switched in/out parallel load resistor,
-Switched cathode bypass cap. (Not as keen on switched cathode resistor, adds switching noise when cathode bias changes)

If you are not keen on a fully unbypassed cathode, you can also use differing bypass cap values in series with a bypass across the smaller value connected to ground for silent switching. When bypassed you get the top cap's value, when bypass is lifted, the series caps' value drops to close to  the smaller value. Use a series cap calculator:

https://www.omnicalculator.com/physics/capacitors-in-series

You will have a tonal change, e.g. a 22uf with a .68uf, the larger cap selection will fill in more bass.

[SEL49]

Instead to use a voltage divider I'll consider to use a Split Plate Load Resistor with it's advantages

Just a thought... A split load resistor IS a voltage divider. Think about it. 

[kagliostro]

Yes, but it don't requiren to add other resistors

If the switch is placed after the decoupling caps the DC will not affect the switch

Franco

[SEL49]

Yes, but it don't requiren to add other resistors

Sure it does. The little circuit you posted has two resistors in the plate. And there's another two resistor voltage divider on the grid of that cf.

[mountainhick]

If the switch is placed after the decoupling caps the DC will not affect the switch

Franco

EDIT: Ah so, something like this?

[kagliostro]

@ SEL49

Yes, you are right, because in the original use the path of the signal is across a single resistor but in this use it may be taken after the second resistor

About the CF, as I told, it was only an example, not mandatory to have a CF, use which stage you prefer

@ mountainhick

Yes, that is what I was suggesting to give a try

My preferred try, however, will be Geezer double "MV"

If you want to give a try to the Hiwatt way, may be It will be better to use a relay instead to have se signal arriving till the pedalswitch

Franco