Please Help me to Understand this Phase Inverter.

[HaroldBrooks]

Sorry, this may be a repeat question, but I will try to be more specific in my questions this time around, as I am still lost.

1. What is the function of the .005uf cap just after the first (pin #1) plate (I believe one function is to block DC as a coupling cap) , along with the 6.8 Meg resistor that is connected to the 100k resistor to ground ? I'm not sure how that all works ! It looks like a voltage divider to me, but I'm not used to seeing those types a values (100k for a cathode resistor ?!?!), and it's got me wondering about it all, and it makes me realize I don't understand how the #7 grid gets a proper signal from this, or how the cathode voltage still works with such a big 100k resistor to ground.
2. What about the the tiny .05uf cathode resistor bypass capacitor over the 6.8k resistor on the pin #3 cathode. I've never seen that small of a cap ever as a bypass. How does that work and does that value make sense ? Does this PI arrangement need that small of a bypass cap on the first half, or is it just that to add top end to that one side of the PI arrangement ? Seems weird.

I apologize if the answers are easy, but I still need to understand this arrangement fully if I want to make any value changes or mods. So far I increased the bypass cap on the #3 cathode from .05uf to a more standard 1uf and it added a boat load of low bass (much needed in this amp) and a good deal of gain as well. I am perplexed at that result as well, because 1uf is also a bit small ! I would typically be adding a 20uf cathode bypass cap or more to up the bass. That was a great change though all in all, and it matched what I was looking for.

Another change went sour however, as I believe when I increased the size of the .005uf cap off pin #1 to a .02uf cap, it created a poor sounding distortion (abrupt and harsh). I was trying to boost the bass going through the PI tube but it failed. Just tonight I put that Pin #1 cap back to the original schematic .005uf value, and all is well again.

I'm sad to say I am lost with this, and I want to get smart and listen carefully to any comments regarding this type of PI arrangement, and as always, thanks for your expert tutelage and the Help on this great board !!!

[Enzo]

Pins 1,2,3 are just a gain stage, then the 0.005 couples the signal from that plate to the grid of the phase inverter triode.

6,7,8 is the phase inverter, a plain old split load type. 0.005 caps couple the power tube grids to the plate and cathode of the inverter.

From Valve Wizard:
http://www.valvewizard.co.uk/cathodyne.html

The gain stage has a 6.8k cathode resistor, and the small cap is all it needs. Growing the bypass usually grows the bottom end, but you can do that at most stages.

Again, only one side of that tube is phase inverter, the side on the right. The left side is just a plain old gain stage.

[Dave H]

I increased the bypass cap on the #3 cathode from .05uf to a more standard 1uf and it added a boat load of low bass (much needed in this amp) and a good deal of gain as well. I am perplexed at that result as well, because 1uf is also a bit small ! I would typically be adding a 20uf cathode bypass cap or more to up the bass. That was a great change though all in all, and it matched what I was looking for.

The way it's drawn is a bit confusing but as Enzo says it's a gain stage followed by a cathodyne PI (similar to the 5E3).

I think you've landed on a good value for the bypass cap. On the plot below green is 0.05u, blue 1u and red 20u. 0.05u cuts most guitar frequencies and 20u lets too much useless subsonic bass through.

[eschertron]

The one thing that strikes me as odd is the 6.8M grid leak (?!) bias of the concertina stage. Maybe done to save 1 resistor? I can't imagine it's better than either of the two example schems shown on Merlin's page.

[Helmholtz]

Not topic-related, but the first thing I would change in this amp is raising its input impedance to at least 500k. A 47k/100k input kills PU resonance and acts as if the guitar's tone control was rolled down quite a bit.
Also a higher input resistor will allow the .005µ input cap to pass more bass.

[catalin gramada]

It is huge to develop some voltage necessary to establish a bias point for section used as phase spliter. As you can see the grid do not benefit by other reference relative to cathode: it is ac coupled from previous stage and only reference to grid relative to cathode is those huge 6.8M. Some microscopic grid leak current will develop enough voltage over 7 M to create a bias point reference for the tube. Think is called contact bias or so...In rest is pretty common gain stage followed by concertina type phase spliter inverter.
However should be redrawn, you can twist you neck trying to read it...

[catalin gramada]

Think the best thing can do is to change those old resistor with a fresh high quality one ( e.g.10M metal film) and replace/well decouple its B+ supply point with a good quality fresh electrolithic...and you can add a supplementary supply HF decoupling in parallel, say 100nF film cap directly solder over the stage.

[Randall]

Also, notice how the standby switch works by connecting the two out of phase signals from the PI, effectively cancelling each other out. I like it. but I wonder if it's noisy.

[Justin Thomas]

Could just put a crossline MV in, since it's the same thing... I don't know why it should be any noisier thsn other types of Fender-style Standby switches. Half of those I meet pop conspicuously, though never obnoxiously...

Justin

[HaroldBrooks]

Not topic-related, but the first thing I would change in this amp is raising its input impedance to at least 500k. A 47k/100k input kills PU resonance and acts as if the guitar's tone control was rolled down quite a bit.
Also a higher input resistor will allow the .005µ input cap to pass more bass.

Thanks Helmholtz, I recognized that as a problem early on, and tried to install a 470k input resistor to ground early on, but picked up a ton of hum, and this amp is super quite in it's stock form. I did back off the resistor to a more conservative 100k (vs the stock 47k) and the amp picked up quite a bit of gain and the hum is still very low, as it was stock.

I also changed the 'Treble' Channel capacitor to roll off starting around 300hz (forgot what value I used), as that makes the treble input much more useful for what I do. The amp in it's current slightly modified form will clip nicely with a guitar plugged straight in with no boost, and from top to bottom it sounds pretty good.

Not sure how much I am squashing my main guitars (Yamaha are my workhorse guitars, vintage Gibsons on rare occasion), but I am open to the idea of using a larger resistor if I could get rid of the input hum. It's not from the pickup, but from the input stage itself, and I am not sure what is causing it. I believe someone said a 10 to one input ratio is effective, and if that's the case I am close by using a 100k resistor to ground on the input.

Thanks for your great suggestion, as always !

[HaroldBrooks]

These types of switches are a bit noisy (as in pops) and weird when they get old. I have two of them, and both "leak" some signal across even in the standby mode.

I've become super good at knocking down my guitar volume, and kocking off boost at the end of each song, so I don't really need much of a standby, and when I leave the stage between sets with my amp still on, it's usually so loud in the venue, that I can't hear any noise or hum from my amp just a short distance from any stage - (of course bear in mind I am half deaf at this point ! LOL, thanks to playing a cranked Marshall plexi in my early days.)

I think if you start with a new good switch the standby works fine ( for a while that is ), on these old Valco amps. I think I've seen some schemes to lessen the "Pop" on this very board, and I may have to investigate how that was done, just for the fun of it.

[Helmholtz]

Thanks Helmholtz, I recognized that as a problem early on, and tried to install a 470k input resistor to ground early on, but picked up a ton of hum, and this amp is super quite in it's stock form. I did back off the resistor to a more conservative 100k (vs the stock 47k) and the amp picked up quite a bit of gain and the hum is still very low, as it was stock.

Higher input impedance increases sensitivity to electric interference and sometimes (depending on amp layout) necessitates shielding of the input wiring. So you need to decide if the tone improvement with higher impedance is worth the extra shielding effort.