The "huge cathode cap method" on 18watters... does it work?

[Silvio55]

Isn't the load on the NT 470K just like the 18W? the 200K is the MV pot. When I changed the gridstoppers on the NT they didn't changed the tone in any noticeable way, in fact I didn't noted any change at all.
Yes, the NT has the .1 caps and the 18W the .01, just like the schems.
I have the feeling that the problem is still the crossover distortion, but I don´t have acces to a scope, so I can't be for sure, but it sounds exactly like the COD buzz, I just don´t know why the PR mod don´t solve it, even with 11V zeners. When I put the 150R bias resistor it made the buzz worse, even with the PR mod.

[Chuck H]

On the NT the 470k resistors are followed by another capacitor. WRT the grid this eliminates those resistors as a DC reference. So each 200k track of the MV pot IS the 0VDC grid reference from each power tube grid to ground.

It would be normal for the NT to have less crossover distortion than the 18 watt because I'm betting the PI signal in the NT is a lot less than the 18 watt. Almost all of the NT's distortion is prior to the power tubes. Almost none of the 18 watt's distortion is. Still, I would expect the PR mod to tame the buzz to almost none. Maybe recheck polarity on the diodes? Just generally make sure everything is as it should be. Maybe a sound clip?

[EFK]

I have to second what es345 is illustrating above. I have gotten to the point where I literally can't stand the sound of an amp with a typical cathode bypass cap. I don't know if what I hear is crossover distortion or some other artifact, but adding cathode bypass caps to my ear adds a lot of harshness and that nasty ass clacky sounding fizzy crap that trails notes until you drown it out by completely cranking the amp. I experimented a lot and found that I seem to prefer adding a cap of approximately .68 in value, oddly enough (odd in that it is the same as Marshall preamp values). I wouldn;t think that such a small value in the power amp would make any difference, but it does. When overdriven, it does add something of a singing quality/sustain to the top end but without all the added nastiness of the big cap that always seems to manifest as the amp begins to break up.

Another way around that crappy "crossover" type sound is to lower the B+ voltages and bias hotter, closer to 100% at idle. But you can't get away with doing that with higher plate voltages, and maybe that's a part of the problem there - I know I often build cathode bias amps with PT that are really putting out too much plate voltage unless I go with a less efficient rectifier aka 5U4 or 5R4 (I'm dealing w/ big bottle amps here so a 5Y3 isn't really an option).

[Silvio55]

On the NT the 470k resistors are followed by another capacitor. WRT the grid this eliminates those resistors as a DC reference. So each 200k track of the MV pot IS the 0VDC grid reference from each power tube grid to ground.

Ah, now I got it! thanks!

It would be normal for the NT to have less crossover distortion than the 18 watt because I'm betting the PI signal in the NT is a lot less than the 18 watt. Almost all of the NT's distortion is prior to the power tubes. Almost none of the 18 watt's distortion is. Still, I would expect the PR mod to tame the buzz to almost none. Maybe recheck polarity on the diodes? Just generally make sure everything is as it should be. Maybe a sound clip?

The PR mod sure helped with the NT! I cheked, for the twentieth time the polarity of the diodes and they´re fine, the cathodes (the marks) facing each other, the 1N4007's anodes to ground, checked the diodes with the meter and they are also fine. Maybe I´ll put a switch on the diodes to see if the PR mod is making any difference.
The fact that putting a larger bias resistor made the buzz worse makes me think it´s a cross over distortion problem.
I´ll see if I can make a sound clip later, I can´t crank the amp right now!

[Silvio55]

I don't know if what I hear is crossover distortion or some other artifact, but adding cathode bypass caps to my ear adds a lot of harshness and that nasty ass clacky sounding fizzy crap that trails notes until you drown it out by completely cranking the amp.

My problem is actually when I crank the amp (it´s a non MV amp), I don´t hear any fizzyness with the volumen at lower settings.

Another way around that crappy "crossover" type sound is to lower the B+ voltages and bias hotter, closer to 100% at idle.

My dissipation at idle is around 14 watts, more than 100% (12W)!!

[Chuck H]

My problem is actually when I crank the amp (it´s a non MV amp), I don´t hear any fizzyness with the volumen at lower settings.

My dissipation at idle is around 14 watts, more than 100% (12W)!!

Actually, that 14 watts is for the whole tube. So your plate dissapation is probably just at 100% or so. el84's, even fixed bias, always seem to be biased too hot when they're actually not. Probably because the screens idle between 1.5 and 2 watts.

[Silvio55]

Actually, that 14 watts is for the whole tube. So your plate dissapation is probably just at 100% or so. el84's, even fixed bias, always seem to be biased too hot when they're actually not. Probably because the screens idle between 1.5 and 2 watts.

I used the weber bias calculator for that, I think it assumes 5% of the cathode current is screen current.

[Chuck H]

Gotcha

[Silvio55]

Here´s the sound clip:
http://k005.kiwi6.com/hotlink/9e3as1...wlite_buzz.mp3

[Chuck H]

Other than the excessive bottom end for the distortion level making it flabby (stay calm) that actually doesn't sound like "bad" crossover distortion. It could almost qualify as "swirl". Which is, IMHE, rare with el84's. They usually sound more mosquito-y. More of a phase shift at high, buzzy frequencies. What you have may actually be more of the buzzies that occur with excess bass as it distorts the PI. Looking at the schem I can see a few things I might try. It would be great if you could scope the thing and see the level of crossover distortion. Then, if it's actually a small amount as I suspect, you could focus on other possible causes.

[Silvio55]

Other than the excessive bottom end for the distortion level making it flabby (stay calm) that actually doesn't sound like "bad" crossover distortion. It could almost qualify as "swirl". Which is, IMHE, rare with el84's. They usually sound more mosquito-y. More of a phase shift at high, buzzy frequencies. What you have may actually be more of the buzzies that occur with excess bass as it distorts the PI. Looking at the schem I can see a few things I might try. It would be great if you could scope the thing and see the level of crossover distortion. Then, if it's actually a small amount as I suspect, you could focus on other possible causes.

Sadly, I don't have acces to a scope. What do you think I could try here to reduce the bass, Chuck? Reduce the bypass cap on V1?

[Chuck H]

That's probably where I'd start. Maybe try 2.2uf to 4.7uf for C11. Also, C18 is a bypass cap. It should be larger than it is to balance the bottom end. In theory this could help to cancel distortions in the bottom end. So I might try changing C18 from .01uf to .1uf.

[Silvio55]

Well, I found that I made a mistake, actually two on the layout, I put C1 where C18 should be and vice versa, and instead a .022uf cap the guys at the electronics supply shop had sold me a .22uf cap, so I was running with C1 .01uf and C18 .22uf. I broke the .01 cap when I removed it and didn´t have any spares, just a .1uf cap, so I put a .022 at C1 and .1 at C18, reduced the value of V1 bypass cap to 10uf and now the bass is tighter than before, much better sounding! Still that swirl, but not that buzzy like before.
I would like to try the cathode zener mod (is that the Chuck H mod? ), but I don´t have 5W zeners now, when I get them I´m surely going to give it a try.

[Chuck H]

That swirl sound, when it isn't too buzzy, is actually a desireable effect for a lot of players. Because it tends to be drive specific it can be used sort of like a touch wha that responds to your pick attack. You'll hear it in almost any Trainwreck clip you listen to. I agree that at face value it seems like something you want to "correct". But try just playing your guitar without listening for amp problems (almost impossible, I know). You might find that it works in your favor. The reason reducing bass reduced the fizzyness is that overshoot is much worse when there is too much bass. That spike sounds like fizz. 10uf is still plenty large. Let's just say that the 50uf stock value was redundantly large. 10uf is still basically fully bypassed WRT guitar frequencies. Do try a 2.2uf in that position. Here's a clip (of a very good player) using a Trainwreck Express. It's easy to hear the swirl and crossover distortion (listen close to the ending notes as they trail off). It's also easy to hear the player using it to his advantage. These are some coveted amplifiers. And they obviously "suffer" the same problem as your amp

[kg]

I don't know why removing the bypass cap stops crossover distortion. Is it because crossover distortion is more analogous between the tubes and therefor cancels when not bypassed???

I'm wondering if Silvio55 is actually hearing diode clipping atrifacts, as suggested above. The values for the PR mod are amp specific. If we had a schematic with voltages...?!?

well, the grid conduction/diode clipping in a high Z/high DCR grid circuit CAUSES crossover distortion, so i think you're on to something.