Rebuild. 5e3 preamp + 5f1 output stage.

[utervo]

I am still worried about the gain structure with the extra triode compared to 5E3 and 5F1. In what way can I deal with it?

Does it make sense to make it a cathodyne phase inverter and just join the signal and feed to the 6V6 grid?

Edit: That would result no signal exept noise...

[Lynxtrap]

I am still worried about the gain structure with the extra triode compared to 5E3 and 5F1. In what way can I deal with it?

There should be many options. You could skip one of the inputs, and use the extra tube half as second gain stage, leaving out the third stage and the second tube alltogether. But it might not be exactly like the 5E3, but the amp isn't exactly a 5E3 anyway.

Or look at other SE amps with several gainstages for inspiration and ideas. The THD Univalve comes to mind, even if that is quite a different amp.
Or browse the AX84 projects for similar ideas.
Or like said previously, build the thing and try out lower gain tubes.

[utervo]

Thanks for all the help.

If I use the fourth cap on the filter section, the 5Y3GT effectively sees a 32uF cap?

[Firestorm]

Lots of switchable bypass cap options. Added film bypass caps parallel that are always on in case it would help to retain highs that the 5F1 was missing. Any problem paralleling like that that?

Parallel caps add so I don't see any advantage in having a 0.1 in the mix. By itself, it will only boost dog hearing frequencies. Maybe consider 0.68, 2.2, 4.7 and 10, switchable. For an interesting way to use variable bypass on the output, search for the VHT Special 6 Ultra schematic.

Is it ok to place the 100k plate resistors head to head like I have done, any experience?

Not problem, just takes up more board space. Where are you connecting your plate resistors to B+?

Does B+ get easily contaminated by AC hum?

No. B+ is an AC ground.

[utervo]

Parallel caps add so I don't see any advantage in having a 0.1 in the mix. By itself, it will only boost dog hearing frequencies. Maybe consider 0.68, 2.2, 4.7 and 10, switchable. For an interesting way to use variable bypass on the output, search for the VHT Special 6 Ultra schematic.

Thanks, I will check the schematic. I have dogs but they don't like guitar sounds too much. I am planning to use a simple SP3T switches to use caps one at a time.

Not problem, just takes up more board space. Where are you connecting your plate resistors to B+?

The normal place at the end of the power supply 22k resistor node.

[printer2]

I don't know about you guys but I see no problem with the four triodes in the first schematic. The dual input section has no more gain than a regular triode input of a Champ. The second triode is also regular 5E3/Champ fair, no more excessive gain here yet.

Next comes the PI stage, but in this case you are only using one output. Skip the added gain that you have this stage set up for, set it up with the regular Cathodyne values with 56k and 1.5k resistors. You get the same overload of this stage as you would in the 5E3 but with a gain of one. Is that not really what you are looking for? You could add or skip the NFB if you want.

[Firestorm]

I don't know about you guys but I see no problem with the four triodes in the first schematic. The dual input section has no more gain than a regular triode input of a Champ. The second triode is also regular 5E3/Champ fair, no more excessive gain here yet.

Next comes the PI stage, but in this case you are only using one output. Skip the added gain that you have this stage set up for, set it up with the regular Cathodyne values with 56k and 1.5k resistors. You get the same overload of this stage as you would in the 5E3 but with a gain of one. Is that not really what you are looking for? You could add or skip the NFB if you want.

But what would be the point of having a unity gain stage? You call it a "PI stage," but of course there is no PI; this is a 5F1 output.

[rdjones]

I don't know about you guys but I see no problem with the four triodes in the first schematic. The dual input section has no more gain than a regular triode input of a Champ. The second triode is also regular 5E3/Champ fair, no more excessive gain here yet.

Next comes the PI stage, but in this case you are only using one output. Skip the added gain that you have this stage set up for, set it up with the regular Cathodyne values with 56k and 1.5k resistors. You get the same overload of this stage as you would in the 5E3 but with a gain of one. Is that not really what you are looking for? You could add or skip the NFB if you want.

But what would be the point of having a unity gain stage? You call it a "PI stage," but of course there is no PI; this is a 5F1 output.

There may be something to having multiple consecutive stages where the clipping behaviour is staged in such a way that each clips slightly rather than a single stage that clips hard.
What this might add to a single ended Champ you'll have to let us know.

As far as "excess gain" look at the second stage of the Brown Deluxe with the voltage divider plate load to reduce the gain of the stage.
This presumably to compensate for the for the lesser attenuation of the simple tone control (as opposed to a full tone stack).

rd

[printer2]

Pretty much as explained above. While the 'PI' stage is unity it has under half the signal swing of a regular gain stage. A bit more going on in a P-P amp in overload that you would get with this stage to a SE output but nothing you can do about that. You could just leave out the stage or use it as a cathode follower, part of a reverb circuit, all kinds of things. But since the intent was to build a 5E3 SE amp this seems like something to try.

[Lynxtrap]

I don't know about you guys but I see no problem with the four triodes in the first schematic. The dual input section has no more gain than a regular triode input of a Champ. The second triode is also regular 5E3/Champ fair, no more excessive gain here yet.

Next comes the PI stage, but in this case you are only using one output. Skip the added gain that you have this stage set up for, set it up with the regular Cathodyne values with 56k and 1.5k resistors. You get the same overload of this stage as you would in the 5E3 but with a gain of one. Is that not really what you are looking for? You could add or skip the NFB if you want.

The "PI", turned cascaded third gain stage, was the problem here. Not the first two.
I understand your point about overloading the triode sounding like a Deluxe. But it simply is not a Deluxe, because it's single ended. I don't see how reducing swing would be of any benefit?

[utervo]

Next comes the PI stage, but in this case you are only using one output. Skip the added gain that you have this stage set up for, set it up with the regular Cathodyne values with 56k and 1.5k resistors. You get the same overload of this stage as you would in the 5E3 but with a gain of one.

You mean like now in the schematic? Please see attachment if I've drawn it correctly.

I've tried to simulate the missing 6v6gt so that the signal leaks to ground.

Is that not really what you are looking for?

Yes most likely it is. Merlin Blencowe's site shows how a dc-coupled cathode follower distorts signal, could be interesting in that position also: http://www.valvewizard.co.uk/dccf.html

I have been wathing amp videos lately and I seem to notice that there is something in most single ended amps that I don't like when distorting compared to push and pull amps. Is it the lack of sag or PI or what I don't know.

One I like very much is Victoria Regal II, there is something with that amp that makes it sound so smooth. Has anyone taken a look?

[utervo]

I wonder if Victoria Regal is running quite low voltages, or the tubes are biased to clip symmetrically, or both...

[utervo]

Can anyone check if the cathodyne is drawn correctly?

What do I need to add to be able to adjust 6v6 bias to symmetrical clipping?

[utervo]

Anyone willing to see if the schematic is ok now before I order parts?

[rdjones]

Anyone willing to see if the schematic is ok now before I order parts?

I don't see anything that would keep it from working as discussed.
That's a lot of switches and caps, though.

You can use a single resistor for the "phantom PI" lower half, a single 220K.

rd