NFB/Presence control for a Cathodyne PI build?

[Gainzilla]

Hi All,

Long time lurker, but new member. I've learned a ton of stuff here so far, so thanks for everything!

Ok, here is my issue: Due to a shortage of available triodes for the features I want for a new build, I'm considering using a cathodyne PI. The issue with this is I'm not sure how to implement NFB and Presence controls.

I've seen Orange and Fender (princeton I think) put the NFB on the cathode of the gain stage right before the PI, so that gives me a starting point. Still, I'm unsure if the usual values work and what the sonic differences are.

Then there's the presence. I'm at a loss here. Any ideas would be greatly appreciated.

BTW, the basic topology is sort of a 15 watt, cathode biased (EL84) Naylor Super Drive.

Thanks in advance!

[J M Fahey]

... shortage of available triodes for the features I want for a new build

Not sure about what you mean.
You wonīt use a triode for your cathodyne?

I'm considering using a cathodyne PI.

You are choosing an inferior design, with lots of problems.

The issue with this is I'm not sure how to implement NFB and Presence controls.

Here you are.

I've seen Orange and Fender (princeton I think) put the NFB on the cathode of the gain stage right before the PI, so that gives me a starting point. Still, I'm unsure if the usual values work and what the sonic differences are.

Then there's the presence. I'm at a loss here. Any ideas would be greatly appreciated.

*If* any of those amps has some killer sound which you want, just build any them by the book.
This is *old* technology, settled 50 or 60 years ago, not much unexplored territory to discover.

BTW, the basic topology is sort of a 15 watt, cathode biased (EL84) Naylor Super Drive.

Why not just build one?

With the 15W power amp, of course.

Personally I wouldnīt want to reinvent the wheel, unless I had very solid reasons to do so and enough background to be able to pull it successfully.

But just to be different .... mmmmmhhhh.

[Gainzilla]

Hi Juan,

Thanks for taking the time to comment!

My design looks like this: Gain stage (switchable) --> Gain stage --> Gain Stage --> Cathode Follower --> Tone Stack --> FX loop send --> FX loop receive --> PI --> Power Amp. That's 8 preamp triodes, by my count. I currently only have 3 preamp tubes and 2 EL84s and don't really have room (or enough available current from the PT) for an extra tube. In order to build this amp, given my current constraints/budget, one solution I'm considering is to implement a mosfet Cathode Follower to drive the tone stack, and try a Cathodyne PI. Alternately, I could try a passive FX loop, but I'm not sure how that will play with my effects (Delay pedal mainly). If I do that, I can have a standard LTPI. Then at least I'm in more familiar territory.

I'm not sure if it's an inferior design. However, the Princeton, Deluxe Reverb, and many Orange amps all seem legit. YMMV, of course.

In general I agree with you. Hell, why don't I just buy the amp that I want? Because I enjoy the quest. I used to play professionally, but now I have young kids and a "real job", and am currently not gigging. I need an outlet or I'll go freaking insane. Also, it's pretty hard sneaking $3k past the wife.

I'm not sure I agree about the unexplored territory, however. I'm a designer professionally, and in the world in general, most innovation happens by standing on the shoulders of giants. Combining known ideas into something new, or highly useful, or that solves some new problem. I don't see tube technology as being any different. Sure, from a physics standpoint, or even an amplifier topology standpoint, we are mostly dealing with designs created in the 40s. But when you combine those "ingredients" in new ways, or stir in new technology, Innovation can happen. I had no idea I needed heated car seats until I had them.

So here I am. Fun little side project. I want to see how far I can get with this. I could try and squeeze in another tube, but my PT would be running beyond it's spec. Maybe that's ok, though. Thoughts?

Cheers!

[Tom Phillips]

...I want to see how far I can get with this. I could try and squeeze in another tube, but my PT would be running beyond it's spec. Maybe that's ok, though. Thoughts?

It can be a fun challenge to see what you can create out of the parts you already have available. However, my thoughts are that you will be happier with the results if you scrounge or buy parts more suited to your needs. (Like the PT) You can do that for way less than the cost of buying a new fancy amp. Post some examples of what you are after. A schematic or a block diagram will be much easier to follow than a written dialog description of the amp. Another comment is that you seem to be focusing on the limitation of your parts rather than the ultimate sonic capabilities you want. If you tell us more about the tone goals then someone may be able to suggest some existing projects that you can use as a basis for your new amp. You can always tweak the design but you don't need to start from scratch with a new design
Cheers,
Tom

[Richard]

A cathodyne PI will not drive your output stage as hard as a long tail pair, whether or not this is an issue is up to you. You can add the presence control anywhere after the resistor (between the transformer output and the other end) in the NFB loop. I'm surprised I never noticed that the NFB is tied to the cathode of the previous gain stage in the Princeton. Does anyone know if the OT outputs can be reversed so that the NFB could be tied to a different stage (for proper phasing)?

[Tom Phillips]

...Does anyone know if the OT outputs can be reversed so that the NFB could be tied to a different stage (for proper phasing)?

Yes. The leads of the OT can be reversed if you need to change the phase of the feedback signal for any reason including to use a different injection point.

[frus]

A cathodyne PI will not drive your output stage as hard as a long tail pair, whether or not this is an issue is up to you.

you only need like 20 V peek to peek to drive the EL84's to the edge of overdrive, that's easy enough even for cathodyne. If you used, say, 6L6's you would need 90 Vpp and that's a bit hard on cathodyne

anyway, it would give you a different sound, different texture if you like, which you may or may not like

on the other hand, long tail has just a tad lower gain than an (unbypassed) gain stage + cathodyne (I'm thinking like 22 versus 30 or something like that), so probably you'd have enough gain, considering the EL84 are 3-4 times easier to drive than the big bottles

[Chuck H]

Regarding the presence control...

You can't really implement it in the same way as you would with the typical LTP/Bassman/Marshall type because, WRT applying NFB to an ungrounded preamp stage, bypassing with a cap to ground will both increase HF by removing it from the NFB loop AND by bypassing the cathode resistor for that gain stage. This may or may not sound good, but my first consideration is that reducing HF NFB has a particular sound and revoicing the preamp has another. It wouldn't be the singular effect of a presence circuit for better or worse. There IS a way to implement the usual presence effect without notably bypassing the gain stage. You could split the NFB circuit series resistor and tank highs from the loop there. The resistance to either side of the presence circuit should then be sufficient to avoid notable interaction with the output load or the preamp cathode voicing.

EDIT: The above info applies to an unbypassed cathode for the triode feeding the cathodyne PI. In the case of the Princeton, for example, there is a partially bypassed cathode circuit on that triode. In this case the impedances look to be suitable for implementing a typical presence circuit from the NFB entry node as usual.

[Gainzilla]

Thanks for taking the time to respond, everyone! Lots of great advise.

Ok, taking a step back here. I would love to build a 15 or so watt SuperDrive. But not necessarily for super crunchy distortion. I have other amps that can do that. What I'm missing is a really nice clean. Not squeaky clean, but slightly broken up clean, then roll back the volume a tad if needed. I'd also love a little grab and go amp that's also versatile enough to leave in the studio with a mic on it and dial up a range of tones. For me the naylor is a great starting point.

The loop is something I could compromise on. I could go passive for now, and add a dumbleator later, or use mosfets gor the send/recovery stage. But I might discover that the delay pedals work just fine in there. I'm operating under the assumption that I'll need those stages though.

Something else I'm considering is omitting the high gain stage. I could always hit tge amp with a pedal to push it over the top. Another thing I'm wondering is whether I really need the cathode follower to drive the tone stack. I was planning on trying a MOSFET there, though. Between that and the MOSFET I should have enough triodes to do the LTPI.

However, I'm not opposed to trying a cathodyne. I'm not looking for metal with this amp, and my thought was that with the el84s being easier to distort, and the cathode bias, maybe it would be ok for my goals...

Btw, my transformer set is a hammond 125E, and 169ex.

Thanks again!

[Gainzilla]

See the attached schem for reference:

[Chuck H]

This is only my very opinionated and subjective opinion, but...

If you will never be clipping the PI then a cathodyne type is fine. And with EL84's you're almost certain to have enough drive from any respectable preamp. If you will be pushing the PI such that it clips I think a LTP is going to sound better. As Juan mentioned, the cathodyne can be problematic for some things. In this case you'll likely have at least some preamp clipping, certainly some power tube clipping. Arranging the circuit gains to get this and NOT clip the PI at all could be a tricky dance, but it may be possible and I even think the results could be very good. I couldn't suggest a circuit or values without building it for myself and adjusting it by scope though.

[frus]

but you're not saving any triodes with cathodyne, so why use it?

[Chuck H]

There actually isn't a cathodyne in the latest proposed schematic. But I've seen a few designs where a cathodyne is used with EL84's without the need for a stage dedicated to drive the cathodyne because EL84's require so little drive voltage. In this particular case then it IS possible to save a triode. IME you can still lose some things. One is how hard you can drive the power tubes. EL84's do have a very nice, characteristic sound when pushed hard. If you can only get them to full power, but never hard clipping, you lose that. If you have a design that is a high gain preamp this isn't an issue since all the clipping is supposed to be from the preamp anyway. If it's a design that has hard padding, like the use of an effects loop, a cathodyne won't assist in gain recovery. So a dedicated driver would be necessary and then you DON'T save a triode. But you probably wouldn't need ALL the gain so it might be possible to dual purpose the driving triode.

[J M Fahey]

In fact triodes can be saved, by replacing less critical stages with Fets or Op Amps.
1) use a Mosfet cathode follower to drive the tone stack.
2) use a similar one to drive the effects loop send .
Bingo, one full 12AX7 less

That said, wonīt waste time demostrating it here, but cathodynes are the single *worst* type of PI when you overdrive the power stage, by far.

Princetons were designed to be used clean or at most very slightly crunching, when overdriven they are often sent to a Tech to solve a problem (will post scope screen captures later) which some Techs waste time trying to solve, more experienced ones brush them aside saying (correctly) : "thatīs the way they are" .

As of old Orange amps, I always thought that they were quite poor designs, with their James type tone controls, (more useful for Bass or Jazz guitars than for Rock) and .... you guessed it .... cathodyne PI .

As a whole they ended up sounding good (although more as a clean sound than overdrive) because they were built with high quality, expensive parts, such as Partridge transformers, and mainly the impressive Fane speakers (used also by Hiwatt) , much superior to any Celestion of that era, and never used by, say, Marshall or Vox because of the price.

[Gainzilla]

@ChuckH - thanks for the info. I might be inclined to try it just for giggles. The Phaez daisy cutter uses this topology as well, but this amp is optimized for preamp gain. The general tone about the cathodyne seems to be that it can work well in certain conditions. It feels reachable to me, but probably not easily at my level of understanding. BTW, I'm from the Northwest too (eastside). Cheers!

@Juan (frus... Pretty much everyone. Lol). Ok, you have convinced me. Tried and true LTPI it is! I was ready to drop the loop if need be. However, if the mosfets are truly an option, I'm not opposed to giving them a try. I've got some on the way from Mouser, but they won't be here until Monday.

I would plan on using a MOSFET for cathode follower like in the second attachment, and I guess the loop driver stage like the first attachment.

Any issues just replacing the loop stage with this one? These are non-critical stages, but should I expect any issues, totally or otherwise? I should only be getting around 250v there, which is far lower than shown in the schem.

Also, fwiw, I've never been a fan of the orange sound. They do have some decent high gainers now, but those all have LTPI. Still sound fizzy though. Lol

One observation: Man, this chassis is small. 14x6x2... It seemed pretty spacious in the previous SE build. The chassis was an overstock from Kitty Hawk that I bought maybe 12 years ago, and it was a universal head/combo chassis, so there are tube holes on the back and the bottom. Looks good in the cab, but take it out and it looks like Swiss cheese. Lol I bought all mil-spec resistors, Mallory 150 poly caps, so hopefully this should sound pretty good.