Please Peruse This Schematic & Tell Me What You See?

[johnDH]

No prob! Here's what I got:

View attachment 925162

It just models the passive parts. I'm assuming the output impedance of the stage before is set by R1, at 100k?

The lower blue low-pass trace would get boosted up by the following gain stage, whereas the high-pass stage looks to just have a buffer after it.

Its easy to run variations, so let me know if youd like to see any!

cheers J

I looked up a tube stage calculator online, which predicted that with 12ax7's the gain before the filters, and also the one on the low pass, will have an output impedance of 68k, and a gain of 30db. Ill try to set those up together with the mixing stage ,to see how the two signals combine. The low pass amp stage is inverting too, so phase and phase shift is going to be significant

 

[The Ballzz]

@Bendyha
In reference to the quote you have at the bottom of your posts: Harlan Ellison was a brilliant individual and it is refreshing to know of someone who even knows who he was! Some of his observative visions could/should be re-visited, to possibly and partially inspire a way out of the mess that humanity has created for and/or of itself! You're obviously no dummy either!
I Appreciate Your Input,
Gene

 

[johnDH]

This plot has the active stages in, represented by equivalent opamp circuits, plus the two tone pots. The red and the blue are the raw versions of each tone, and the in-between curves are sweeping the pots one up and one down. So the mix is like 50:50, and there'd be a louder version if both were at max:

 

[andrewRneumann]

Ahhhh, but! How much gain do you guesstimate gets recovered going through the added gain stage of V2B, just before the Master Volume?
Just Askin'?
Gene

V2B has a gain factor of approximately 32. Let's say your guitar is strumming along at 0.1v. V1A brings that up to 3.2v. After splitting and filtering and recombining, that's back down to about 1v. After V2B amplifies it, it's about 32v, which is plenty to overdrive the phase inverter and output valves if desired.

 

[andrewRneumann]

This plot has the active stages in, represented by equivalent opamp circuits, plus the two tone pots. The red and the blue are the raw versions of each tone, and the in-between curves are sweeping the pots one up and one down. So the mix is like 50:50, and there'd be a louder version if both were at max:

View attachment 925253

That looks very similar to what I posted in #23. There does not appear to be any notching or scooping. When the controls are matched, whether at 25%, 50%, or 100%--the frequency response is very even.

 

[johnDH]

That looks very similar to what I posted in #23. There does not appear to be any notching or scooping. When the controls are matched, whether at 25%, 50%, or 100%--the frequency response is very even.

Yes, and I didnt even see that you had done that before! But, our two models seem to be almost identical, even down to how we both made engineering assumptions to get the result with the tools available

 

[Bendyha]

@Bendyha
In reference to the quote you have at the bottom of your posts: Harlan Ellison was a brilliant individual and it is refreshing to know of someone who even knows who he was! Some of his observative visions could/should be re-visited, to possibly and partially inspire a way out of the mess that humanity has created for and/or of itself!..........
Gene

Coincidently, Harlan Ellison's novel, "I have no mouth, and I must scream" was presented as a radio drama last weekend, on bbc. If you fancy a half-hours' entertainment. It was recorded in 2002, and stars Ellison himself...well worth a listen. I was in the workshop making a roller bridge for a lap steel when it played.

 

[andrewRneumann]

Yes, and I didnt even see that you had done that before! But, our two models seem to be almost identical, even down to how we both made engineering assumptions to get the result with the tools available

For some reason, the Falstad simulator will only simulate pot rotation from 0.5% to 99.5%. It seems miniscule, but I think that is the main reason our results are different. That last 0.5% seems to make a big difference!

Are you using a free program?

 

[Lynxtrap]

Let's not forget what matters most: How does it sound?

 

[The Ballzz]

Let's not forget what matters most: How does it sound?

Well Sir,

First, I must say that while I understand the operational (actually putting sound/guitar through the preamp) significance of the kindly provided research and graphs, you folks are miles above my technical understanding. I truly appreciate all of the fabulous input towards helping this relative hack (me) through understanding the implications. While my bench and fabrication skills are good, I can follow a schematic, but I get lost on what affect component value choices make in a circuit. Rather than spending my last bit of time (I'm a fairly old fart) studying to learn all the ins & outs of electronics theory, I simply want to play my guitar through unique amplifiers (not commercially available) that I have the skills to "physically" build with a bit of technical assistance in the design process.

That video is a good example of why I want to "attempt" to recreate such a beast, but in a vastly lower wattage package, with say, a pair of 6V6s. While I'm well aware that some of the qualities are from the massive amount of power provided by the power amp section, it seems that much of the uniqueness is created in the preamp. There is a little bit of out of phaseness and partially cocked wah sound that I'm attributing to the already discussed phasing issues in that preamp. There is a magically raw character to the sound! Of course, Johan's great skills

I'm guessing that plate voltages in the preamp and PI are fairly high and also guessing that keeping that voltage high, while having suitable voltage for the 6V6s may be a bit of a challenge?

Thanks For All The Help,
Gene

 

[andrewRneumann]

I'm guessing that plate voltages in the preamp and PI are fairly high and also guessing that keeping that voltage high, while having suitable voltage for the 6V6s may be a bit of a challenge?

My 2 cents:

I don't think it will be a challenge! 2x 6V6s won't need as much voltage swing to overdrive them, although they present a heavier load with 100K grid leaks. You might even get too much 6V6 overdrive for your taste.

I have found that a filter choke for the preamp section is a great way to filter ripple and maintain high voltage. They really aren't that expensive and are small because the preamp doesn't draw a lot of current--you can tuck one in almost anywhere.

If I gather, most of the distortion and overdrive is coming from the PI and power tubes. To drive the power tubes hard, you want to keep the plate voltages on the PI high. It's kind of a balancing act though and might require some experimentation... increase the PI plate voltage and you'll get less PI overdrive and more power tube overdrive. Decrease the PI plate voltage and you'll get more PI overdrive and less power tube overdrive. The original Pig may just have found the sweet spot!

Good luck and please update us on your progress if you start this build. If you want to emulate the PI / power tube gain structure of the 2x KT66 in the OP, but use 2x 6V6s, you might want spend some time learning about load lines and estimating headroom. Or... just build it and start experimenting with different circuit resistances until you find your personal sweet spot.

 

[johnDH]

For some reason, the Falstad simulator will only simulate pot rotation from 0.5% to 99.5%. It seems miniscule, but I think that is the main reason our results are different. That last 0.5% seems to make a big difference!

Are you using a free program?

I think there's no fundamental differences in the models. Before, I had 100k pots and now i look again and I see they are 1M. I also have a higher output Z on the buffer circuit than on yours. I see the value you have and agree, but added something for the output being not to the cathode but 1.5k lower. Would that make a difference? I guessed but I dont know.

I'm using a free Spice program called 5Spice. Its been withdrawn now but I'm used to it and it does what I need. The free version lets you work on a panel area as you can see on my picture, and the small area was the limitation for the free version. But you can squeeze as much as you can into it, which is great for looking at guitar and stompbox circuits, parts of amps and I used it to design my attenuators.

But I think most people use LTSpice for a free program, which is more versatile but more clunky, like a 1990's code. 5Spice looks about 10years later with handy drag and drop features. Id expect new codes would share your design data on Facebook, draw animated circuit diagrams in 3d and send ransomware to your Grandma.

Here's the sweep I get now:

To compare he models, I had a look at a setting with full bass and minimum treble, and looked at the differences between the signal levels at 200 and 2000 hz. The difference, about 26db, is very close to that on your model, and the not quite full 0-100 range would explain a slight difference.

So Gene @The Ballzz , setting aside the analysis, what I think we see here is a pretty good well designed tone circuit. There's a very flat level response available at any setting where bass control = treble control, and you can tweak from there.

The frequency at which the two controls seem to meet looks to be about 200hz.

On my chart, there's a line for 50% 50%. You can turn them both up to 100% for the same curve 6db higher.

if you wanted to tweak it, you can for example, reduce all caps from 4.7nF to 2.2nF, and now that 200hz point shifts up to about 430-450 hz, and nothing else unexpected seems to happen (I tried it)

I don't think we are seeing significant phase cancellation effects. It seems that the phase changes between the two circuits are mostly negated by the inverting stage on the bass side.

But I think there is a significant difference from this to most amp TMB tone circuits on Marshalls and Fenders etc. While this PIG is giving a very well considered potentially flat response, the tone circuits in more common amps do not seem flat, as shown with tools such as Duncans Tonestack Calculator. Here's a plot for a Marshall tone stack (Fender ones look more scooped):

 

[The Ballzz]

@johnDH
I installed TSC on my Windows machine, but rarely even fire up that computer, for multiple reasons. There is however another nearly identical one online called TSC in the web with some slightly different features, with some added devices:

https://www.guitarscience.net/tsc/m...3=22n&RB_pot=LogB&RM_pot=Linear&RT_pot=Linear

What is very surprising is what settings are required to achieve a flat response. It should be noted that we can't assume the "slider" positions in the middle (5) to be flat as on a graphic EQ. And although many players don't really want a "flat" response, the way I end up with the contols on many Marshalls they end up pretty close, according to the TSC.Definitely the mids considerably higher than the bass and treble. It can be fun to play around with and it's also pretty cool how just changing the value of the pots can make such dramatic differences!
Just Noticin'
Gene

 

[Lynxtrap]

That video is a good example of why I want to "attempt" to recreate such a beast, but in a vastly lower wattage package, with say, a pair of 6V6s. While I'm well aware that some of the qualities are from the massive amount of power provided by the power amp section, it seems that much of the uniqueness is created in the preamp. There is a little bit of out of phaseness and partially cocked wah sound that I'm attributing to the already discussed phasing issues in that preamp. There is a magically raw character to the sound! Of course, Johan's great skills

I'm guessing that plate voltages in the preamp and PI are fairly high and also guessing that keeping that voltage high, while having suitable voltage for the 6V6s may be a bit of a challenge?

Thanks For All The Help,
Gene

The "magically raw character" in the video is no doubt the output stage being driven hard, partly (the way I see and hear it) as a result of low filtering in the power supply.

2x6V6 will not sound exactly like 4xKT88, but I think you will be able to capture quite a bit of the character of it by selection of component values. It does have a bit of "maxed Tweed"/Neil Young character to it.

It will still be loud though. I still think a post phase inverter master volume might be a nice feature to get some overdrive out of the PI at lower volumes.

 

[The Ballzz]

The "magically raw character" in the video is no doubt the output stage being driven hard, partly (the way I see and hear it) as a result of low filtering in the power supply.

2x6V6 will not sound exactly like 4xKT88, but I think you will be able to capture quite a bit of the character of it by selection of component values. It does have a bit of "maxed Tweed"/Neil Young character to it.

It will still be loud though. I still think a post phase inverter master volume might be a nice feature to get some overdrive out of the PI at lower volumes.

I use a really stellar attenuator design, pioneered by the brilliant @johnDH that allows the use of any amp in it's "sweet spot" at nearly any volume required.Of course, by the time you get down to TRUE bedroom levels, all bets are off, but still the best out there! I've already built two that are installed in two of my amps (5E3 Tweed Deluxe & DSL20CR) and am working towards building as many more as are required for every amp I own and/or plan to build! I'm currently laying one out to fit into a smallish head cabinet for my "someday to be finished" marriage of a Low Power 5E8A Tweed Twin preamp to a 5E3 power section! The @johnDH attenuator design is the greatest thing since sliced bread and by far the best and most liberating pieces of gear I've ever owned, in 50+ years of "Screechin' & Squawkin'" with guitars through amplifiers! I've not yet met any version of pre or post PI master and even power scaling that didn't mess with tone, dynamics and feel more than the @johnDH attenuator design. John has been very humble and courteous about keeping most discussion of this secret weapon to one specific, VERY LENGTHY thread over at the Marshall forum! More folks should give his design a try! It ain't your Grandpa's attenuator!

However, I do agree that during at least the building and testing stages of the LITTLE OINKER, it might not be a bad idea to include both pre & post PI masters, for finding just the right balance of PI and 6V6 overdrive! Thanks again for the encouragement and participation in my madness (oops, I meant "thought") process.

I used to think I hated 6V6s as being sometimes too mushy and fuzzy and other times too harsh and brittle. Then I realized it was because many run them at too low or too high voltages, and filter them too much. It seems there is fine line drawn at about 365vdc to 375vdc and between just enough filtering and too much. At least to my ears!

And yes, I'm just a layman/novice/hack at amplifier design, but I know what "Screeches & Squawks" the best!

John Is My Hero!
Gene

 

[drmordo]

Here is the full Jaffe article that Lynxtrap's post#14 refers to, that Mashall is said to have based the idea on.

View attachment 925190
View attachment 925191
View attachment 925198

It would be really cool to convert this to transistors and either box it as a stompbox eq or add a distortion/fuzz to the end of it.

 

[Lynxtrap]

It would be really cool to convert this to transistors and either box it as a stompbox eq or add a distortion/fuzz to the end of it.

There is a solid state version floating around the net.
http://www.hobby-circuits.com/circuits/audio/tone-and-volume-control/930/paraphase-tone-controller

@The Ballzz A good attenuator should do it! Better than a PPIMV as output stage overdrive is the main thing here.

 

[The Ballzz]

There is a solid state version floating around the net.
http://www.hobby-circuits.com/circuits/audio/tone-and-volume-control/930/paraphase-tone-controller

@The Ballzz A good attenuator should do it! Better than a PPIMV as output stage overdrive is the main thing here.

As a relatively ignorant fool, how else can one (at least temporarily for testing purposes) "adjust" the balance of how much overdrive is being created in the phase inverter, vs power tubes? IE: How hard the PI is driving the power tubes? Is it done by changing the values of the tail resistors of the PI?
Just Askin'
Gene

 

[andrewRneumann]

As a relatively ignorant fool, how else can one (at least temporarily for testing purposes) "adjust" the balance of how much overdrive is being created in the phase inverter, vs power tubes? IE: How hard the PI is driving the power tubes? Is it done by changing the values of the tail resistors of the PI?
Just Askin'
Gene

Maximum clean output from the LTP PI can be obtained by using a smaller tail resistor and/or larger anode load resistors. There is a limit though… hard to say where it is… where the tail is too small (and/or plate resistors too large) and total output drops. The opposite it true if you wish to lower the output of the LTP.

You can also hijack the output of the LTP by decreasing the grid leak resistance in the power tubes to bleed off signal. This would have the added benefit of reducing bias recovery time in overdrive.

A more power tube overdrive oriented sound might also be obtained by trying a lower ra (anode resistance) or lower mu set of valves for the LTP.

All of these changes would affect the tone of the amp as harmonic content would change. Lots of options, no hard answers that I can see.