Quilter Zero crossing simulation

[Steben]

Hi Guys,
Long time no see....
SS emulation of "tube sound" is one of my favourite topics.
I stumbled upon the quilter / old pignose circuits and put the little zero crossing circuit in a spice.(nevermind the clipper, which is not quilter)
It is ment to simulate the dynamic crossover distortion.


The results when driving it with a 5V p-p signal and a 0.5V p-p signal are:

[Steben]

I've put a resistor accross the diodes and by making this variable one can alter the crossover width. Or in other words:it is a crossover effect control.

[elseif]

Coincidently, I did some LTspice modeling pf Pat Quilter's patent last week, though I got different results.  I was going to use this with a Nord C2D Hammond organ clone to get some 122 type of overdrive.  It needs fiddling.  With the values chosen, the coupling caps had to be large-ish to get decent low-frequency response.

[Digital Larry]

I'll be honest, I never heard of anyone trying to get crossover distortion.  Are there some "famous examples" of this sound?

[Gus]

Is this the patent the sim is based on?
https://patents.google.com/patent/US9008333?oq=9008333

I don't understand what is new in the linked patent.

[bool]

I built a ratnest demo for testing couple years ago, but found no use for it. It makes the sound to appear a little saggy and more overdriven than it really is. If anything, it made the tone prone to "dissapear" in a daw mix. I dissasembled it quickly.

[ElectricDruid]

There are a couple of basic problems with crossover distortion:

1) You get more distortion (relatively) at lower signal levels than at higher levels. This is the opposite of the effect we expect and doesn't map at all well to the natural way that "louder" tends to mean "more harmonics" and "softer" tends to mean "less harmonics".

2) As the sound dies away, it'll disappear below the crossover threshold. That's tantamount to deliberately adding a "splatty fuzz" fade out.

Not to say that these are reasons not to experiment with it, but they might be things you want to consider.

[Mark Hammer]

I'll be honest, I never heard of anyone trying to get crossover distortion.  Are there some "famous examples" of this sound?

One needs to distinguish between X-over distortion in a power amplifier, and X-over distortion in a pedal.  You're right in assuming that few pursue X-over distortion in a power amp.  However, I find it can sound pretty decent in a pedal with bridge pickups.  It certainly contributes to the legendary tone of the Boss HM-2 Heavy Metal pedal, is the basis for the ZVex Machine pedal, and I've used it on my Contrafrizz adaptation of the Gretsch Contrafuzz. (The X-over control is on the upper right hand side.  The X-over is courtesy of Ge diodes, while the clipping is silicon.  That's how I can still get X-over on top of regular clipping.).

[Digital Larry]

That's actually an interesting and subtle effect THAT I NEVER HEARD OF BEFORE.  Certainly I knew about crossover distortion, but only something to be avoided, as with all other things relevant to guitar amps that you SHOULD do.

As with overdrive, how smooth the corners are at the inflection points has a big impact on how harsh the circuit sounds.   I thought that sounded pretty good and an unusual effect in that it smoothly gets more distorted as the note decays.  Which is exactly what you'd expect.

- adjust straw in mouth
- hands on hips
- regional idiom in the general vein of "I'll be danged"

[ElectricDruid]

- adjust straw in mouth
- hands on hips
- regional idiom in the general vein of "I'll be danged"

Lol! exactly!

[PRR]

What, is this crossover week? From another hangout:

"Someone just asked me if it would be possible to "dial in" a variable amount of crossover distortion..."
https://groupdiy.com/index.php?topic=72632.0

[Elektrojänis]

While the other stuff is interesting too, I think many of you are missing the point of the OP. He is speaking of _dynamic_ crossover distortion. So The Crossover distortion is changing acording to the signal.

It is probably part of some tube amps sound especially when the poweramp is driven to saturation. Good explanation can be found here: http://sound.whsites.net/valves/analysis.html#s9

I linked to the part about the dynamic crossover distortion... The whole article is interesting read too. It's basically an analysis of the JCM800. While the writer doesn't really approve a lot of the design decisions in that amp, he seems to describe the workings of it in a very detailed way and we can use the information how we like.

While I have never played on JCM800 I can imagine that the kind of thinning of the note transients that this king of dynamic crossover distortion could produce might be actually beneficial for some type of harder rock tones.

[Steben]

correct.

At low levels and hardly any high amount of distortion, there is no crossover distortion.
Fact is, the Quilter circuit has a fixed reaction towards the input level. Adding a variable resistor from anode 1 to cathode 2 brings in a control.

[Steben]

I'll be honest, I never heard of anyone trying to get crossover distortion.  Are there some "famous examples" of this sound?

You're right in assuming that few pursue X-over distortion in a power amp.

The happy few 

Seriously, adding dynamic crossover is what happens in the circuit in front of the poweramp in Peavey's Transtube and in the Quilters.
It adds a kind of fuzzy hair on heavy played parts on top of clipping. Along with some loss in power (yes, a kind of compression). It certainly happens in many (older designed) cathode biased amp. But even fixed bias tube amps (plexi, jcm) have it, linked to blocking distortion ion the grid of the power tubes.

[bool]

And that "fuzzy hair on top" makes it to kind of getting lost in mix. I guess that the whole point of ... iconic (?) heavy amps (like mesas) was to "get to the point" of a damped palm mute with minimum loss and maximum gut-punch. I guess that's polarly opposite design philosophies. ymmv.

[Mark Hammer]

What, is this crossover week? From another hangout:

"Someone just asked me if it would be possible to "dial in" a variable amount of crossover distortion..."
https://groupdiy.com/index.php?topic=72632.0

Didn't you get the memo from the OECD?  I was able to get the "International Crossover Distortion Week" t-shirt and headband last month, in anticipation.  Of course, like all slogan apparel, it'll be on the discount rack at Dollar General (and Giant Tiger in Canada) in a few weeks.  Honestly, in the Twitter/Instagram era, great passions can be displaced and die so quickly.  You have to work hard to keep up.  But enough jibber-jabber.  I'm off to buy my "National Better Op-Amp Month" t-shirt.

[Steben]

And that "fuzzy hair on top" makes it to kind of getting lost in mix. I guess that the whole point of ... iconic (?) heavy amps (like mesas) was to "get to the point" of a damped palm mute with minimum loss and maximum gut-punch. I guess that's polarly opposite design philosophies. ymmv.

Of course! Mesas have tons of preamp again though. Completely different philosophy. But, if I may, a philosophy which brings tube closer to ss. Remember there is no wrong.

Fact is, a stepless control of X over in and out would add to the usefulness in a versatile amp.

[bool]

Ok, if you say so.. I tried that same trick you did (a trimpot across diodes, it's not rocket science, you can also trim the "load" that diodes "see") when I played with that circuit and my findings were that it is detrimental to what I want to hear as a "good tone". ymmv of course

A better way to add some fine "hair" to a (recorded) guitar track was historically to slightly load the channel (or a mic-pre) or to use a comp with very short attack/release (this will distort a little). In another thread here there was a discussion on driving a LTP bjt-pair and some comments on the supposedly good "hair" that such configuration provides. Which shouldn't surprise, because a discrete LTP-into-diffamp configuration is in like 95% of mixers out there; in most PA's and sound systems that used to "power" bands. Even modern "micpres on a chip" use that topology. My take is that our ears are accustomed to hearing that particular "hair" over the basic amp tone.

The question is are mic-pres really trully 100% "neutral clean" when you blast them with the signal from a dimed marsh/mesa/what have you. I know for a fact that engineers used to dime the pre gain when recording driven gtrs; not that this was needed for technical reasons.

[Steben]

https://www.youtube.com/watch?v=s2XeAEFSsNY
FFW to approx 1:20

[Steben]

Ok, if you say so.. I tried that same trick you did (a trimpot across diodes, it's not rocket science, you can also trim the "load" that diodes "see") when I played with that circuit and my findings were that it is detrimental to what I want to hear as a "good tone". ymmv of course

A better way to add some fine "hair" to a (recorded) guitar track was historically to slightly load the channel (or a mic-pre) or to use a comp with very short attack/release (this will distort a little). In another thread here there was a discussion on driving a LTP bjt-pair and some comments on the supposedly good "hair" that such configuration provides. Which shouldn't surprise, because a discrete LTP-into-diffamp configuration is in like 95% of mixers out there; in most PA's and sound systems that used to "power" bands. Even modern "micpres on a chip" use that topology. My take is that our ears are accustomed to hearing that particular "hair" over the basic amp tone.

The question is are mic-pres really trully 100% "neutral clean" when you blast them with the signal from a dimed marsh/mesa/what have you. I know for a fact that engineers used to dime the pre gain when recording driven gtrs; not that this was needed for technical reasons, resulting in the highest  amount of X over at lowest gain.

Depends on what you mean with "resistor accross diodes". The resistor I suggest controls the leakage the charge of the caps in front of the diodes see. It alters the max charge in a dynamic response. low level = no X over.
A resistor in parallel with back to back diodes in series with the signal controls the portion of X over, but again with the all or nothing response.