Newbie falls down the 4049 rabbit hole...

[ThermionicScott]

Hey folks, I'm Scott!  Mostly a tube guy, but since even a 5-watt tube amp is loud when cranked up around the house, I've been dinking around with my pedals lately.  And decided pretty quickly I wasn't happy with the sounds I was getting from them.  I tried a few tone circuit tweaks on the Boss Metal Zone, and am on the "Diezel mod" currently.  I widened the midrange hump on my Tube Screamer and now it shares a lot of overlap with my OD-3.   

I like guitar sounds in that zone where light picking yields a pretty clean sound, but digging in on a note gives a little scream and strumming a chord yields a nice crunch.  Think Malcolm and Angus in their early years.  But that setup always seems to bring out an abrupt transition in all my solid-state distortion pedals, even the ones with "soft clipping."  So I've been on a quest to find something with a smoother, less distinct transition between states.  The Blackstone overdrive has really impressed on the demos I've watched, but there's a lot going on under that hood, so I decided to go back to the beginning of that thread and try building a Tube Sound Fuzz first:



(It's almost exactly this circuit, but without the "lead" option:  http://3.bp.blogspot.com/-leeS_gB54pI/UXMQJmrknBI/AAAAAAAAAmw/aYVmIaWG3L0/s1600/Screen+Shot+2013-04-20+at+6.00.13+PM.png )

I'd never tried a solderless breadboard before, and I'm sold on this thing for prototyping.  To my amazement, it worked the first time!  Some sonic impressions and a couple questions to follow.

[willienillie]

Think Malcolm and Angus in their early years.

No pedals, just straight into plexi Marshalls (supposedly some with KT66s).  If you come up with a solid state pedal that sounds like that, you'll be a rich man.

The closest thing available?  I don't know.  Maybe a Marshall Guv'nor, but I've never tried one.

[ThermionicScott]

Think Malcolm and Angus in their early years.

No pedals, just straight into plexi Marshalls (supposedly some with KT66s).  If you come up with a solid state pedal that sounds like that, you'll be a rich man.

The closest thing available?  I don't know.  Maybe a Marshall Guv'nor, but I've never tried one.

Oh yeah, I'm under no delusion that I'll be the one to capture that whole sound in a pedal!     Just seeing what tinkering I can do to get closer to that, ermm, "envelope."

[Ben N]

For me, the best edge of breakup tones from a pedal seem to come from JFETs--ROG Peppermill, Fairfield Barbershop, and so forth. But JFETs are a pain in the ass, especially now that that the through-hole ones are near extinction.

[iainpunk]

For me, the best edge of breakup tones from a pedal seem to come from JFETs--ROG Peppermill, Fairfield Barbershop, and so forth. But JFETs are a pain in the ass, especially now that that the through-hole ones are near extinction.

i know, i am saving up this month and next to stockpile different jfets so ill have enough the rest of my life.
already did that with tubes a few years ago, i got a bunch of PCC82 PCC83 and PCC85's stockpiled, since the P series is harder to find and they have 2x4.5V // 9V heaters.

i would recommend jfets over the 4049 for 'edge of breakup' sounds, if you get them to bias just right. asymmetry also helps with that edge of breakup.

cheers, Iain

[ThermionicScott]

For me, the best edge of breakup tones from a pedal seem to come from JFETs--ROG Peppermill, Fairfield Barbershop, and so forth. But JFETs are a pain in the ass, especially now that that the through-hole ones are near extinction.

i know, i am saving up this month and next to stockpile different jfets so ill have enough the rest of my life.
already did that with tubes a few years ago, i got a bunch of PCC82 PCC83 and PCC85's stockpiled, since the P series is harder to find and they have 2x4.5V // 9V heaters.

i would recommend jfets over the 4049 for 'edge of breakup' sounds, if you get them to bias just right. asymmetry also helps with that edge of breakup.

cheers, Iain

Hey, I've got boxes and boxes of tubes with oddball heater voltages (it tends to be all-or-nothing when you buy out an old TV/radio repairman's stash) -- want me to look for anything? 

[Ben N]

Thermionic indeed!

[duck_arse]

Thermionic indeed!

that guy! great scott, he was nothing but trouble.

[ThermionicScott]

Thermionic indeed!

that guy! great scott, he was nothing but trouble.

I've spotted a "thermionicx" in some past threads.  No relation, I assure you!   

[iainpunk]

For me, the best edge of breakup tones from a pedal seem to come from JFETs--ROG Peppermill, Fairfield Barbershop, and so forth. But JFETs are a pain in the ass, especially now that that the through-hole ones are near extinction.

i know, i am saving up this month and next to stockpile different jfets so ill have enough the rest of my life.
already did that with tubes a few years ago, i got a bunch of PCC82 PCC83 and PCC85's stockpiled, since the P series is harder to find and they have 2x4.5V // 9V heaters.

i would recommend jfets over the 4049 for 'edge of breakup' sounds, if you get them to bias just right. asymmetry also helps with that edge of breakup.

cheers, Iain

Hey, I've got boxes and boxes of tubes with oddball heater voltages (it tends to be all-or-nothing when you buy out an old TV/radio repairman's stash) -- want me to look for anything? 

nah no need, i have well over 400 tubes of about 100 different types in storage. even some röntgen tubes, mercury vapor rectifiers, a deuterium lamp and a lot of TV tubes.

cheers, Iain

[ThermionicScott]

Hey, I promised sonic impressions and some dumb questions, so here goes:

I like the subtle transition to breakup on the TSF "rhythm" setting.  You can hear the signal get thicker and woolier as it approaches the clipping point, and then the decay reverses that.  Just what I've been wanting in that area, we're on track.  Turning up the gain a little, power chords get more and more dirt, but there's something I don't like about the sound at this point.  For a lack of better words, a buzzy character in the distortion for the lower strings that has been present to some degree in all of my other drive pedals.  To be fair, is this just how distortion is when you are barely past the clipping point?  It's been a while since I've had the occasion to use really loud amps with friends.   

Switching the 1M pot to a 10M resistor for the "fuzz" setting is fun.  At least with my guitars, it's not a super-saturated endless sustain kind of fuzz, but the name is accurate.  I wonder if the "Tube Sound Fuzz" name is really missing some punctuation.  You've got "tube sound" on the rhythm setting, and "fuzz" on the other.

I'm glad I didn't try to immediately build it into a pedal, since I'd like to play with a few more circuit variations first.

So here's a generic inverter stage that could be the second stage of the TSF/Red Llama:


* I know (or at least assume) that C1 affects the low-end response, but how do you calculate the corner frequency when there isn't a resistor to ground (or positive voltage source)?  Ditto for C2 in the feedback loop.  Do they interact at all?

* R1 is used to set the gain, and I'm beginning to wrap my head around how this is done with op-amps.  But here again, how do you calculate the gain when there isn't a resistor in series with the input? 

* Bonus question:  In tube guitar amps, we tend to shy away from negative feedback except as needed since this linearizes the signal.  Instead, we'll typically use cookie cutter circuits for amplifier stages with zero negative feedback, and then cut down the level with pots or resistor dividers if need be.  Why isn't this done more with SS distortion pedals?  Doesn't all that feedback contribute to the abrupt clipping thresholds that we then try to "soften"?

Thanks.  I'll keep doing searches in the meantime.   

[Rob Strand]

For a lack of better words, a buzzy character in the distortion for the lower strings that has been present to some degree in all of my other drive pedals.  To be fair, is this just how distortion is when you are barely past the clipping point?

It's normal to get a more buzzy character when you clip harder.

You can use a low-pass filter to shave off some of the nasty harmonics.


The low-pass filter cut-off frequency is f3 = 1/(2*pi*R*C).    You can choose R = 10k but the best value depends on the circuit.
An f3 of 6kHz to 7kHz will just shave off some nasties without affecting much.  For R=10k that gives C = 2n2 to 2n7.    If you like a bit more of a darker sound then perhaps 4kHz to 6kHz; for R = 10k that's C = 2n7 to 3n9.     Anymore than that has a deliberate darkening, which some people like.

So here's where some practical issues come in.   Most of the tube-sound fuzz circuits us a 10k output level pot.   That stuffs up the low-pass filter.  Also if you add a 10k low-pass filter resistor between 4049 output and the 10k level pot the level will drop.  So if you want to keep the 10k pot it might be better to use 100k for the low pass filter resistor and place the filter after the 10k level pot.  In this case all the cap values will decrease by a factor of 10.   An alternative is to use a 100k level pot and a 10k low-pass before the level pot.

You might find a fixed low-pass is enough once you pick the right filter cut-off to suit you.

You can also make the filter variable, this way will need a 100k level pot and say a 10k tone pot.



The other way is to put the tone pot after the 10k level pot, say a 100k tone pot.


Probably best to try a simple low-pass filter add-on and see if you can get closer to what you want.

[ThermionicScott]

Thanks for replying, Rob.  It doesn't seem like high-frequency content is the issue, though.  I plugged in again and tried playing with the guitar's tone knob rolled all the way off, and if anything, the "blatting" on the low notes was even more prominent.  If this link works, it's a crappy sound clip, where the first ~9 seconds is with the tone control all the way up, last ~12 seconds is with the tone all the way down:

https://imgur.com/G4JpozG

Leaving the tone knob up on the guitar and turning down the treble at the amp instead yields about the same result -- blatty sound that's also dark.

Two things I wonder are:
1) Is this from too much bass in the signal?  I haven't implemented any low- or high-cut yet.
2) Or is it clipping too hard?  It seems to me that the TSF really tries to wring as much gain as it can from its two stages, and that causes the first stage to really wallop the second one.

[anotherjim]

What! Imgur hosts soundclips? When did that happen?
Anyway, that sounds like inverter distortion to me. I put the "fizz" down to the dynamic gain of the inverters. It's hi-gain in the signal zero crossing (and brings up lots of harmonics & noise) and low near the peaks which is the cause of the compression effect. I think the fizzyness is due to the cyclic boost in low level detail at every zero-crossing (and that what I see when I put a 'scope on it. A way to tame that is to reduce the feedback resistor for less gain and drive the input harder with a clean boost to accentuate the compression.

To answer an earlier question, when there is no input resistor, the input impedance is the source impedance including whatever the input cap happens to be worth at the frequencies it's given.
If you want less treble out of it, increase the feedback cap and it's the same formula for an opamp low-pass for the corner frequency.

[Rob Strand]

For some reason the sound clip didn't play for me.

[willienillie]

For some reason the sound clip didn't play for me.

It defaults to muted.  Volume control in the top right corner.  Odd place, took me a minute.

What! Imgur hosts soundclips? When did that happen?

It's a video.

[ThermionicScott]

What! Imgur hosts soundclips? When did that happen?

It's just a cellphone video, with the camera facing the top of a box so it wouldn't see anything.

Anyway, that sounds like inverter distortion to me. I put the "fizz" down to the dynamic gain of the inverters. It's hi-gain in the signal zero crossing (and brings up lots of harmonics & noise) and low near the peaks which is the cause of the compression effect. I think the fizzyness is due to the cyclic boost in low level detail at every zero-crossing (and that what I see when I put a 'scope on it. A way to tame that is to reduce the feedback resistor for less gain and drive the input harder with a clean boost to accentuate the compression.

That reminds me:  I need a 'scope!  I'm sure that would show all kinds of useful info when tuning these things.  Do you folks recommend any particular desktop apps for 'scoping sound files?

I know Mark Hammer (and Mr Blackstone, of course) much prefers using a clean boost from an op-amp in front of these things.  I'd been curious why, so I'll be trying that before long, too.   

To answer an earlier question, when there is no input resistor, the input impedance is the source impedance including whatever the input cap happens to be worth at the frequencies it's given.
If you want less treble out of it, increase the feedback cap and it's the same formula for an opamp low-pass for the corner frequency.

Thanks!  I'll do some digging, might have follow-up questions on how to find those impedances.

[Rob Strand]

It defaults to muted.  Volume control in the top right corner.  Odd place, took me a minute.

Thanks.   I actually get a black (dark grey) screen with nothing on it at all!
I have seen that before on that site, at the time I assumed the content was removed or expired.

[Rob Strand]

That reminds me:  I need a 'scope! 

If you have a multimeter measuring the DC voltage on the supply rails and the outputs of the 4049 gates could be useful.

[ThermionicScott]

It defaults to muted.  Volume control in the top right corner.  Odd place, took me a minute.

Thanks.   I actually get a black (dark grey) screen with nothing on it at all!
I have seen that before on that site, at the time I assumed the content was removed or expired.

Should look like this.  A little light did get to the camera when I was recording...