How do you get less distortion out of a tubescreamer?

[saultime]

I'm sorry if this has been covered 1000 times, but I want to get less distortion out of my tubescreamer. Most of the time I run it with the drive knob all the way down.

I came across this Robert Keeley page: http://www.robertkeeley.com/audiokt88/tonecap.html

...and this seems like a good way to go.

1. Which resistor is it that will lower the distortion?

2. If I change the resistor value, will this change the over all tone? I've already changed the .047uf with a .22, and really like the results. I don't want to change that.

Here's a picture of the actual PCB I'm using (from buildyourownclone) so you can direct me to the exact resistors...



Here's a picture of my screamer clone--I'm proud of the paint job, considering it's my first project.

[TryingToDo613]

Smallbear just posted a low noise opamp that will fit that socket. I'm hoping to try it in the buildyourownclone boards myself this week. they aren't expensive. I bet it sounds good. -ph

[SonicVI]

There's a 4.7K resistor that can be decreased to increase distortion or increased to decrease distortion.

[Connoisseur of Distortion]

there's a knob labeled drive that will increase or lower distortion...  

[travissk]

http://www.geofex.com/Article_Folders/TStech/tsxfram.htm

You're concerned with the clipping section, though the whole article is excellent.

I think the 4k7 resistor that SonicVI mentioned earlier is the one on the right half of the board rather than the left half (the one around the opamp). As he said, increase the value of that to decrease distortion.

Another option is to decrease the 51k that is in series with the drive pot. Changing that is essentially decreasing the minimum point of the drive knob.

[MartyMart]

Travissk is correct, if you make that 4k7 a 10k and the 51k a 22k, then
you'll have a much cleaner screamer with reduced max drive
and reduced drive on minimum gain

Marty.

[Melanhead]

hmmmm ... you could do what I did to my clone ... I had a switch to remove the diodes from the circuit essentially making it a booster... It's much louder but no clipping distortion

[brett]

Hi.
IMO reducing the 51k resistor is a cleaner solution.  
For 2 reasons:
1.  increasing the 4k7 to 10k lowers the gain (to half) at both the top and bottom of the range on the drive pot.  Reducing the 51k to 10k reduces the gain at the bottom end to 1/5th, but hardly affects the top end at all.  Ibanez should have built it that way.
2.  that 4k7 is in series with a cap, so increasing it to 10k will reduce the rolloff frequency of the pair to half what it should be.
cheers

[Mark Hammer]

The "ideal" solution depends on what else Saultime wants to do.  For instance, if the goal is to achieve less distortion without regard to overall output level or bandwidth, then either reducing the 51k or increasing the 4k7 resistor will do nicely.  If he wanted a hot but cleaner output, then changing the aforementioned resistors would not be a good solution.  Instead of lowering the river, the strategy would be to raise the bridge - in this case using perhaps a 2+2 diode complement to raise the clipping threshold.  If I'm not mistaken, the alledgedly desirable Landgraff distortion unit uses a similar type of diode complement to facilitate outputs that are both loud and only mildly colored by distortion.

As Brett rightly points out, the change in the 4k7 resistor would also change the low-end rolloff - which is so key to the TS-9's tone.  On the other hand, it drops the low-end rolloff when you increase the value of that resistor.  Some folks find this a desirable change (how many times have we heard complaints about the shrillness or squawk of a TS-9 and requests to eliminate it?)

Perhaps an "ideal" solution would be a mélange of several approaches.  First, changing the 4k7 resistor to maybe 6k8 would lower the low-end rolloff just a bit (500hz, so a wee bit more heft comes in).  Second, change the 51k resistor to 39k.  Again, not a huge change but it makes the minimum gain achievable 6.7 which is going to be clean and enough of a difference from bypass level.  With the retention of the 500k pot, you'll still be able to dial in gains of 80, which can snag you some grit.

Finally, change the 1+1 diode complement to 2+2.  If that seems too clean, go with 1+2.  Both will provide more clean headroom and less clipping when you do hit the limit.

Those changes might get you what you want.  On the other hand, there may still be some issues about whatrepresents the "best" gain pot taper for your needs.  Some folks want more selectivity at the high gain range, and some want more selectivity at the low gain range, and some want a decent spread over the whole range.

[saultime]

I want more selectivity at the lower overdrive range. I consistanly run my tubescreamer with the drive control turned all the way downâ€"and sometimes that’s still more drive than I want. I hardly ever turn it up over 1 o' clock.

It’s plenty loud enough for me though. And I like the EQ curve right now. I put the .047uf  cap on a DPDT switch with a .22 cap so I can switch between a “thicker” sound and the classic sound. I really don’t want to mess with the sound I have there.

I just want to be able to dial in less distortion without changing the overall EQ.

[Mark Hammer]

Then it sounds like you pretty much have the options nicely narrowed down.  First, drop the 51k resistor in the feedback loop down to 27k.  That will set your minimum gain at 6.7 instead of 11.8 where it is now.  If that's *still* too much, then try 22k.  The drop in maximum resistance from 551k to 522k won't make that much difference in terms of maximum gain - 118 vs 112.  If you can hear the difference you have my undying admiration.

The other factor to consider is the pot taper, as noted earlier.  Typically, a linear taper pot is used for the gain control, since that distributes the selectivity of the control relatively evenly over the whole range.  In your case you may want to try a log pot, where the first 2/3 of rotation will make finer differentiations of clean-vs-sort-of-clean than it does now.

[Mark Hammer]

Then it sounds like you pretty much have the options nicely narrowed down.  First, drop the 51k resistor in the feedback loop down to 27k.  That will set your minimum gain at 6.7 instead of 11.8 where it is now.  If that's *still* too much, then try 22k.  The drop in maximum resistance from 551k to 522k won't make that much difference in terms of maximum gain - 118 vs 112.  If you can hear the difference you have my undying admiration.

The other factor to consider is the pot taper, as noted earlier.  Typically, a linear taper pot is used for the gain control, since that distributes the selectivity of the control relatively evenly over the whole range.  In your case you may want to try a log pot, where the first 2/3 of rotation will make finer differentiations of clean-vs-sort-of-clean than it does now.

[vanhansen]

This may sound like too easy a solution but couldn't you just pull out the diode's altogether?  Wouldn't that decrease the distortion since they provide some clipping themselves?  Seems logical to me.

[Mark Hammer]

Most likely that would take away too much of the distortion.  AFAIK, the TS-9 is not like the Distortion+, where completely removing the clipping diodes still leaves you with a dirty signal.  It sounds to me like the request is to have *some* clipping, just not too much.  

Just as an aside, many folks seem to think the Blue Magic design from Jake Nagy fits that bill nicely.  Just as another aside, for years, I played a guitar with an onboard preamp.  Did wonderful things to my amp, but when I tried a TS clone, I hated it.  When I finally picked up a cheap second guitar with passive pickups and tried the TS, the gentler "push" that the TS received made it sound MUCH better to my ears.  Sometimes less IS more.

[aron]

I'm just scanning but you can put resistance in series with the diodes. That will do it too.

[petemoore]

This may sound like too easy a solution but couldn't you just pull out the diode's altogether? Wouldn't that decrease the distortion since they provide some clipping themselves? Seems logical to me.
 Diode threshold
 Gain setting
 V+ voltage 12? [components rating permitting]
 Resistance at input or between stages
 Actually since it's in the box I'm just typing to type...so, OT.
 interesting approach I think.
 Start with it as clean as it can get like a booster W/T.C.
 Then see what it's like at 9 or 7V, with many different thresholds and assymetrics [diode clip adjust pot..offers control on the clipping amount of the diodes as 'clipping central'] you might find that having two gain presets on an SPST, and top mounted clipping adjust to be useful.
 I have a 3 knob DIST+ with 'twangy bite' mod, see AMZ Lab Notebook.

[dev]

That is a very easy fix. You can use a jumper (a wire) in place of the 51K resistor at one end of the drive pot.  If you never max out your gain and you want more preciece control over the min gain, use a 250k or 100k pot instead (log tapper may help).  Remember with the gain at zero you still have at minimum a 51k resistance, so something less than that is what you want.  If you ever want to do the dual cannel overdrive thing you can switch in another pot in series with your smaller drive pot for extra grit and noise at times (go crazy here 1M).  Shure it's been done but so has pizza and I don't hear anyone complainig.

good luck,
Dev

[Melanhead]

This may sound like too easy a solution but couldn't you just pull out the diode's altogether?  Wouldn't that decrease the distortion since they provide some clipping themselves?  Seems logical to me.

that's what I said too ... No-one listens to me

.... I figure it could be set up to boost the amp and get that little bit of drive from it rather than the pedal, as long as it's a tube amp he's driving...