Wider gain range for Red Llama

[AM]

Oh yeah, I see what you mean. Thanks Doug.

[AM]

I built the circuit but I felt very experimental and did lots of substitutions. One of the most radical was to replace the 1M gain pot with a 500k one. This was done out of necessity though because I realized I had no 1M pots. I like how it sounds with the 500k pot. I can't explain it theoriticaly though. A pot is like a variable resistor. So by decreasing the value in half do I practically let less signal going through the pins that the pot is connected to and let more signal to ground? In other words is this a way to reduce the maximum gain of the circuit?

[MarcoMike]

I didn't really understand the "more current to ground" thing... in this circuit the gain pot is not acting as a voltage divider as the volume pot does... all the maximum/minimum gain story can be summarized as:

the more the resistance, the more the gain.

R2 sets the minimum gain, remove it to get the lower possible. the pot sets the maximum gain, use a 2M or more if you can find it to get higher gain at max. with a 500K pot your maximum gain is lower.

(I removed R2 on mine)

[AM]

I didn't really understand the "more current to ground" thing...

That's because my limited knowledge of electronics doesn't allow me to understand and explain a lot of things in a comprehensive way....

the pot sets the maximum gain, use a 2M or more if you can find it to get higher gain at max. with a 500K pot your maximum gain is lower.

That's exactly what I was trying to say. Frank, Doug and Mark had already answered my questions about controlling the minimum gain. For setting the maximum gain I could only assume that if you affect the lower gain by changing R2 you should be able to affect maximum gain by changing the pot. Thanks for answering. You gave me a theoretical backup in my empirical observations.

[AM]

I have put the pedal together and playing it for a couple of days.
Here are some the changes I did. Lowered C1 and C4 some. Increased C3 (almost doubled it). Decreased C2 just a tiny bit. Used an electro cap for C5. With the above changes it's definitely quieter than all the other Llamas I've built so far and retains the pick attack sensitivity this circuit is known for. I also lowered the gain pot value by half. Less max gain but that's fine for me since I didn't like the stock circuit maxed out. I like the sound so far a lot.

I also used a 150k input resistor instead of the 100k as R1 in the schematic.

ONE FINAL QUESTION:
Did I overdo it by placing that 150K as R1? I've never seen such a value as an input resistor in any of the overdrive circuits I've studied so far?
Normally I would just try a smaller value and not ask but I've soldered everything on vero so it's not very convenient unsoldering and soldering back. If you people think 150K is too much I have some 68K`s around. Would that be a huge difference though?

Edit: Schemo link so people don't have to search for it:
http://gaussmarkov.net/layouts/redllama/redllama-schem.png

[Mark Hammer]

"I like the sound so far a lot."

If it is not overheating, nor emitting magic blue smoke, no further justification is needed.

[AM]

No smoke of any color so far Mark  I was basically worrying if there is a potential of pickup loading, or similar implication by using an 150k as input resistance and also if anybody has any theoretical way to figure out what would be the practical difference between the 150K and a 68K. Unfortunatelly I can't figure something like this out without taking the pedal apart and solder away.

[AM]

Today I had the chance to play the pedal at loud volumes and through a couple of different setups.
I have a few observations:
First of all the level of noise has been dropped dramatically comparing to my previous builts of the same circuit. I guess raising C3 a lot and not using any ceramic or tantalum caps has contributed to that.
Chords ringing a bit clearer comparing  to the bypassed signal when playing on the neck position (Humbuckers at both bridge and neck guitar).
It's hard to explain that one. The circuit adds a tiny bit of high end that brings out the individual notes of the chords. Kind of adding some sparkle.

Now, I'm not sure whether this is true or my ears started playing tricky games on me after playing for quite a few hours but I have the feeling that it thins out the signal just a tad. I started having that feeling when rolling off my guitar's volume knob. I'm not sure if there is some bass loss or I perceive that extra sparkle I mentioned above as a bit of bass loss. I'm talking  about puting the pedal under the microscope here, so the differences between bypassed signal and signal going through the circuit with gain at minimum are very very minimal. I have reduced C1 and C3. I think I ended up with something like 0.047uF for C1 and 0.022uF for C3. (stock values were 0.068uF for C1 and 0.033uF for C3 for the Red Llama and 0.1uF and  0.047uF for the Tube Sound Fuzz). I still have that 150K as an input resistor. As I said I'm not sure whether that final observation about some bass loss is just me getting paranoid after test driving the pedal for hours or it's really there. It makes me wanna go for a bit more experimentation though. I want to get that circuit as close to perfection as possible for my rig. Do my cap value choises sound OK? I know I have more or less lowered the values by half from the Tube sound fuzz and by a third from the Llama but is that difference on the verge of keeping the bass response right? The original values had more bass than most people wanted. Is it possible that the 150K resistor at the input tames the signal too much and when I roll back the guitar's volume knob it gets too weak? Do I need therapy ?

[AM]

Ooops..I typed my last post after a very long day. I read it today and spotted a couple of mistakes. I mixed up C3 and C4. So, the values I was talking about are:
My project: 0.047uF for C1 and 0.022uF for C4.
Red Llama: 0.068uF for C1 and 0.033uF for C4
Tube Sound Fuzz: 0.1uF for C1 and  0.047uF for C4.

I also had a look at Mark Hammer's Fortyniner. Mark has used 0.1 uF caps in both positions. I don't know if that is for compensating for something that the booster does to the signal though...

[Mark Hammer]

I use 0.1uf simply because I found the stock values in the original TSF circuit were depriving me of bass.  A .1 value doesn't deliver gobs of bass, but you don't feel like you're missing it.  That being said, I usually play through 8" speakers, so don't go by my opinion.  In a 4x12" context there may well be more audible bass.

More critical to tone, I think, are the feedback caps (C2/C3).  My own preference when dealing with multi-stage overdriving is to "spool out" the top end bit by bit.  That is, earlier stages have lower treble rolloffs such that later stages are fed with more lower-order harmonic content than higher order.  In the case of the Red Llama, the 51pf and 100pf values provide estimated rolloffs of 2.8khz (at max gain) followed by 1.6khz.  At min gain, the first stage has a treble rolloff starting around 31khz, reducing to 5.2khz when the gain control is halfway.  Some folks may find that values of 68-82pf may well be more to their liking; e.g., 68pf in stage 1 and 82pf in stage 2.  That would get you a 2.1khz rolloff at max gain in stage 1, and a 1.9khz rolloff in stage 2.  Obviously, there are a great many who like the stock sound of the RL, so this is not necessarily an "improvement to a flaw" in the stock circuit, merely a twist and slightly different flavour.

If a person finds the bass end wanting, there is nothing wrong with inserting a gentle mid-scoop between C5 and the Volume pot.  That's how lots of other pedals achieve a big bottom sound - by attenuating some of the mids.  Lord knows there is enough signal amplitude in the circuit to weather a bit of passive loss.  You can find lots of great ideas for producing a midscoop over on Jack Orman's AMZ site.

Alternatively, I find that a SWTC does a nice job of trimming unwanting top end from a hot output such that I don't feel the need to trim the treble at the guitar or amp when kicking in the overdrive.

[AM]

Mark, thanks for typing all this info. After reading your text a few times I understand that I had the wrong idea about how C2 and C3 work. I thought that by increasing their value, one cuts hi frequencies. By reading your text I understand that the more one increases the C2 value and decreases C3 the more high frequencies are removed. Did I get this right?
I did some work on a couple of amps a few years ago and I remember some people supporting the idea of leaving plenty of high frequencies at the first stage and then cutting them down at the second stage. I noticed that by follwoing that method the sound could get a bit shrill in some cases. Your explanation about your preference in cutting more high end at earlier stages in order to maintain lower-order harmonic content than higher order sounds like the direction I would like to go too.

I don't know how you calculate the treble roll off but if I would swap the 220pF I currently have at C3 with the 47pF at C2 and go with C2=220pF and C3=47pF would that work? I realize that in your calculations you used the 51pF and 100pF stock values and then 68pF and 82pf. In both cases the sum of the values is almost the same. The sum of the values I used it exceeds the 150-151pF total (267pF).  That could give an interesting twist in the overall sound or totally kill the high end You used 330pF at both places in your 49er. That makes me think is more a matter of balance between the two caps than the actual sum of values. Is there some formula to calculate those roll off frequencies?
I have your SWTC schem in front of me. It looks like a great idea and the circuit definitely has more than enough gain to spare.

p.s: Sorry, I have asked you way too many questions about this project. I feel bad I've taken a lot of your time.

[Mark Hammer]

Don't sweat the question-asking.  It's what we do here.

The calculation of treble rolloff folows the time-honoured formula of F = 1 / [2 * pi * R * C] , where R is megohms and C is microfarads.  With a 1M feedback resistance and 100pf cap in parallel, you can see that F = 1 / [6.283 * 1 * .0001] = 1591hz.

Increasing the value of the feedback cap lowers that rolloff point, and making the cap value smaller increases the treble rolloff point.  naturally, since R is in the formula, the rolloff moves downward as R gets bigger too (note that the 220pf - 390pf used in the same locations in the 49-er is used in tandem with a 100k feedback resistor, hence the opportunity to use a larger cap value).  Since the Gain control is in the feedback loop of stage 1, as you turn the gain up, the treble gets rolled off at progressively lower and lower points as gain moves upward.  Since greater gain results in more harmonic content and perceptible treble, that gain-compensated tonal change that takes place is desirable.  (You might note that the identical phenomenon happens with the venerable Tube Screamer as well). 

The question for me is whether the range over which the tonal compensation occurs is ideally suited to "field conditions".  For instance, should one stick a booster in front of the RL, the chances are pretty good you might have the gain on the RL turned down a bit.  With the Gain turned down, the amount of treble rolloff in that first stage may be insufficient to bring out the best in the second stage.  This is why I suggested the cap change I did; it permits better restraint of treble content in the first stage under moderate-gain conditions.  Again, I can't emphasize enough that this is an adjustment to suit particular operating conditions and not a mod that will always be guaranteed to yield better tone.  I'm very certain that Jeorge Tripps did considerable empirical work in identifying optimal cap values, and certainly the number of people that the stock unit makes happy bears witness to this.

If you hope to achieve some "bite" from your own RL build, then I would suggest not using anything larger than 100pf in the first stage or second stage.  The 68pf/82pf values I suggested were simply to shift the "balance of treble control" so that stage 1 held the reins a little tighter, and stage 2 held them a little looser.  Some folks may find that they prefer the original 100pf value in stage 2 to keep a "rounder" tone, and that the stock value eliminates the need for a variable tone control.

Do take a look at Jack Orman's SWTC II article over at AMZ.  There are a great many options for simple-but-effective tone control that can help to deliver more sonic options without having to add a lot of parts or eat up too much signal, and that can simply wedge between the circuit output and the volume pot.

Good luck with your build.  I hope you nail the tone you seek.

[AM]

Mark,
Thank you so much! You have made clear a lot of things to me. Now I can do a bit of calculation before I solder away every time and also understand more when studying the circuits of my favorite pedals. With all the info I gathered from this thread I think I will be able to get the tone I want this time.

[unclejustin]

I just wanted to say thanks to everyone who posted on this thread. I had a very similar issue where I wanted less gain. I built an RL clone from the generalguitargadgets kit and which is pretty similar to the layouts posted here. I ended up pulling R2 and replacing it with a jump wire. Now I can draw the gain down where I want it. Next I'm going to sway the 1M pot for a 250k so I can have a little finer control since I never bring the gain above 25% now.

Thanks again everybody.