Kliché Mini Mod for more bass in the distortion, higher treble gain with the tone pot and a smoother overdrive

mattdrago

New member
This is my first post here and I've been really liking the builds I've been making from PedalPCB, so I wanted to help contribute to the community! I was searching the forums and couldn't find a post describing these types of Klon mods, which I thought was odd, so I figured I'd make a post myself. These are pretty simple and common Klon mods, I simply am listing the values for each mod that I personally think sound the best. I have more complex modifications that I can describe in a separate post if people are interested, but I figured these mods would be the most accessible to people building this kit and the least invasive. I choose these specific values based on the formulas that I've included, so if you're interested in what I used for reference and figuring out your own mods then I highly recommend this analysis of the klon circuit: https://www.electrosmash.com/klon-centaur-analysis

A quick summary of how to do these exact mods;
  1. Change C15 from 3.9nF to 4.7nF for more bass. Higher values = more bass response. Possible values include 5.6nF, 6.8nF, or even 8.2nF.
  2. Change R23 from 4.7k to 10k for more treble gain with the tone control.
  3. Use Germanium diodes that have a Vf = 0.3V for a little smoother, less aggressive distortion. You can look up the Vf of most diodes online, but it's best to measure them with a multimeter since germanium diodes typically have very large tolerances that can result in two diodes from the same batch having significantly different Vf values. **Note** The higher Vf of the diodes used generally results in less clipping of the signal, therefore less overdrive and increased output gain. I typically use D9E diodes with Vf measured around 0.24V, but even the small change to diodes with a Vf = 0.3V created so much more gain that the volume pot had to be set barely more than fully counterclockwise just to get it at unity gain. If you want to reduce the overall output gain caused by this, read about that below.

If you're interested in the details of these mods and the formulas used for calculating the selected values, read on! Please note that this is a simplified analysis of an ideal circuit, so actual measured results can vary.

For more bass in the distortion, I used the very common mod that's done with Klon clones and simply change the value of C15 (3.9nF). Some people use a toggle switch to switch between 2 different capacitors to get this bass boost, but I just used the formulas to approximate the value I wanted and used that. The tone control in the circuit is an active high pass shelving equalizer, so it simply passes any frequencies before the cutoff frequency and all frequencies after that become boosted or reduced. The formula for calculating the cutoff frequency is fc = 1 / (2*π*R*C). This is the general formula for basic first order RLC filters, which are used in tons of pedals everywhere. The famous Moog ladder filter is actually just 4 cascaded 1st order low pass filters filters that have global feedback ie. resonance control.

1. Using the original PedalPCB values, the cutoff frequency is fc = 1 / (2*π*R21*C15) = 1/ (2*π*100,000*(3.9*10^-9)) = 408Hz.
I've tried mods using 5.6nF, 6.8nF and 8.2nF and I personally like using 4.7nF, which results in fc = 1/ (2*π*100,000*(4.7*10^-9)) = 338Hz. It's subtle, but noticeable if you were to compare them side by side. If you still want more bass, simply use 5.6nF or 6.8nF instead, since the increased capacitor value here reduces the cutoff frequency of the filter. I personally don't like going up to 8.2nF since the cutoff becomes 198Hz, which captures too much of the low end that I don't like when I'm trying to cut through the mix with an overdrive.

2. For greater gain from the tone control, I changed the value of R23 from 4.7k to 10k. The gain of the shelving equalizer described above is calculated by looking at the maximum and minimum gain you would get when the tone control is turned all the way down and all the way up. You can calculate this using the formula:
  • Gvmax= (RTreble+R23)/R20
  • Gvmin= R23/(RTreble+R20)
Using the original PedalPCB values, we get the following;

  • Gvmax = (RTreble+R23)/R20 = (10,000+4,700)/1,800 = 8.16 (+18.2dB)
  • Gvmin = R23/(RTreble+R20) = 4,700/11,800 = 0.4 (-8dB)
Changing R23 to 10k gives Gvmax = 11.1 = +21dB. This change proportionately affects the gain of the tone cut as well, resulting in Gvmin = 0.847 = -1.4dB. It doesn't seem like much, but I personally always have the tone cranked up anyway so I wasn't worried about the gain lost in treble cut. If you want more treble gain than this, you could try 20k to get a max treble gain of 24.4dB. Granted this will pretty much remove any treble cut possible, but if that's what you want then try it out!

3. For a smoother and less aggressive distortion, you can change...the germanium diodes of course! I'm going to oversimply things here, but what matters right now is that they are germanium diodes and the value of Vf, or the forward voltage drop measured across them. Basically the lower the Vf is of the clipping diodes, the harder the signal clips and the more distortion you get. Conversely, a higher Vf results in softer clipping and less distortion. I have always used D9E diodes or 1N34A diodes that had a Vf measured between 0.23V to 0.25V. This time I decided to use a matched pair of germanium diodes with a higher Vf than normal, since I wanted a klone that would be a little softer and not so harsh when I really crank up the gain on it. I used a matched pair of 2AA112 germanium diodes that had a Vf measured at 0.304V and 0.302V. Choose a germanium diode that has a measured Vf closer to 0.3V to get a noticeably smoother distortion; I personally didn't mind losing a bit of distortion due to the higher Vf here, since I would never turn the gain up past 3 o'clock anyway. I think the higher treble gain mod helps to compensate for any distortion that might be lost here as well, but if you feel like there's not enough distortion then just use germanium diodes with a lower Vf. I like both diodes to have nearly identical Vf because I like the sound of the symmetrical clipping here, but again it's all subjective and up to you.

That being said, a higher Vf means the overall output volume is increased as well, since the signal has more headroom and isn't being clipped as hard. There's different ways to approach this and reduce the output gain, like changing the resistance across the volume pot or making a voltage divider using resistors at the output. I haven't done this mod yet, but if people are interested then I will make another post going over different options for doing this.
 
This is my first post here and I've been really liking the builds I've been making from PedalPCB, so I wanted to help contribute to the community! I was searching the forums and couldn't find a post describing these types of Klon mods, which I thought was odd, so I figured I'd make a post myself. These are pretty simple and common Klon mods, I simply am listing the values for each mod that I personally think sound the best. I have more complex modifications that I can describe in a separate post if people are interested, but I figured these mods would be the most accessible to people building this kit and the least invasive. I choose these specific values based on the formulas that I've included, so if you're interested in what I used for reference and figuring out your own mods then I highly recommend this analysis of the klon circuit: https://www.electrosmash.com/klon-centaur-analysis

A quick summary of how to do these exact mods;
  1. Change C15 from 3.9nF to 4.7nF for more bass. Higher values = more bass response. Possible values include 5.6nF, 6.8nF, or even 8.2nF.
  2. Change R23 from 4.7k to 10k for more treble gain with the tone control.
  3. Use Germanium diodes that have a Vf = 0.3V for a little smoother, less aggressive distortion. You can look up the Vf of most diodes online, but it's best to measure them with a multimeter since germanium diodes typically have very large tolerances that can result in two diodes from the same batch having significantly different Vf values. **Note** The higher Vf of the diodes used generally results in less clipping of the signal, therefore less overdrive and increased output gain. I typically use D9E diodes with Vf measured around 0.24V, but even the small change to diodes with a Vf = 0.3V created so much more gain that the volume pot had to be set barely more than fully counterclockwise just to get it at unity gain. If you want to reduce the overall output gain caused by this, read about that below.

If you're interested in the details of these mods and the formulas used for calculating the selected values, read on! Please note that this is a simplified analysis of an ideal circuit, so actual measured results can vary.

For more bass in the distortion, I used the very common mod that's done with Klon clones and simply change the value of C15 (3.9nF). Some people use a toggle switch to switch between 2 different capacitors to get this bass boost, but I just used the formulas to approximate the value I wanted and used that. The tone control in the circuit is an active high pass shelving equalizer, so it simply passes any frequencies before the cutoff frequency and all frequencies after that become boosted or reduced. The formula for calculating the cutoff frequency is fc = 1 / (2*π*R*C). This is the general formula for basic first order RLC filters, which are used in tons of pedals everywhere. The famous Moog ladder filter is actually just 4 cascaded 1st order low pass filters filters that have global feedback ie. resonance control.

1. Using the original PedalPCB values, the cutoff frequency is fc = 1 / (2*π*R21*C15) = 1/ (2*π*100,000*(3.9*10^-9)) = 408Hz.
I've tried mods using 5.6nF, 6.8nF and 8.2nF and I personally like using 4.7nF, which results in fc = 1/ (2*π*100,000*(4.7*10^-9)) = 338Hz. It's subtle, but noticeable if you were to compare them side by side. If you still want more bass, simply use 5.6nF or 6.8nF instead, since the increased capacitor value here reduces the cutoff frequency of the filter. I personally don't like going up to 8.2nF since the cutoff becomes 198Hz, which captures too much of the low end that I don't like when I'm trying to cut through the mix with an overdrive.

2. For greater gain from the tone control, I changed the value of R23 from 4.7k to 10k. The gain of the shelving equalizer described above is calculated by looking at the maximum and minimum gain you would get when the tone control is turned all the way down and all the way up. You can calculate this using the formula:
  • Gvmax= (RTreble+R23)/R20
  • Gvmin= R23/(RTreble+R20)
Using the original PedalPCB values, we get the following;

  • Gvmax = (RTreble+R23)/R20 = (10,000+4,700)/1,800 = 8.16 (+18.2dB)
  • Gvmin = R23/(RTreble+R20) = 4,700/11,800 = 0.4 (-8dB)
Changing R23 to 10k gives Gvmax = 11.1 = +21dB. This change proportionately affects the gain of the tone cut as well, resulting in Gvmin = 0.847 = -1.4dB. It doesn't seem like much, but I personally always have the tone cranked up anyway so I wasn't worried about the gain lost in treble cut. If you want more treble gain than this, you could try 20k to get a max treble gain of 24.4dB. Granted this will pretty much remove any treble cut possible, but if that's what you want then try it out!

3. For a smoother and less aggressive distortion, you can change...the germanium diodes of course! I'm going to oversimply things here, but what matters right now is that they are germanium diodes and the value of Vf, or the forward voltage drop measured across them. Basically the lower the Vf is of the clipping diodes, the harder the signal clips and the more distortion you get. Conversely, a higher Vf results in softer clipping and less distortion. I have always used D9E diodes or 1N34A diodes that had a Vf measured between 0.23V to 0.25V. This time I decided to use a matched pair of germanium diodes with a higher Vf than normal, since I wanted a klone that would be a little softer and not so harsh when I really crank up the gain on it. I used a matched pair of 2AA112 germanium diodes that had a Vf measured at 0.304V and 0.302V. Choose a germanium diode that has a measured Vf closer to 0.3V to get a noticeably smoother distortion; I personally didn't mind losing a bit of distortion due to the higher Vf here, since I would never turn the gain up past 3 o'clock anyway. I think the higher treble gain mod helps to compensate for any distortion that might be lost here as well, but if you feel like there's not enough distortion then just use germanium diodes with a lower Vf. I like both diodes to have nearly identical Vf because I like the sound of the symmetrical clipping here, but again it's all subjective and up to you.

That being said, a higher Vf means the overall output volume is increased as well, since the signal has more headroom and isn't being clipped as hard. There's different ways to approach this and reduce the output gain, like changing the resistance across the volume pot or making a voltage divider using resistors at the output. I haven't done this mod yet, but if people are interested then I will make another post going over different options for doing this.
I built this pedal last week. Your suggested changing C15 from 3.9nF to 4.7nF sounded much better to me. Gonna look for those Ge diodes. Thanks for posting.
 
Last edited:
a note about klon diodes ... the original trace measurements (350 mV) were taken at 5 mA ... if you're measuring with a typical multimeter, you're probably testing at 1 mA and so your Vf should be lower ... how much lower depends on the performance curve of the diode (If/Vf) ... ballparking of 300mV is on the higher side of that range

speaking of this performance curve, the more gradual the curve the smoother the clipping comes on. some schottkys considered as a sub for germanium wrt forward voltage have a severely steep curve which means they begin clipping MUCH sooner than germanium. good for ring mod/octave, bad for smoothing out the hard clipping arrangement of the klon

matching diodes for clipping is NOT necessary. you wouldn't audibly notice asymmetry until you're in a significant difference in Vf and that wouldn't occur within using the same quantity of the same part. about the only time you're looking to match diodes up is for ring modulation/octave effects like the green ringer, foxx tone machine, or the octavia ... so long as you can get them within 10mV of each other all is good for that sort of effect.
 
Last edited:
a note about klon diodes ... the original trace measurements (350 mV) were taken at 5 mA ... if you're measuring with a typical multimeter, you're probably testing at 1 mA and so your Vf should be lower ... how much lower depends on the performance curve of the diode (If/Vf) ... ballparking of 300mV is on the higher side of that range

speaking of this performance curve, the more gradual the curve the smoother the clipping comes on. some schottkys considered as a sub for germanium wrt forward voltage have a severely steep curve which means they begin clipping MUCH sooner than germanium. good for ring mod/octave, bad for smoothing out the hard clipping arrangement of the klon

matching diodes for clipping is NOT necessary. you wouldn't audibly notice asymmetry until you're in a significant difference in Vf and that wouldn't occur within using the same quantity of the same part. about the only time you're looking to match diodes up is for ring modulation/octave effects like the green ringer, foxx tone machine, or the octavia ... so long as you can get them within 10mV of each other all is good for that sort of effect.
Very good point! I was operating under the assumption that one would already know the threshold voltages of whatever diodes they choose to use, but you'd also need to know the current across the diode for the corresponding Vf. I wanted to somewhat simplify my descriptions of what I did so that someone who isn't as familiar with all the technical stuff can easily find exactly what I did and replicate it without much thought. You're right though that the typical multimeter is measuring the threshold voltage with a different current than what these diodes would actually have in the circuit; even if they did measure around 5mA it wouldn't be a very accurate representation of the Vf response. There's some really simple circuits online that anyone can breadboard and are meant to measure the Vf across a diode at multiple different currents and then plot them on a graph to see the full Vf response across a wide range. I did this and took measurements at 0mA, 0.1mA, 1mA, 2mA, 5mA and 7mA; plotting the measured Vfs on a graph vs. the corresponding mA values are a good way to get an average to use.

Regarding the diodes being closely matched or not, I just so happened to have those 2AA112 diodes as a matched pair so it just worked out that way, but I've also never tried putting a pair of diodes into a klon that had significantly different threshold voltages to see what kind of asymmetrical clipping would occur. I didn't want to give the impression that using mismatched diodes would be good or bad, I just liked using the matched pair of 2AA112 diodes more than the other combinations I tried lol. You're absolutely right though that it probably wouldn't make much of an audible difference! I think a good rule of thumb for asymmetrical clipping diodes is to just keep the higher Vf diode within twice the value of the lower Vf diode; so if one diode clips at 0.3V, I'll usually choose the other diode to be something that's no more than 0.6V. I've tried some weird diode combinations for asymmetrical clipping on a tubescreamer mod once and when I tried some diodes with vastly different threshold voltages, it sounded all messed up and didn't clip right, just sounded bad lol.
 
Back
Top