You can build an awesome-sounding designer distortion pedal for about $45. (Joe’s hand sold separately.)
Have you been reading about DIY guitar effect projects for years and thinking about giving one a shot?
Here’s a good excuse: Premier Guitar just published my “Build Your Own Stompbox” article, which just may be the most ambitious project I’ve ever undertaken for a guitar mag. And there’s more to it than just building a great-sounding distortion pedal from scratch: It’s practically a “how to build your own guitar effects” course, covering everything from sourcing parts and prototyping through boxing and troubleshooting. You’ll learn how all the key components work, and how to choose the ones that best suit your needs. Most important, you’ll start customizing the circuit from the project’s first stages using techniques you can apply to all your future builds.
The project is based on the simple but powerful Electra distortion circuit, the same one used in many boutique overdrives. Yup — it’s the same circuit that I’ve been featuring as Project #1 in Tonefiend DIY Club for a couple of years. But this revised and expanded version is far superior to the original, thanks to lots of help from my talented PG colleagues. Art Director Meghan Molumby did an amazing job assembling the massive build guide PDF (which you can download for free here). The layout is incredibly user-friendly, and you’ve never seen nicer-looking schematics and diagrams. The text got a thorough going-over from ace editors Andy Ellis and Shawn Hammond, so there are far fewer of my usual boneheaded typos. And there’s a cool video demo by Nashville ace John Bohlinger, so for once, you don’t need to watch one of my twitchy performances to hear how it sounds. What’s not to like?
But wait, there’s more: My pals at Mammoth Electronics have put together a $45 prepackaged kit to spare you the hassle of sourcing all those pesky little parts and drilling your own enclosure. There are lots of cool optional upgrades, like painted enclosures, premium switches, and bitchin’ knobs. Details here. (Neither I not Premier Guitar have any financial stake in the kit. I just asked Mammoth to create one as a convenience.)
Seriously — this one’s good. If you’re finally ready to burn your house down master that soldering iron, well, zero hour is here!
Probably the best intro to building your own pedals I have ever seen. Great job Joe!
Congrats! Maybe my memory is not long enough nor detailed enough, but this feels pretty unprecedented for a high-gloss guitar magazine. Even post-Occupy, it feels rare to find content in one that isn’t the echoing BUY BUY BUY of the pro-consumerism editorial policy, much less an article as genuinely community-oriented as this.
Thanks. No surprise that that’s the focus of consumer mags, since they subsist on ad income. Though it must be said that readers/consumers really love content about new gear. (I was humbled years ago by a Guitar Player mag reader survey revealing that far more subscribers read the ads than the editorial.) Anyway, this project was possible largely because Premier Guitar is web-based, and I could link to a dense 30-page PDF that would never fit in a standard print mag. Plus, the editorial staff was 100% supportive of this venture, and they made if way better with their contributions. They’re cool, smart people who really try to put the reader first.
Thank you so much!
I’ll second that! This is fantastic. Thanks for taking the time to put this together.
What timing! I actually just dove into your DIY club projects last week, and completed Project 1 just a few days ago. I put in switchable input caps and clipping stages, as well as a third switch that adds an extra 1N914 diode in series with one half of each clipping stage. It was lots of fun and I can't wait to do the next several projects. Thanks so much for putting together the DIY club projects!
In fact, I have a question: I'd like to try my hand at building a small preamp for an acoustic piezo pickup. I'm guessing that I could simply adapt Project 3's circuit to the purpose. I'd like to make the input impedance 10M, too. How would I go about doing that? Do I simply add a resistor at the input to ground?
The short answer to that question is no.
Project 3 shows two circuits built around an N channel Junction Field Effect Transistor. In this case a 2N5457 (unfortunately like many JFETs this is an obsolete part and can be hard to find).
N channel JFET’s require that the gate terminal be negatively biased in respect of the source terminal in order to function as active amplifiers. The precise amount of bias varies a lot from device to device and this can make them difficult to use in 9V circuits. Out of any random collection of 2N5457’s it is likely that there will be some individual devices that require such a high bias that they just won’t work in most 9V powered circuits.
The first circuit shown in project 3 is not a normal configuration for an N channel JFET because the gate and source are effectively at the same voltage (0V or ground) so there is no bias voltage. This circuit will conduct and produce a clipped signal at the drain only when the input signal swings negative of ground. Positive input swing will be clipped at +0.3V at the input and will appear at the output as negative clipped at 100mV or so. As far as I can tell this first circuit is just a rather peculiar clipping arrangement.
The second FET circuit can have the gate negatively biased with respect to the source, depending on the individual device Vp. The two 1M resistors put half the supply voltage onto the gate, so +4.5V in respect to ground. For a 2N5457 with a Vp of 1.5V and a 4.7K from source to ground, the source will therefore sit at around 5V making the gate 0.5 volts negative of the source. This second circuit is configured as a source follower and it therefore has slightly less than unity gain. Its input impedance is equal to the two 1M resistors in parallel = 500K. You could increase the input impedance by making these two resistors larger but they would need to be the impractical value of 20M each to get to 10M input impedance, and in any case this circuit has no gain.
If you want to build a piezo pre-amp I’d suggest you either invest in a DIY kit or find a circuit online specificaly for a piezo pre-amp and build that. And watch out for kits / clone circuits that seem to show a high value input resistor to obtain a high input impedance and then cock it all up by fitting a 1M pop stopper resistor to a bypass switch.
Many thanks Terry. I was wondering about that myself re: biasing the first transistor in the circuit, but it’s been a long time since my last physics class, so I didn’t think I should ask! How does it function as a clean boost then?
Well that’s the thing. It doesn’t function as a clean boost. It can’t.
It is capable of amplifying the negative peaks of an input waveform, but the positive peaks are always clipped.
The odd thing is that in Joe’s demo of his pedal the booster does sound clean. Joe would have to chip in with any further explanation of what’s going on with his particular pedal.
To make that booster into a correctly configured, high input impedance stage, with gain you would need to add a resistor from source to ground. Make it a preset so bias could be set. Make the drain resistor to +9 a fixed value. Then run an input resistor from the gate to ground, which could be a fairly high value – 4.7Meg should still be OK pushing it to 10Meg might be going a bit far, and de-couple the input from the gate with an input capacitor, say 0.1uF. You could get more gain by AC bypassing the source resistor with say a 10uF electrolytic cap.
If you did that you would wind up with a high input impedance, single JFET stage with gain that would closely resemble the triode input stage on most valve amps, but of course running on 9V with a JFET instead of half a 12AX7 dual triode valve. You would likely have to select a suitable 2N5457 and set the source resistor trimmer for optimum performance.
OK just done a quick and dirty simulation of what I’m suggesting so I know this should work with a typical 2N5457 JFET –
0.1uF as an input cap to the gate
4.7M from gate to ground
25K trimmer from source to ground (set to around 15K as a starting point)
47K from drain to +9V
10uF elec output cap, + to the drain, – to top of a 100K log output level pot. Bottom of pot to ground, wiper is the output.
That will give you a basic pre-amp with a 4.7Meg input impedance and a gain of around 2. For a gain of roughly 20, bypass the source trimmer from source to ground with another 10uF elec cap.
Thanks again for doing the simulation, Terry. From the little bit of brushing up I’ve been doing online, your design appears to be much in line with a self-biased FET amplifier. I’d probably also place a source follower after it as well to buffer the output, and I’m guessing the buffer circuit in project 3 is suitable for that.
I’m quite curious about the circuit of Project 3 now! After I box up project 2, I’m going to see what happens when I breadboard it.
Hi Richard,
I think I now see what is going on in Joe’s circuit. JFET manufacturing specs are quite wide and the Vp for a 2N5457 can vary between 0.5V and 6V. If you happen to have a 2N5457 with a Vp that is at, or very near, the low limit for Vp, in Joe’s booster circuit the internal channel resistance of the FET itself is enough to result in the gate being biased negative. As the trimmer in the drain is adjusted the current through the FET varies and the internal source voltage drop changes, varying the bias on the gate. So there is a limited amount of adjustment that can be made to achieve the ‘best’ bias for JFETs with Vp’s that fall within a low range. The trouble is that adjusting that trimmer for bias will also have an affect on the circuits gain.
Although it kinda does work for JFETs with very low Vp and is a circuit with minimum parts count, as a design it is a long way away from best practice. Also it has no gate pulldown resistor (and no input de-coupling cap) so it is relying on the guitar electronics to hold the gate at ground.
If you do try out this Project 3 booster results will vary depending on the particular characteristics of the JFETs you happen to use.
Yes you could use the buffer circuit in project 3 as a buffer, although again if you happen to use a JFET with high Vp it won’t work because the source voltage of the source follower sits at 4.5V (the two 1M resistors divide the 9V supply to 4.5V) plus a value that gets bigger for devices with higher Vp. At some point the source voltage gets too close to the 9V supply and the signal output clips on the positive peaks.
JFETs are a bit of a PITA to use in low voltage pedal circuits because you need to select them for these low Vp values.
So what you’re saying is I should have bought more than two 2N5457s then 🙂
That all makes sense to me. I think I’ll try breadboarding your booster circuit as well and compare the two. Maybe I’ll get lucky with the JFETs, but otherwise it’s likely your circuit that will go into the eventual box! As for the buffer, I take it the appropriate thing to do to adjust the bias is to use a potentiometer as a voltage divider so I can find the appropriate bias for the JFET?
Or you can build the cicuits and use sockets for the JFETs. Pop some JFETs in and measure the voltage from ground to source. As long as it is not too far from 1/2 the supply you have room for the signal on the drain to swing positive without hitting the supply voltage (or ground) and clipping. It does mean that ideally you need a bag of JFETs so you can cherry pick the ones that will work well in 9V circuits.
With Joe’s buffer circuit, since it has a fixed resistor divider, to adjust bias you could put a trimmer in between the 1Megs. Or trim the source resistor.
Trimming the source resistor makes much more sense, doesn’t it… plus I already have a couple 10K trimmers lying around.
This has been awesome. Thanks so much for your answers and advice, Terry. I’ll name my booster pedal after you 🙂
Hi guys — I’ve been away, helping out my ailing dad. Sorry I haven’t contributed to the thread here — but Terry’s doing a better job that I ever could! Thanks, man, for covering for me! 🙂
Hi Joe,
Glad I didn’t piss you off by horning in ! I was beginning to worry I was coming off like that annoying Harry Enfield character (did you get that show over there?) – Now I do not think you wanted to do that. What you want to do is … here let me do it ….. BANG! … mayhem ensues
Anyway I learned a bit more about JFETs from the discussion.
Naw, most Americans have no idea who Enfield is. But I was in a Brit band, innit? Don’t get me started on Vic Reeves! 😉
I always hear that JFETs are unpredictable, and that you need a trimpot somewhere to guarantee they work properly in these circuits. But the J201s and 2N5457s I come across always seem to work fine. Maybe I’m just lucky.
Great pedal Joe, I built the pedal yesterday after reading your article in PG.
Well written article, easy to follow.
Awesome! You’re the first person I’ve heard from who’s completed the project. Any suggestions? Any parts that could have been presented better?
https://www.youtube.com/watch?v=uIUAr-y6T50
Joe, you are my hero! I've been trying for years and was able to almost complete the project (you can see in demo, I didn't get the switch, AC input or stero input jack to work…but I'm going to breadboard it next time). I love how you start out with a working circuit and then add on to it. You project opened up a whole new fuzzy world.
One question — way no ground to the output jack? I didn't see in the pics and I didn't find that step in the verbage. So, when I backed up I used a mono input and nixed the switch to get it work….and it sounds great! Thank you, thank you, thank you!