doorstop electronics


music electronics that're noisy and weird 

Doorstop "Nucular" 

    This combies a simple Tube Screamer-style distortion pedal, a Big Muff-style tone control, and a notch filter (with switchable center frequencies) at the end.  Schematic here.  (Note that the LM7809 regulated power supply - absolutely necessary! - is not shown.)  I used a rail-to-rail quad opamp, the TS974.

Passive notch filter

A simple notch filter using a nine-position rotary switch, a Cinemag 1.2H inductor, and plenty of love.  Schematic to come.

Yet another Goldtooth Softy

This time, a one-bottle variety, and running on 24VDC.  Yes, schematic to come.

Here it is.   Nothing special here, just two simple RC-coupled gain stages with a "Big Muff Pi" style tone control between them.  This little circuit gives, at most, 20 dB of gain ... enough to push the front end of a guitar amp slightly.  Great for a bright tone on your guitar, I think.

Here's a demonstration of the tone you can get from this design. It really breaks up heavily with P-90 pickups.

Dual Pentode Mic Preamp

Here's a new, simple design for a two-tube mic preamp.  I have not built it ... yet.  But it sure looks purdy.


A new Goldtooth Softy

Having got some useful feedback on the Goldtooth Softy and possible uses, I have made some radical changes, and learned a helluva lot about triode stages in the process.  Below is the schematic of the latest version.

The previous design had two triode stages (with a switchable, additional triode stage) and a cathode follower.  There was too much gain with three triodes, and the sound was a little thin, too, so I am going to use only two triode gain stages, each with half as much gain as in the original design.  That leaves me with two triodes for the output stage.  Having read generally unfavorable comments of the cathode follower I decided to attempt a cascode stage (SRPP) in place of the follower.  The jury is still out on that one - my ears aren't trained well enough to be sure of the difference.  But at least this approach allows me to recover a little gain (about 20 dB) after the tone control.

To reduce noise I put a balanced input transformer from edcor at the front end, which hopefully will provide with a little better noise rejection in a studio setting. A simple tone control should add some flexibility as well.

The bass-end rolloffs of the triode stages in the previous design were far too steep for some ears, particularly because the preamp found use as a line preamp to add "color" to a studio mix.  So I redesigned the cathode rolloff, with a switch to knock out the bass a little in the first stage. 

One thing that I learned in this process was a mistake I had been making in calculating bass-end rolloff caused by the cathode bypass capacitor.  Previously I had believed that the rolloff was calculated via the standard formula:

f = 1/(2*pi*R*C)

This is correct, but the calculation has to take into account the output impedance at the cathode, which is determined independently of the cathode resistor.  Output impedance at the cathode is

Z = (ra + Rp)/(mu +1)

where mu and ra are interal characteristics of the triode, and Rp is set by other design factors.  In this case, Z at the cathode of the 12U7 triode is 1071.  Determining the bass rolloff for the cathode capacitor is therefore the first formula above, but the value for 'R' is the parallel resistances of the output impedance and cathode resistor.  A 1k resistor and 1071-ohm impedance results in a parallel resistance of 517 ohms.  Thus, a 2.2 uF bypass cap produces a rolloff of 139 Hz.  Summed with a 4.7 uF cap (totaling 6.9 uF) the rolloff becomes a kick-drum safe 44 Hz.

(EDITED to remove the retarded mistakes.)

UPDATE:  Burleigh Seaver, of Shortstack, and I discussed what cheesy endorsement line I could attribute to him for this little pedal.  This is what he come up with: "When I push up the fader, all I hear is the guitar.  It's like I'm INSIDE the amp.  Amazing!"

Guitar Player couldn't say it better.

Vox AC-4 Clone

I finally finished a small tube amplifier, a clone of the unheralded Vox Ac4, to which I made a few small changes.  The original (which I have never used or owned) seems to have received some negative reviews, probably due to the shoddy construction techniques used by Vox on this practice amp.   But if built from scratch using fairly easy-to-find parts, and with a few basic changes, it is really a terrific amplifier. Though the circuit is fundamentally similar, it sounds nothing like a Fender Champ.  It makes a nice little alternative.

The basic schematic for my amp is below, but you can find two drawings of the original schematic at Blue Guitar Schematics.   

I made a few changes.  If I ever have a need or urge to open it up again, I'll make a few others.

The circuit has been panned for being far too quiet.   Well, that is an advantage sometimes, but to boost power slightly I used two EL84 power tubes, arranged in a parallel single-ended formation.  It gets about 8 watts as a result. If I had designed it as push-pull, I would have had 10 or 15 watts. (But then I'd be cloning the AC-10 or AC-15!)  Installing two tubes in parallel required me to halve the primary resistance.  In the original circuit, the grid's resistor to ground is placed after the the grid-stopper (100k) resistor.  That produces a resistor-divider effect that cuts the volume by almost 10%!.  Ludicrous.  I moved the 1Meg resistor to the input jack.  On the EF86 preamp stage I used a 1 meg screen resistor instead of 5.5meg, for a little bit more gain.  Finally, I used a larger power transformer, running about 30-40 more volts through those tubes.

People sometimes have said that the original amp was a little  noisy.  This is probably because of poor shielding and construction technique by Vox.  I installed a choke to provide additional power supply filtering, which was probably overkill.  In addition, unlike Vox, I used a thicker chassis that surrounds the entire circuitry, providing better shielding and thus more noise rejection.

I read that the original amp's sound is too thin, or alternatively that the sound is too farty.  So I used a beefy Hammond 125ESE output transformer (much larger than Vox's) and a 12-inch (not 8-inch like with the original) speaker, and the result is probably a warmer sound.  To ensure that I kept the brightness, I swapped out the 0.047 coupling cap (between the EF64 and EL84) with a 0.022 poly.

The tremelo is okay, but nothing special.  I used a larger-value pot for more control over (and more range in) the speed, but you can get away with a 1 meg pot.  I really dislike the tone control, since it has an extremely hard knee and is not versatile.  I advise leaving it out, or using a linear pot rather than a logarithmic one.  Or borrowing a different control, like from the Fuzz Jade  schematic.

Other changes:  I installed 1k screen resistors on the EL84s, to prolong tube life.  A footswitch for the tremelo.  And of course I used a three-prong power cord.

I think the nice sound comes largely from the EF86 preamp tube, plus driving the EL84s in a class-A arrangement.  Putting a little clean boost in front of this amp results in a terrific distortion sound (very Vox-like) but on its own, this amp is pretty clean, even at full gain.  In retrospect, I would use a smaller grid-stopper resistor (probably as low as 56k) and/or a slightly larger cathode resistor on the EL84 tubes.  I would definitely use a different tone circuit, with a switch to remove it completely.  I would probably also leave out the tremelo, and instead install an extra gain stage using the 12AX7.  Live and learn....

EDITED for grammar and clarity.

ADDITIONAL THOUGHTS:  You might ask why I have a choke and a power resistor in series for filtering the power supply.  Well, the answer is that I could not find a power transformer that suited the purposes (i.e., enough current at the right voltage).  One feature I wanted for this amp was the ability, if I wanted to, to remove one of the EL84 tubes for a return to a slightly quieter amp.  In retrospect, I don't see the point, but ....

Any way, I found a slightly overrated power transformer, and had to use the 1k resistor as a drop-down to get B+ under the max rating for EL84 tubes.  If you can get your hands on a power transformer that can reach about 100 mA at 500 VCT, then you could use a choke filter without a drop-down resistor (say, a Hammond 159M).

Dual preamp, revisited

I made some long-planned additions to my dual preamp, already discussed below.  I added a switchable limiter to each preamp, based off of a circuit that allegedly simulated the compression effects of saturated tape.  That circuit can be found at "LXH2" website, here.

(The extra mic input is for future expansion.)

I can't say whether the circuit really produces a "really convincing replica" of analog tape compression, since I've never recorded to tape.  But it's a nice, elegant circuit and I needed some kind of limiter, and this one works wonders for me.  Here's the circuit with my changes (click on it to see the whole thing):

I think the resistors "soften" the limiting effect of the diodes, and in practice it produces very little distortion as you approach the threshold.  Once you reach the threshold, distortion is very dramatic, but there is enough wiggle room that the effect is just perfect for my home recording system.  With this limiter I can keep most of my recording levels in the "yellow" (as opposed to red).  I also removed a feedback resistor from the second op amp stage to cut back on the unnecessary  gain in the original circuit.

Not that this matters, I think, but I used a TL2074 (rather than a simple TL074) for my design.  The +/- 12VDC power supply provides - just barely - enough headroom.

Harmonic Goldtooth

A slightly enhanced version of the Harmonic Percolator. I love the clangy, scratchy Steve Albini guitar sound, and the HP is allegedly the pedal he uses to get there. Schematics for the HP that I've seen on the web produce a pretty boring germanium-heavy fuzz. But check out Tim Escobedo's site (Circuit Snippets, to the right) cause it has a quicky and dirty silicon version. I didn't put the clipping diodes in front either. I put a mosfet clean boost in front of it and swapped in a few different coupling caps. Here's a picture of one I built for a friend:

The knobs say "Pregain," "Volume," and "Bright." I need to work on my labeling skills. I found the old, out-of-production transistors for the heart of the circuit on ebay, and they make a noticeable difference from silicon equivalents.

This is a really sensitive circuit that had me swapping values in and out at various points for days. At some point I guess I decided I was close enough. The pedal gets an appropriate mix of heavy and clangy sounds when mixing bridge and neck pickups on my tele '72 custom.


Goldtooth Softy

This odd box has two 12U7 vacuum tubes inside it. These dual triodes were designed for car radios and operate on 10-15V. I learned about them from the nifty Sopht web site  (hence the "Softy" name) and picked a couple up. This runs strictly off a 500mA wallwart at 12V, through a simple 7812 regulator. Because I can power the filaments off of the same supply as the "B+" and because the filaments don't take much current, the circuit is pretty simple and doesn't need heavy iron.

At first I wired both triodes in each tubes up for gain stages, but that produced so much gain that it began to oscillate. I backed it off by making the last stage a cathode follower, The knobs are ordinary volume rolloffs between the first and second, and aftet the fourth triodes, respectively, so one knob is gain and the other volume. It's a perfect overdrive to put before a tube amp input, so often the volume knob becomes a gain control of its own. As with almost everything I've built, there are no tone knobs.

The horrendous paint job was the product of two cans of spray paint, simultaneously wielded. It could have worked out better. I haven't put a stomp switch on yet. The indicator light is a blue LED. I love those things.  Here's some interior pedal pron. It was messy, but I did the wiring in a day, I think. When I get my scanner hooked up and working, I'll post a schematic (which is, after all, why I decided to do this web site).

UPDATE. Here's the schematic.

Dual preamp

Texas Instruments makes a little 8-pin chip preamp that's very clean - the Burr-Brown INA217.  It can accept balanced inputs without a transformer and needs only a few external components.  The datasheet includes the basic schematic, which also shows how to drop the  coupling cap on the output by using a FET op amp to eliminate any offset voltage.  I like that because I figure, since capacitors can often introduce some kind of unwanted filtering, it's best to keep them out when possible. 

Anyway, I bought two and used an external laptop power supply (bought for only $9 from and rackmounted it.  The power supply is surprisingly quiet.  Having finished it off, I think I can recommend the circuit very strongly for anyone who wants to DIY a mic preamp but doesn't want to sweat a tube circuit or burden themselves with a hefty boutique discrete design.

Here's the data sheet for the INA217.  I used the schematic found in Figure 4 (page 7), but I used non-polarized 1 uF blocking caps for a high pass filter with cutoff at about 100 Hz.  For R7 (the gain pot) I think I used a 10k pot.  The op amp was just a TL072; as the datasheet says, the op amp is not in the signal path, so you can use any op amp you like, probably.  

And now that I have a couple preamps that are external to the soundcard, I really have been able to test the sound of my homemade mics.  Below is a clunky little acoustic guitar performance, double-mic'd by an SM58 and by my homemade mic (the fugly yellow thing you can see in another post).

Here are two samples: