page: 1 2 3 4 5 6 7 8 9 10 11 12

Mini-amp files:   Miniamp.png | 000076   8/11/104 4:41 pm
Our family is doing a cross-country trip by minivan, and I wanted to bring something to play with on the way, since I'll be gone for 3 weeks. I put together a 2W miniamp from an LM380-based schematic I had. It's a pretty meat and potatoes unit, with a pair of op-amps forming the input gain stage and baxandall-type boost/cut EQ section. The speaker is a 6" Marsland full-range with a great big magnet, which has a surprising amount of bass for its size. The closed-back cabinet helps this along a great deal, and the amp is actually pretty damn loud.

I tried covering the thing in vinyl leatherette, but the small space made it difficult to make things fit snugly, mostly because you couldn't stretch things tightly. Finally, I used black automotive running board paint for the rear panel. This has a nice rubbery pebble-grain texture very similar to black tolex and is very easy to work with.

The cabinet is cheap particle board, and uses glue and screw for fastening. The baffle is painted with some nice thick flat "blackboard" paint we had. This is paint you can use to make a blackboard/chalkboard on a child's bedroom wall. The covering is just some heavy wire mesh staple-gunned to the baffle. The controls are mounted on the back panel, and nicely recessed so they don't get banged around on the trip.

The unit is powered by a pair of 4-packs with C-cells, for a total of 12vdc power. I hot-glued the battery packs to the side of the case. By all rights I should have screwed the chassis-plate for the controls to the back panel, but I was in a hurry and hot-glued it. I didn't stick in a wallwart socket but I should have. Maybe when I get back. The batteries are accessed and replaced by undoing the 4 screws holding the back panel on.

The more interesting thing is the heat sink on top of the LM380. I found this black anodized unit at a local electronics shop in their surplus bin. It bolts to the board on each/either end of the chip. I put some thermal compound between chip and heatsink, and was fortunate to have enough clearance on each side of the chip for this thing. Seems to me this could work very nicely for a pair of LM386's or other 8-pin DIPs when used in bridge mode, provided you plan out the layout wisely to allow clearance.

All in all, a pretty simple build, decent sound, and good portability for modest cost.

Little Leslie file:   Leslie.pdf 000075   7/29/104 3:21 pm
(NOTE: Graphic PDF file requires right-clicking and saving to hard drive for foolproof viewing.)

I acquired an older smaller Leslie type speaker this year. This is likely the smallest such speaker of its type, and uses an 8" speaker and what I like to call the "cheese wheel" rotating baffle. The unit has to be powered by a separate amplifier since it has none of its own.

Since most of the online stuff doesn't really give a good sense of the mechanisms, I thought I'd take a few inside closeup shots and post them.

What you can see is that the baffle is belt driven by motors off to the side. The "fast" motor is the one closest to the chassis, and the slow motor sits atop that, turning the shaft of the fast motor by an idler wheel (which you can just make out between the two motors).

The speaker sits behind the baffle board, with a bracket mounted over top of it, along with the pulley wheel for the belt-drive mechanism. A second bracket sits on the other side of the moulded styrofoam "cheese wheel" baffle. A spindle/axle runs through the rotating baffle and is held in place (with bearings, of course) between the outer bracket and the one just over the speaker.

As you can imagine, with so many things in the way, upper treble is not this things finest attribute, but it sounds great, is easily powered by small single 6V6 guitar amps, and is reasonably light to carry around, even though I made the cabinet over top out of MDF.

Dano-tech!! file:   Danotech.pdf 000074   7/29/104 2:42 pm
(NOTE: Graphic PDF file requires right-clicking and saving to hard drive for foolproof viewing. It's about 2.5meg)

I love making guitars, but quite frankly, the effort involved in making bodies is a real deterrent, both in terms of the mess, the space needed, and the stink (I don't always have the best ventilation). There is also a lot to learn in terms of working with finishes.

This led me to consider using the same fabrication technique as is used by Danelectro guitars (a much higher-tech version is also being used by Reverend guitars these days). This used what was essentially a "garbage wood" core, with a masonite top and back glued to the core. While softer, lighter woods can sound pretty decent, they don't hold up well to use (I had an all pine "Les Paul" years ago and can vouch for that). Masonite (the very fine pressed particle board that is partly wood, partly glue) provides a nice smooth, durable, and *inexpensive* surface. It also takes a paint finish very easily because it is essentially nonporous.

Although Masonite is much heavier than actual wood, use of a core and solid covering means that the weight issue can be offset by taking big chunks out of the core and "chambering" the body.

Why not just BUY a Danelectro, given how cheap they can be? Good question. Well, they always come with lipstick pickups, and I was interested in other sorts of instruments that haven't been made by Danelectro techniques yet (next up is a fake Vox Phantom!). So I stuck on the pickups of my choice on the body shape of my choice.

Since part of my mission was to sidestep spraying finishes, I used plain old formica to cover the top and back, and white edging to do the sides. Danelectro also uses a white plastic edging but not the same type. I used stuff from Home Depot that is preglued with heat-softened glue. Just line it up, fire up your heat gun and press it on, then trim away the excess.

On top of a masonite surface, the formica is pretty durable....and pretty, too. I've included a shot of a "Dano-Caster" in progress to show how nice some of the wood-grain formica can look. Closer inspection of the LP clone will reveal some cosmetic flaws around the edge, but that is more a testament to my desire to make this in a hurry for a road trip, than any indication of what is and isn't possible with the technique. It would be nice to use a router and install some creme binding along the edge to complement the pickups. Maybe later.

The total construction time for the red guitar was maybe 20 hours in all, spread over a few weekends and evenings, but I imagine someone with the proper tools (a band saw and sanding drum would have helped) who knew what they were doing could do it in less time. Much of the time was spent on correcting mistakes. Obviously having a finished neck and other parts sped things up.

For now, it plays pretty good, though I think neck angle needs to be worked on a bit. Happy to put that Bigsby to work, though. It's been in hiding for a good 10 years.

I included a picture of my new "Dano-Paul" next to my very overhauled, but much beloved, mid-60's Epiphone Coronet.

Ultraballz file:   Ultraballz.jpg 000073   7/3/104 2:13 pm
My Baseballs clone. Added the following controls:

1) Moved the two 10k filter trimpots to the front panel to stagger and retune the filters differently

2) Variable attack time control.

3) Slow/fast decay switch; brings in additional envelope-lag capacitor.

4) Split switch. Routes each filter output to a separate jack (slightly visible on left).

I also changed the stock resistive-divider "fuzz" path to a diode clipper.

VERY flexible pedal.

The Roseyray file:   Roseyray.gif 000071   3/12/104 2:55 pm
This is a simple circuit with a surprising amount of flexibility. It is basically a dual diode clipper, with two cascaded op-amp sections.

I originally called it the Voodoo Plus because I started out with the Voodoo Labs Overdrive as a template, but figured it was a little unprofessional to use a name that might be confused with theirs, so I changed it to the "Roseryay". Why the "Roseyray"? In hommage to the wonderfully shlocky horror movie called "The Thing With Two Heads" with Rosey Grier and Ray Milland as a self-conflicted science experiment gone wrong (Ray's racist white head is grafted onto Rosie's black football player body). Part of this thing is refined, and part is able to snap you like a twig, and it can't make up its mind which it is, yet each needs the other.

The original Voodoo Labs Overdrive, on which this was based, was designed to provide a warm overdrive by means of a diode clipper stage, similar to the MXR Distortion+ and the DOD 250. Because diodes to ground set a ceiling on maximum output level, the clipping section had a second op-amp based gain recovery section to boost the signal by a factor of 4 or so so that even mildly distorted sounds could push an amp a little harder than the MXR or DOD units might. Smart idea.

Here, the second op-amp stage is adapted for another nastier purpose. Instead of providing a modest clean boost with some additional lowpass filtering to keep things "warm" (the VDL-OD had a rolloff around 4.8khz on the second stage), the clipped output of the first section is boosted by a factor of 36 and clipped a second time by a pair of back-to-back red LEDs, for another type of distortion, in this case much harsher because it is also boosting and clipping the added harmonic content of the first stage. It also has a higher rolloff frequency (6.6khz) to keep more of the sizzle and burn. Additionally, the first stage is capable of higher gain than the VDL unit, and is more similar to the MXR Distortion+ in that regard.

The truly interesting part is the tone control, though. It is much like the Big Muff Pi tone control, which pans between the outputs of two filters (highpass and lowpass) for different tones. In this case, however, it pans between the outputs of two simple filter sections, but each filter takes its source from a different clipper. The first clipping section feeds a simple lowpass filter to one side of the tone control, while the much hotter second clipping section feeds a mid-scoop (notch) filter lifted from the old Univox Superfuzz. Unlike the Big Muff, which pans between the bass and mid/treble versions of the same distortion signal, this control pans between a warmer less distorted version and a much toothier version. Although it feels like a transition between bass and treble, at no time is the bass ever lost.

What is also very nice is that since the distortion in the second section depends very much on the gain of the first (remember, the gains are multiplied), turning down the gain of the first section radically changes the distortion generated by the second stage, especially since the LEDs have such a high clipping threshold. What this means is that the kinds of sounds obtainable with the tone control, and the "channel-balance", changes depending on gain settings. This yields a really amazing range of tones from just two knobs (gain, tone), that can attain hint-of-dirt Tube Screamer type tones (with more bass) to Mick Ronson-type round grind to shrieking death-metal gain.

Finally, this thing has a nice big, fat, volume boost.

Infinitely better schematic file:   INFLSCHM.gif 000070   2/12/104 7:18 pm
This is a retouched and editted schematic for the "Infinite Flanger" project that appeared in HSR (Home & Studio Recording?) magazine in 1985. The design is by Paul Williams and the entire article, layout, and parts list can be found in a zipfile at:

The schematic is shown in that zipfile but is a small portion of the page. Here it is extracted from the page, touched up for better legibility and printing, and I have replaced all the component numbers (e.g., R25, C10, TR2) with component values (100k, 2u2, BC182, etc.) plus a few stage labels. The intent is to make it easier to follow, and hopefully easier for the clever among you to reconfigure to use more readily available BBD chips. The original did not clearly indicate electrolytic capacitor orientation. I have tried to do that here to the best of my knowledge, but it is still worth checking over should you decide you want to make one.

The Infinite Flanger is a BBD-based "through zero" flanger. This mimics what you used to only be able to do with true tape-hub flanging. Here the signal source that is delayed can not only "catch up" in time with the straight signal, but move ahead in time before it start to sweep back and lag again. It accomplishes this by having the "dry" signal actually delayed by a very short fixed amount. This way, as the swept delay approaches the shortest delay it is capable of, it can actually be delayed by a smaller amount than the fixed delay.

The effect as it passes "through the zero point" is really quite striking. You used to only be able to get this by tape flanging, or by using two completely separate delay lines, a splitter, and a mixer. The I.F. does us the convenience of packaging it in one convenient device. At the moment, there is only one analog pedal on the market that can do this.

Note that there is NO LFO in this project. It is designed to be swept either by hand or by means of an external control signal. That has some shortcomings but ultimately makes for a more flexible design. Note as well that the through-zero aspect is most useful at very slow sweeps. This is the classic "jet-plane" swoosh people associate with flangers (but somehow are never quite able to get). If your intent is to use a flanger for pleasant quasi-Leslie speaker sounds, or chorus-like effects, the through-zero aspect offers little or no sonic benefit.

Dual loop selector file:   DUALLOOP.png 000069   11/27/103 8:34 pm
This schematic comes from an *old* Japanese projects book. Old enough that I had to flip around the PCB mask to make it PnP-compatible.

It uses FETs (2SK30) to provide bypass switching and LED indicator switching from a SPDT stompswitch. It probably isn't the most transparent loop selector on the planet (relays would be better), but it has a buffer (2SC945, but a 2N5088 or 89 will probably do nicely, and you could probably adapt a FET input buffer for even higher input impedance, although what's there is pretty decent), and some other nice features, plus it will run off a single 9v battery (which relays won't).

You will need 8 jacks for this, since there is a normalized (but not foot-selectable) in/out loop in the middle of it. You can skip the 3rd non-selectable loop, though, and just stick with the master in/out and 2 loops. That will reduce cost and package size and simplify construction a bit.

Probably the best part is that a nice cheap SPDT switch finds a proper use, and that while it uses a lot of parts, they are generally very common ones. Probably the worst part is that most folks have plenty of 10uf, 47uf, and 100uf caps in their bins but not many 22uf. There is also the question, for purists of what sorts of signal levels FETs can tolerate before clipping.

Still, all in all, a decent project, and quite capable of being stuffed into a 1590BB with jacks, battery and all. I have not personally tested it but I assume the publishers checked it for errors.

As with anything posted here, do not assume that scanned layouts are perfectly to scale. You may have to resize.

page: 1 2 3 4 5 6 7 8 9 10 11 12