Messages

Rookie mistake.

Eh, I miss stuff all the time...

Your perf work is really good. Mine rarely looks that tidy.

Man, a 1938 Russian rifle ... probably saw some military action in the 80s. It's really 1838's technology!

in Russia, rifle shoots you!

I'm here all week, try the veal.

There's something wrong with your regulator ... actually a couple things.

First, the center pin does not appear to be connected to ground. Fix that first and see if it results in 5V output.

Actually, there's a whole host of places you should be getting ground but you don't seem to be. What are the voltages on your transistors? Did you triple check all ground points on the layout with your multimeter?

Second, there are two types of regulators, the "L" which doesn't have the heatsink and without the "L" and they have opposite pinouts. If grounding the center pin doesn't fix it, you might want to turn the regulator around.

The chip is not a voltage inverter. It's a charge pump. It boosts the voltage on the audio path to +18V.

Finally, can you snap a clear picture of the underside of the board?

EDIT: Remove the PT2399 while you're getting the regulator working. Feeding it 9V could result in a dead chip. Get the 5V working first and then you can put it back in. Have another PT2399 ready to go just in case this one fried.

Dat kingslayer!

What is the "magical" property of the OA126? I have a metric crap ton of Germanium diodes available here and I could measure to see if they meet the magic spec. Is it FV drop? If so, what is it?

It's a germanium Zener diode. The Fv is .65 despite being germanium, and it also has some indeterminate reverse bias voltage that I haven' been able to measure (at any rate it exceeds 3v, which is all my multimeter can measure).

They also sound really frigging good imo. Soldering two germanium diodes in series for the same Fv is similar but doesn't quite have the same bite.

I bought an extra batch of about a hundred after I did the diode kits, some went to friends, and the rest will be pried from my cold dead hands. I don't know if my source was the same as Chromey's, but they are gone.

Post your voltages, too.

this circuit should have very low gain (a hair over 1x) when in vibe mode. Tremolo mode does have more gain, due to the trimmer.

Hope everything works out for you, Curt. Hang in there and good luck with the new business.

Well that's nifty.

This is normal. PT2399s are noisy, with increasingly bad fidelity at higher settings without heavy filtering, and the Cave Dweller does not heavily filter the single repeat from the "Echo." Brian calls this a "medium-fi" delay, but I would describe it as low-fi for the most part. Honestly, though, it's kind of part of its charm.

There's a perf layout for this one in my library with a halfway decent mod for making the mids switch a real second channel/leads boost (rather than just swapping out the volume and boost pots). I found the circuit almost unusable without the mids switch on, so I hardwired it "partly on" with a higher cutoff (to boost the treble) and made the mids switch turn it "more on." I actually did this mod right before selling it on the forum (I think Layton got it?) and almost regretted letting it go after that since it was so much better.

In the original, there are very, very minor differences in the capactitors associated with the two channels. One big reason the channels sound different is because the circuit is considerably darker at lower distortion levels.

Also, like all 4049 circuits, the thing is a "little" noisy. I think the reason it's not as noisy as the Red Llama isn't that it's better engineered or anything, it's just that it has a steep treble cut at the end of the circuit. I would actually describe the bandwidth of the circuit as "crappy."

aesthetically, I have always loved how Jack Deville Electronic does their layouts.

Yeah, Jack's layouts are works of art. Look even more closely and not only are all the resistors facing the same way, but they are grouped as much as possible by value.

Oh this should be fun.

All the Boss pedals are exceptionally well-engineered. And yet sometimes they manage to use a crappy buffer, despite being able to make an almost perfect one (the TU-2).

You also have to define "better" in terms of engineering. There are multiple types of engineering that go into a pedal:
1) Industrial: can this pedal be cost-effectively manufactured? Can results be duplicated on a large scale? What are the additional time, labor, and parts costs associated with improving consistency of a particular design (e.g. if you can make two pedals sound identical by twittering several trimpots, is that worth it?).

2) "trade dress" or something like that -- some companies have physical limitations. Most boss designs must be able to fit in a certain sized enclosure. MXR pedals are all either 1590B or 1590BB. Moore everything is in a 1590A, almost always with a specific knob layout. EHX does whatever the heck it wants most of the time. A cost-benefit analysis must be done to determine which corners can be cut to meet this need. And while some people on forums might scoff at it or carp about how it affects quality control, recognizable enclosures are hugely important to the vitality of a particular company.

3) Lifetime: How long will these parts be available? Will we be able to source them while they're available? What is the plan if this changes? This is a huge one for delay pedals: It's why all the major manufacturers have slowly moved toward digital, and it's also why cool audio chips get used even though it's still possible to get panasonic chips. You don't want to have to do limited runs just because something sounds better to someone somewhere.

All of these are on top of "does it sound good"? And there are many, many other things to consider.

Here's my personal philosophy:
1) How hard is it to track down a problem if it happens?

2) How many things can go wrong?

3) If a part is likely to blow up (MOSFETs and CMOS chips, I'm looking at you), is there a substitute, or a way to protect it? Sometimes this one has really good results -- my Hamlet delay came about looking for a way to avoid using a compander to avoid distortion, because they blow up easily and cost a lot of money. I ended up with a single LED (2c) being the magic solution.

Also, sometimes "bad" engineering is the secret to a sound. The Fuzz Face has low input impedance -- bad for guitar pickups. But it sees the pickups as a current source -- awesome for touch sensitive fuzz. The RAT uses a horrible, low-fidelity op amp ... and won't sound right without it.

The Kingslayer is just "different" from the klone, and this is where designer choice factors in. It will accomplish the exact same thing 99.9% of the time with a simpler schematic. I'm a big fan of that. It will also do "more" with minimal additions (switches and diodes). I'm a fan of that, too. Is +9V/-9V the " 'best' possible thing" that could be done from a design perspective? Maybe not, because we know we can squeeze 18V out of the positive rail like in the klon, so if you think that the best possible way to build that circuit is with the highest possible voltage from a 9V source, then you will think that Brian made the wrong choice in the Kingslayer. But the situations where that comes up are few and far between. The other difference is that the overdrive section won't be clipping the op amp as much in the kingslayer. That's another choice, not a design flaw or benefit. If you think that op amps shouldn't be clipped, you'll think Brian made the right choice for the kingslayer. And so on.

True enough, but we all know the typical TGP poster has superhuman golden ears.

Are these the same people that can hear 6" cable differences after a buffer and use power conditioning chords on their amps?

I am a fan of this build.

But then I have heard others saying that this approach is a hack and not recommended ... and I do wonder what would happen if you inserted a stereo jack ... hmmmm?

Who told you that? Using the ring as a switch with mono plugs is one of the main uses of stereo jacks. What about it could someone possibly believe is a "hack"?