Ludwig Phase II Clone Debugging and Miscellany

[Keppy]

One thing that made this a little abstract is that the signals are AC coupled from the collectors of Q7 and Q8 and so the raw DC levels were removed, leaving only the AC portions, which were square waves, albeit of the wrong sizes.

Yeah. What concerned me most was that the transistor that was "off" was still swinging half the supply, so you couldn't really turn either animation function off, although according to Dino this isn't all that noticeable. I guess at some point the last layer of weirdness just kind of blends in.

[digi2t]

OK, a 15v zener is a definite improvement where switching is concerned. I've been going over videos of my original units, and some bleed through was normal. That's why I considered it normal on my clone as well. As Kep says, part of the weirdness factor. Nope, I've looked at it again, and the original is much, much cleaner. I initially looked at the wrong part of the video, and then it also hit me... the switching on the original was purely mechanical.

Anywho, I've soldered in a 15v zener, and the switching is much cleaner now. No fuzz repeats in filter animation mode, and no filter animation in fuzz repeats mode. Both... well... is both.

I know this is supposed to be "a good thing", but it's almost clinical now. The switching is now as clean as the original. I neglected to take voltage readings after the fact though. I'll do that tonight.

Just kidding... it's a keeper. Add it to the doc R.G.

P.S. In case I never mentioned this before; if any of you are wondering what the difference is between the "Fuzz", and "Voice Fuzz" settings, this is what I've found. When using the animation in "Both" mode, putting the unit in "Fuzz" mode, the fuzz knob acts as a gain control for the fuzz. At 0, you get a real squarewave repeat, with nothing between the beats. At 10, the fuzz fills in the spaces between the repeats.
In "Voice Fuzz" mode, the Fuzz knob acts like a balance control between the filter animation, and the fuzz repeats. With the knob at noon, you get a mix of both. One end is only fuzz repeats, and the other is only filter animation.
Total coolness!

[Ry]

So I built mine from the first layout, and it looks like the connections on R64 in it are different from the second run that RG posted about.  Am I missing something, or is R64 not connected the same way in the first run?

[digi2t]

I've amended my previous post. Unfortunately, I must have taken a hit of acid just before writing it... or it's simply age ravaging my mind. Sorry guys.

So I built mine from the first layout, and it looks like the connections on R64 in it are different from the second run that RG posted about.  Am I missing something, or is R64 not connected the same way in the first run?

Looks to me like the traces are a bit different, but R64 is still in the same place, doing the same job. You should be able to use R.G.'s illustration for the zener placement on either version board.

On the last batch of boards, I think the traces got shifted around a bit so as the Fast/Slow indicator light switching could be switched around, and there was some added power options added to this board as well.

[R.G.]

Digi's right - I had to mess with some other stuff and squash things into slightly different places to accommodate the cleanup changes, but the R64 and ground connections are close enough to work the same.

Yell if there's a problem.

Note that you can tack-solder the zener on either the top of the board where the other parts are, or on the bottom where the copper traces are. That last might be more convenient as you might not have to remove the controls and switches from the enclosure. If you do that, you might want to use a photo/graphics program to flip the illustration over to see how the copper would look from the bottom so you get the correct pads.

[Ry]

I spent a good hour or so comparing the version 1 and 2 boards for this, and I don't quite see that the R64 connection is the same.  Are you saying that I should be able to tack the zener onto the R64/C29/Q17 signal pad on the Ver 1 board?  Sorry to sound dense about this...

[digi2t]

I spent a good hour or so comparing the version 1 and 2 boards for this, and I don't quite see that the R64 connection is the same.  Are you saying that I should be able to tack the zener onto the R64/C29/Q17 signal pad on the Ver 1 board?  Sorry to sound dense about this...

I think you're looking at the wrong board. We're not talking about the smaller filter board (I think you're looking at RF64/CF29/QF17). We're talking about the power board (the larger board). Both versions have R64, and R29 in the same place, so this illustration that R.G. posted before will work (notice that this is the board with the pots on it).



I've retested the voltages on Q7/Q8 collectors with a 15v zener, and as Kep spec'd, this is what I get now (values rounded);

Q8
- Fuzz Rpt. = 0v/15v
- Both = 0v/15v
- FFM = 15v

Q7
- Fuzz Rpt. = 15v
- Both = 0v/13v
- FFM = 0v/13v

[Ry]

Hmmm, I am looking at the larger board, and it's significantly different from that diagram.  Unfortunately, I don't see a working link on GEO to the version 1 build file that I'm using.  I'll dig a bit more, I must be missing something.

[R.G.]

Hmmm, I am looking at the larger board, and it's significantly different from that diagram.  Unfortunately, I don't see a working link on GEO to the version 1 build file that I'm using.  I'll dig a bit more, I must be missing something.

Did you get one of the manufactured Version 1's, or did you etch your own?

I still have the source files for the first version.

[Ry]

I etched my own from the very first build file.

[digi2t]

Crap... hadn't thought about that. Should look something like this, I've drawn in he zener.

[Ry]

Awesome, thank you   

I'll tack on the zener as soon as I get my Mouser order (I didn't have any appropriate values in my stash).

[Keppy]

Ry, that board doesn't appear to have Q7/Q8, which means you don't need that zener. Your board appears to predate the mod that necessitates it.

[digi2t]

Ry, that board doesn't appear to have Q7/Q8, which means you don't need that zener. Your board appears to predate the mode that necessitates it.

Isn't that what q3 qnd q4 are doing?

[Keppy]

Q3/Q4 are the flipflop. Q7/Q8 are added-on transistors that allow one side of the flipflop output or the other to be turned off using only one switch pole. See pg. 28 of the current build doc.

It looks like that is the version of the board that switched the animation using a DPDT on/off/on switch (or 4PDT on/off/on to include LED).

EDIT: The line above should read DPDT/4PDT on/ON/on.

[Keppy]

I just installed a 2N2646 in place of the 2N6027 and removed Ralt. The animation is working now, so it appears that was the problem. Ry, did you order a UJT? Smallbear carries them if you need one. I believe someone found it someplace else as well back in the earlier days of the project.

[R.G.]

Q3/Q4 are the flipflop. Q7/Q8 are added-on transistors that allow one side of the flipflop output or the other to be turned off using only one switch pole. See pg. 28 of the current build doc.

It looks like that is the version of the board that switched the animation using a DPDT on/off/on switch (or 4PDT on/off/on to include LED).

Yep, that's it. It sidestepped the introduced Q7/Q8 issue but needed the hard-to-find switch.

I just installed a 2N2646 in place of the 2N6027 and removed Ralt. The animation is working now, so it appears that was the problem. Ry, did you order a UJT? Smallbear carries them if you need one. I believe someone found it someplace else as well back in the earlier days of the project.

I'll do some more digging to get the PUT to work. The 2N2646 is quite hard to find. I've actually had the 6027 working on a breadboard in the "alt" circuit, although not in a Ludwig. There appears to be some fancier signal throughput to the flipflop needed.

[digi2t]

My apologies for the mis-call then.

At first glance, it looked like the old q3/q4 were the new q7/q8. I should have looked at it a little harder.

[Ry]

Thanks for digging in, guys.  I'll get mine on the bench in a couple hours, I need to wrap my head around the build again.

[Keppy]

I'll do some more digging to get the PUT to work. The 2N2646 is quite hard to find. I've actually had the 6027 working on a breadboard in the "alt" circuit, although not in a Ludwig. There appears to be some fancier signal throughput to the flipflop needed.

Yeah, the 6027 oscillator was ticking just fine. I don't know why it wouldn't work as a trigger. Not a big enough voltage spike or not a sharp enough edge on the transient, I guess. Then again, when I triggered the flipflop manually, brushing a screwdriver against the trigger point didn't always get it to flip. I needed to plant it firmly on the trigger point, so maybe the voltage spike needs to be a smidge longer. One of us needs to breadboard it along with the flipflop and get it working. I have all the necessary parts, but I don't really know how PUTs work, and it'll be a little while before I can get to it, so if anyone else wants to jump in just say the word. Otherwise I'll try and get to it when I can.