Ludwig Phase II Clone Tech Notes

[R.G.]

One thing to note is that Dino's collector reads 9.6v. Mine reads the supply voltage ~35v or so. If anything, though, it seems like that would give me MORE current at the base, not less.

Take a 1M. Solder it on the bottom of the board from the collector of the add-on transistor to the base of the add-on transistor. I bet you get signal. Whether that's original or not is moot at this point. It will get you running.

[digi2t]

One thing to note is that Dino's collector reads 9.6v. Mine reads the supply voltage ~35v or so. If anything, though, it seems like that would give me MORE current at the base, not less.

There is a 10K resistor between the 35vdc and the collector. Wouldn't that cut down the voltage to 9.62? I'm taking a pot shot here. Bear with me, I'm learning to fish.

maybe it has to do with the division of being parallelled?  what's 1,000,000 divided by 150?

GAWD I am such a noob, open mouth, insert shoe store  After some studying, yes, parallel resistors do represent a different resistance. 150K and 1M give you about 130K. Sorry guys. Going to bed less stupid tonight... again!

[R.G.]

There is a 10K resistor between the 35vdc and the collector. Wouldn't that cut down the voltage to 9.62? I'm taking a pot shot here. Bear with me, I'm learning to fish.

Ohm's Law again. The voltage across a resistor can be - anything at all. If there is zero current, there is zero voltage. So if the transistor doesn't conduct any current, there is no current in the 10K resistor, and its voltage drop is zero. In fact, the measurement that both ends of the 10K are up at 35V is a dead giveaway that the add-on transistor is not conducting at all.

GAWD I am such a noob, open mouth, insert shoe store  After some studying, yes, parallel resistors do represent a different resistance. 150K and 1M give you about 130K. Sorry guys. Going to bed less stupid tonight... again!

We all get Mother Nature's little lessons. What's important is to learn from them. I like to think that the sheer number of scars on my hands indicates some kind of education. 

[Keppy]

One thing to note is that Dino's collector reads 9.6v. Mine reads the supply voltage ~35v or so. If anything, though, it seems like that would give me MORE current at the base, not less.

Take a 1M. Solder it on the bottom of the board from the collector of the add-on transistor to the base of the add-on transistor. I bet you get signal. Whether that's original or not is moot at this point. It will get you running.

I did exactly what you're suggesting prior to the post you're quoting.

Doing this only gave me about .13v on the base. Still no conduction. I could lower the resistor value, but I hesitate to do anything here that is a known departure from the original. Any other ideas? Might I have an error somewhere else creating a voltage divider of the wrong sort?

[R.G.]

I did exactly what you're suggesting prior to the post you're quoting.

Ooops. Sorry. I read that wrong.

OK, found out what I did wrong. I ignored R73 (4.7K). Bad move.

Instead of 1M, put in 47K, 51K, or 68K; 47K preferred.

[Keppy]

Got it! The effect now passes signal. Dry is bright, fuzz works and is a little quieter; voice fuzz is louder, harsher and severely gated.

No animation or formants yet. As far as the animation, I suspect console board Q2, the unijunction transistor.

e = 2.07
b1 = 2.47
b2 = 2.47
(I get the same readings whether animation is on or off. All readings are stable.)

R.G., as I said in an email, the datasheet I found for this transistor listed a different pinout than what is in your layout. Currently, I have the legs bent around to match the pinout on this datasheet: http://www.datasheetcatalog.org/datasheet/philips/2N2646.pdf

Should I pull it out of the socket, unbend the legs and put it in according to your layout?

As far as the formants go, none of them does much of anything. If I turn them all on and move the pedal, I hear a tiny difference. Barely.

[digi2t]

Got it! The effect now passes signal. Dry is bright, fuzz works and is a little quieter; voice fuzz is louder, harsher and severely gated.

You're on the right track Kep. Dry signal sounds right. Fuzz, and Voice Fuzz, might be switched around. In Voice fuzz mode, the control goes from nothing to full smooth fuzz. In Fuzz mode, it ges from nothing, to full fuzz around the middle of the range, to a gated fuzz at the top.

[Govmnt_Lacky]

...I suspect console board Q2, the unijunction transistor.

...the datasheet I found for this transistor listed a different pinout than what is in your layout. Currently, I have the legs bent around to match the pinout on this datasheet...

Keppy,

What is the transistor that is called out in the original unit? I have drawers full of transistors from the last 50 years. Let me do some digging. 

[R.G.]

Got it! The effect now passes signal. Dry is bright, fuzz works and is a little quieter; voice fuzz is louder, harsher and severely gated.

Yee-haaah (as we say here).

No animation or formants yet. As far as the animation, I suspect console board Q2, the unijunction transistor.

Q2 would be the animation thing. Formants will be tinkering with the fall plate filters.

e = 2.07
b1 = 2.47
b2 = 2.47
(I get the same readings whether animation is on or off. All readings are stable.)

R.G., as I said in an email, the datasheet I found for this transistor listed a different pinout than what is in your layout. Currently, I have the legs bent around to match the pinout on this datasheet:
Should I pull it out of the socket, unbend the legs and put it in according to your layout?

I sure thought I re-scrambled those to match the datasheet.  Try this: set your meter to measure resistance and start measuring between pins on Q2 out of circuit. A *uni* junction transistor only has one diode junction, that between the base and the bar of conductive stuff. The bar of conductive stuff looks like a resistor between the two emitter leads, and will measure the same resistance with either direction of the meter leads. The emitter is the only one of the three that will measure open circuit/reverse diode to either of the bases. Once you identify the emitter lead, there is only two possible orientations for B1 and B2.

If the orientation/pinout you're using is right, there's a bug in the circuit feeding the socket. Pull out Q2 and measure voltages on the socket pins. The B2 side should measure 35V, the B1 side should measure zero, and the emitter pin should measure the same as the emitter of Q1, and wire pads 11, 12, and 28. These are all connected by resistors and pulling Q2 gives them no current flow, so they all measure the same voltage.

As far as the formants go, none of them does much of anything. If I turn them all on and move the pedal, I hear a tiny difference. Barely.

I suspect this may be a trimmer setting issue. Let's get the animation working, then we'll hit the filters.

[pinkjimiphoton]

great work, keppy!! the fuzz part working is a big step...that's half the battle, right there. it's a very bright, nasally fuzz, and depending on the fuzz type switch, you should get normal fuzz one way, and better fuzz in the middle as dino pointed out. on mine, it came down to playing with that 150r trimmer,
but...i think i know what the issue may be from your description. i made a mistake yesterday...i confused r65 for r29, cuz i was running around like a chicken with no head.

R65 NEEDS TO BE DIMED FOR THE FORMANTS TO WORK!! try that, don't set it half way, try it pegged...mine only seems to work right damn near all the way up.

hope this helps bud!!

[digi2t]

R65 NEEDS TO BE DIMED FOR THE FORMANTS TO WORK!! try that, don't set it half way, try it pegged...mine only seems to work right damn near all the way up.

Mine was set completely counter-clockwise when I got mine. I did dial it up (or down?) a touch, about 15 - 20 degrees, since it accentuates the thumping of the oscilator in fuzz repeat mode. Any more than that sends the unit into very loud oscilation, which can even be heard in bypass.

R29 is a formant frequency adjustment. Mine is close to the middle, and small adjustments can yield fairly large changes in the vowels produced. Hence my earlier request for multi-turn trimmers on the clone  . R29, R31 (close to the middle), and R20 (about 2/3 clockwise), should get you close. Other than R4, which adjusts the rate range, I've found the other trimmers only to affect frequencies when the unit is in fuzz repeat mode.

[pinkjimiphoton]

weird!!! then perhaps he needs to just about turn it off...mine is at the dimed position, but dimed is actually fully counterclockwise...off maybe a degree or two, that's what i meant by dimed. my bad, dino...i just remember it's at the one extreme. if he's got the trimmer in the middle, that may be why he's getting no formants then maybe?

i wonder if we have different revisions of the circuit...cuz mine, some of the trimmers seem to be set completely different?

[R.G.]

I'm guessing that the individual characteristics of the transistors in the filters may require dramatically different trimmer settings.

In the electronics design world, trimmers are a pain; they are unreliable (compared to a fixed resistor), drift worse, and require a human to set them. The human setting labor is quite expensive. Trimmers are avoided as hard as they can be. So having - what, seven of them? - in this thing means that they had to put them in, and that means that they may be critical, as fits your experience Jimi.

From where the trimmers are in the circuit, it's pretty certain that they interact.

I've been thinking that the best way to get Keppy running is probably to read the voltages on the two working units' transistor collectors, diddle the trimmers to get Keppy's collectors to nearly there, then do fine tweaks. I'm pretty sure the "test points" noted in the schematic can be used for tuning it in, but there may be some complicated procedure which is now lost to us. We'll have to make up our own as we go.

[Keppy]

I sure thought I re-scrambled those to match the datasheet.

'Fraid not. The datasheet has the emitter on the middle pin. Looking at the schematic, your layout has it on the side closest to the fall plate board.

Try this: set your meter to measure resistance and start measuring between pins on Q2 out of circuit. A *uni* junction transistor only has one diode junction, that between the base and the bar of conductive stuff. The bar of conductive stuff looks like a resistor between the two emitter leads, and will measure the same resistance with either direction of the meter leads. The emitter is the only one of the three that will measure open circuit/reverse diode to either of the bases. Once you identify the emitter lead, there is only two possible orientations for B1 and B2.

My meter says I have it right.

If the orientation/pinout you're using is right, there's a bug in the circuit feeding the socket. Pull out Q2 and measure voltages on the socket pins. The B2 side should measure 35V, the B1 side should measure zero, and the emitter pin should measure the same as the emitter of Q1, and wire pads 11, 12, and 28. These are all connected by resistors and pulling Q2 gives them no current flow, so they all measure the same voltage.

The bases look fine. The emitter is at 33.9v, while the emitter of Q1 is at 35.2v.

As far as the trimmers go, I tried all the ones on the fall plate one by one with no success. Maybe I need to combine adjustments or something. Anyway, the quick fix failed, so I'll work on the animation before getting to the trimmers.

[Keppy]

Keppy,

What is the transistor that is called out in the original unit? I have drawers full of transistors from the last 50 years. Let me do some digging. 

I'm actually using the original type transistor - a 2N2646 ($3 from Small Bear). There's just some confusion about the pinout. Thanks though!

[pinkjimiphoton]

hi, R.G., yes, the trimmers are certainly interactive...other than r65, most i think need to be set around the mid point to find the "sweet spot".
if r65 is almost off as dino noted (again, my bad) the next one, after setting the rest mid-point is probably r31 - that seems to set the overall range of the exp pedal, and the working range of the filters in general...there's actually a couple different spots in the travel of that trimmer that go from no sound to sound, and even can make the exp pedal seem to work "backwards"...playing with that may help to define where the sweet spot is to get the formants working.

i'm suspecting from what keppy has reported, that maybe that 150 ohm trimmer is broken or open...he seems to be describing the same thing i did.

keppy, try using a test lead to short across the outside terminals of the trimmer...if you suddenly have formants, that may be the problem. IN the unit, mine at first read fine, then didn't read at all...when i took it out, it read like it was ok!! but it wasn't... as soon as i replaced it, the unit started working again. if that's the problem, it will show immediately by bridging those two contacts. i think it's just wired as a variable resistor anyways.

tomorrow i have a day off, gigging tonite too and the ludwig is in the trunk of the car with the rest of the stuff (tried to use it last nite at the gig, but all the lights and icemakers and crap had the power so dirty it was humming too bad to use), but tomorrow i will gladly open her up, and maybe we can figure out a time to meet via an instant messenger or something so i can take readings for you anywhere in the circuit that you feel may be helpful.

one other thought...do you have another transistor you can sub there? maybe the transistor is faulty? pardon my ignorance please, i am really just barely getting my toes wet in electronics, and don't mean to assume you haven't checked or anything...
but i've found sometimes in my limited experience that one transistor may work fine in another circuit, but may fail to work at all in the circuit i got it for. really weird.
sometimes, just swapping two resistors from socket to socket seems to make a difference.

mojo? beats me!! but i have faith you're gonna figure it out , it'll probably be one of those "aha!" things that makes ya slap your forehead and shake your head.

rock on!! dude...you've come so far, after people trying for literally decades to do this...you're gonna go down in the annals of history as the guy that cracked the nut!


or at least in the annals of the diy movement!

[Keppy]

Thanks for the encouragement Jimi! I'll try your suggestions when I get home.

Unfortunately, I don't have a second 2646 to substitute. I'll order one if no other solution presents itself.

The transistor seems to be intact. I measured a voltage drop of .8v from the emitter to either base, which tests normally when I reverse the leads. Seems normal, but I guess you never know.

[R.G.]

'Fraid not. The datasheet has the emitter on the middle pin. Looking at the schematic, your layout has it on the side closest to the fall plate board.

Bummer. I'll go scramble it now before I forget.

The bases look fine. The emitter is at 33.9v, while the emitter of Q1 is at 35.2v.

That's OK-ish.  I would have thought the emitter pin would be higher, but it's high, much higher than the 2V or so before.  So with Q2 back in socket, the emitter drops to 2V and stays there, right?

I'll have to go do some digging, but with E at 2V and B1=B2 = 2.47, the base/emitter is reverse biased and conduction should turn off unless the current through B1 and B2 is high enough to hold it on.

More when I get a spice model of the 2646 running. Right now the model refuses to work. A 2N6027 model works first time, but that's not the real device.  

[digi2t]

Hang in there Kep, we're all here for you. My unit has been apart since day one of this adventure, just so I can get into it easily and quickly for this project. If you need more readings, just say the word. Tell me what you need, and I'll whip it up for you pronto.

You're close man, I feel it. 99% sure it's some stupidity. With 2 units at your disposal, it can't be otherwise.

[pinkjimiphoton]

amen dat gents!! you're gonna find it keppy, i have absolutely no doubt...

and then you're gonna get all excited, cuz you're gonna have one of the coolest toys on the planet...


and even better, something you made yourself!