Help to understand PCB layout on Cry Baby clone

[Nordskov]

Hi there.

I have this Cry Baby clone "Wah Man" I had the idea of doing some mod on. But after tearing it apart, I realize it has already been modded.
I lack the skills to draw a schematic directly from PCB traces, so I cant really tell what have been done and what have not.
The circuit contains four transistors (2SC1815) so I guess an input buffer has been added.
All in all it looks factory modded as all legs are equal length and apparently machine cut except from the two orange caps and the diode, that have been soldered and cut by hand.
It looks like they got hold of a batch of PCBs and changed the component type and layout. The transistors are rotated 90 degrees according to mounting holes.
The two burgundy caps are 10nF each.






I'll appreciate if you guys can help me and tell what is up and down on this baby.

Cheers 

[lethargytartare]

My suggestion is that you take a look at a crybaby board that is the closest match (gcb-95, I think).  Then get a schematic for that circuit.  Then patiently trace your pedal from input through the circuit and compare to the crybaby schematic.  I bet you're going to find more similarities than dissimilarities.  And the soldering on the back doesn't look very professional near the bottom, so I'd guess an individual swapped in those orange drops.  You could also just compare component values between that and a crybaby -- they're numbered, so you can see how they compare pretty quickly. 

Once you find the differences, you can zero in on what might be going on, and might be able to tell what mods the person was doing.

[Nordskov]

Hi lethargytartare.
Thanks for replying.

I did study many schematics for a whole lot of wahs, including the gcb-95. And mod suggestions too.
And you are right that there are many similarities compared to gcb-95. But its the differences I am interested in. Especially that fourth transistor.
But as some of the components are not the same as printed on the board, they cannot be directly compared by number.
I've tried to do a schematic from the PCB on both paper and the PC, but I gave up pretty quickly. Too many traces and too little intellectual headroom
Those 2SC1815's do confuse me because of the leg layout (ECB).
Maybe you could just tell me if the fourth 2SC1815 is the buffer that R.G. Keen is talking about as an alternative to do a true bypass?
I think the rest is pretty similar except from the input section and component values in the effect section.

By factory modded I meant that the manufacturer improved the design with some of the mods out there. The caps and the diode are definately replaced by a less skilled person - who is not me

[MrStab]

Hi Nord,

Hi there.
The circuit contains four transistors (2SC1815)

are they ALL 2SC1815? is that a NPN transistor? AFAIK, at least newer CryBabies use the MPSA13 Darlington transistor on their input, though a lot of schematics online indicate a basic NPN (older models?).

so here's a suggestion: maybe 2 of those transistors are connected together to act like a darlington, because the manufacturer didn't have access to any real darlingtons? just one possibility.

[lethargytartare]

Maybe you could just tell me if the fourth 2SC1815 is the buffer that R.G. Keen is talking about as an alternative to do a true bypass?

No idea, really.  I'm not that well versed on the concepts like that.  But in your shoes, I'd do what I described -- it's not complex, it's just slow, detailed work.  I googled a pic of a crybaby/95 and on first blush it looks like the boards are identical and most of the components are in the same positions, and are numbered.  So I'd try to draw just those parts that are NOT the same, since clearly there'll only be a few.  That might give you the answer you're after.  I've taken the buffer OUT of a crybaby and converted it to true bypass, but that was some time ago, and I wouldn't immediately recognize those buffer components in a clone...without tracing that portion of the circuit...like I'm suggesting you take a stab at :-)  Good luck!

[MrStab]

i've had a longer look, and i'm even more convinced the first two transistors are configured in a Darlington Pair. the biggest giveaway is those two pins joining only to each other underneath the two transistors, and a rough attempt at tracing the region adds up to that. i can't see the pins properly, ofc, so i could still be way wrong.

there's a resistor to V+ (R2 i think? immediately to the right of the blue film resistor) right after the input cap, as opposed to the 1M8 to Collector then 1k to V+ in the regular CryBaby schem. maybe it's a deliberate adaptation to suit the 2-transistor setup somehow. too late at night to figure that out. well, it's probably still nighttime somewhere.

(using this for reference http://www.electrosmash.com/images/tech/crybaby/cry-baby-wah-gcb-95-schematic.png )

[Nordskov]

Hi MrStab.

Yes all four are 2SC1815.
After looking up darlington transistors I think you are right about that.
Let south tran be Q1 and north tran be Q2.
Input->10nF->1k7->collector Q1->collector Q2->+9V
Emitter Q1->base Q2
Base Q1->1M1 and 100nF->input zero
Emitter Q2->rest of circuit

What is the purpose of making a darlington constellation? Gain?
I guess it's not the buffer Keen mention, so I have to swap the switch for true bypass.

I had my eyes on the Wilde mod, but it seems this is the one the manufacturer aimed for. All the crusial component values are identical. Lucky me
But I guess I better change those ceramics caps to film, especially C1.
The cap leg jumper done underneath was because the copper trace delaminated and broke off.

Thanks for looking into this

[Nordskov]

I finally got the schematics down on paper thx thanks to your link, MrStab. Yehaa!
I take no responsibility for any errors
I see I missed the value of the resistor going to collector of Q4. Its 1k.
I have no idea of the value of the coil.


Three things makes me wonder.
1. There's no impedance resistor from signal to ground
2. The 100nF cap from signal to ground.
3. The link from collectors directly to +9V

[MrStab]

glad you're making progress! i think you may have the pins mixed up on the transistors - how are you reading them? in simple terms, usually the Base pin (input) is in the middle, but with the 2SC1815, it seems to be at the furthest side (see here: http://www.laro.com.pl/pdf/2sc1815.pdf ). the pins also seem a bit bent in order to fit in the holes, which might make it more confusing. in a buffer, the input never goes into the Collector or Emitter as you've drawn. for example:



i don't actually know why they chose the MPSA13 for that. i've read a few places that it's not essential and may have been caused by a mis-reading of "MPSA18" being passed down the line and never corrected within Dunlop. don't quote me on that, though.

in any case, i think you should KEEP that buffer and make it true-bypass anyway. good performance when it's on, true bypass benefits when it's off. maybe i misunderstood, just got the impression you want to remove it (some do).

[Nordskov]

Nope, the leg configuration is 1=Emitter, 2=Collector and 3= Base
Signal goes to middle leg on both transistors. Thats collector.
And your drawing of a buffer makes sense, but thats not the case here.
But as stated in an earlier post, the PCB seems to have been reused with another component layout.

I didnt want to remove it, but I thought I could skip true bypass.
Do you have any clue why they chose this darlington configuration?

Edit: 2SC1815 is NPN type.

[lethargytartare]

http://www.electrosmash.com/crybaby-gcb-95

This has a block diagram of the crybaby circuit, and the buffer section, with a darlington.  Again, if you do them side by side (there are tons of pics of crybabies on google), you're gonna find this is a direct clone of a crybaby.  I agree with MrStab -- the manufacturer simply didn't want to source real darlingtons (cheaper to just add pads to the boards than to source and install an additional part -- you have one unique tranny in there now, but adding a darlington would increase manufacturing costs).

And the parts appear to match up 1-to-1 -- so if you write out the component values with the board numbering, and get the same info for a crybaby, you can even map them by value and see if there are any differences there.

Google "crybaby true bypass mod" and you'll see how to remove that buffer.  You'll also see a few other component changes that then get recommended -- compare those to your parts list in that wahman, and you will see if you even need to make those changes.  What mods were you looking to do?

[Nordskov]

Hi letrargytartare.

Electrosmash.com is one of the sites I have studied very close, thank you. It's a very comprehensive walkthrough of the Cry Baby.
I was looking to add some switches and pots to vary key component values.
That is:
Q factor
Sweep range like this http://www.freestompboxes.org/viewtopic.php?f=11&t=286&start=180 (3. post)
Gain
True bypass
On/off LED
And if the pot is worn the Anderton optocoupler

Thats it

[MrStab]

something seems really weird here. if wired as in your diagram, Q2 would just kill the signal (i think) as Q1's Base is pretty much tied to ground. either i'm wrong (probably) or the we have the pins wrong somehow.

if you plan to remove it, it doesn't really matter i guess!

[lethargytartare]

Have fun -- so many cool things out there that are well documented!  Only suggestion is to do them one at a time slowly so you (a) can backtrack if anything in that board IS deceptively funky, and (b) so you can get a good feel for how they're interacting.  I used to do the true bypass, vocal, gain and volume mods to wahs whenever I'd get them.  You can also try to get your hands on different inductors and see how you like those -- you should see them crop up in the classifieds here often (and ebay, of course).  I'd be interested in trying to jam a few inductors in there with a switch to flip between them -- I've never tried a yellow fasel, just various stock dunlops and a red fasel.

Cheers!
ltt

[Nordskov]

Thanks for your input lethargytartare. I dont plan using a whole lot of cash on this thing, so swapping the inductor wont happen
Changing the switch for a 3pdt is costly enough, the rest are cheap components where a few dollars goes a long way.

@MrStab: Im sure I got it right about the pins layout. I checked and double checked it and checked it again because I think its weird too.
The pins that goes to the right are pin 2 on both Q1 and Q2. And if you look at the pictures they are connected to input via 10nF and 1M8.
But Q3 and Q4 have the right configuration and the four are identical. On the housing it says: C 1815 Y 5D.
Why the darlington pair is oriented that way I really dont know.


I have corrected some minor faults.

[duck_arse]

nordskov, as mr stab says, it can't be. if you follow on your draing from the "in", you come to a solid line connecting to the +9V. this effectively shorts ALL your signal to (ground) a low impedance point. further, your darlington base is connected directly to ground, so there will be no nothing happening there, as the transistor is switched off off off.

from recent experience, we must ask you - please look at the transistors fitted at the input section, tell us what part numbers are printed on. then tell us (again, if necessary, it'll all be in one place then) what datasheet you are looking at.

have you seen the article on wah technology at geofex?

[Nordskov]

Hello duck_arse.

We are all pretty confused here.
This is printet on the transistor housing: C 1815 Y 5D
If I look up "c1815 datasheet" I get this:
http://www.alldatasheet.com/datasheet-pdf/pdf/30084/TOSHIBA/C1815.html
and a lot more saying basically the same: NPN type, pin 1=emitter, 2=collector, 3=base

I have read the article at Geofex and a lot more.

If you look at the pictures in my first post, you can see that the two middle pins are connected directly to the +9V rail. That is the bottom trace seen from copper side.
Pin 3 on Q1 (south transistor) is routed to ground via 22pF and 2.2MOhms in parallel.

I have never heard this wah in action as I got it from a friend who thought it was dead. But it turned out the leads to the pot was broken off.
As mentioned earlier some of the components have been replaced. All other component legs are cut or grinded at the exact same height and parallel to the board. Its a smooth cut surface, not like if you cut with a pair of pincers.
That tells me the transistors are stock from factory.

I can take some detailed closeup pictures if you like.

Thanks

[Nordskov]

Hey I realize I forgot a trace!
This makes a lot more sense. Sorry for the gray hair that bummer cost you, I should have quadrupled checked those traces





Ok back to my previous question.
Does that darlington arrangement make it up for an input buffer or just extra gain? Can I eliminate the true bypass switch (=money)?

[MrStab]

just a quick driveby comment as i'm about to leave for band practice, sorry it's not really helpful, but maybe counterfeit transistors are a possibility. some Toshiba power transistors are faked, but i can't find anything on this particular type. i'll shut my hole now. lol

[lethargytartare]

I have never heard this wah in action as I got it from a friend who thought it was dead. But it turned out the leads to the pot was broken off.

I don't recall you mentioning that before, and it changes a few things in my opinion.

If the only problem was broken lugs, and you haven't heard the unit in working condition, then I think you're doing things out of order.  Your first task should be to make it work so you have a point of reference before you put in any mods.  Don't fix what ain't broke, and don't mod what you don't know you don't like.

You'll know what's stock and not by first comparing component by component between your wah and a stock GCB since the circuits are basically the same.  Then look at the places where the solder has clearly been altered.

BUT that is all secondary to making it work first and foremost if all that it needed was for the pot to be reconnected.

The board has pads for the trannys, so that makeshift darlington seems very likely to be original.

On your hand-drawn schematic, include the silk-screened IDs -- it'll help you when you finally start comparing parts to other wahs.

And, honestly, you're probably redrawing a schematic that exists all over the place already, so you should compare your work to a published gcb95 schem to see how accurate you have gotten it.

Now, you mention cost a few times, so that is why I keep hammering away at this:  make it work in stock shape first.  You may not care about the buffer, and may not care about true bypass.  The crybaby has the buffer and millions of people use it without modification, so you don't HAVE to make that change.  The gain and volume mods usually go with the true bypass.  Vocal and sweep mods are the way to go.

What's the pot/anderton optocoupler thing you have in mind?  I couldn't find anything that fit that reference?