Tyco Pedalflanger

[alanp]

Impressed... it looks like a sewing machine ran down the side

[Govmnt_Lacky]

WOW!!!

A lot of creative inspriation has hit the armdnrdy household lately!!!

I already have enough to build Larry.....   

[armdnrdy]

I reworked the Pedalflanger schematic for negative ground. I figured I might as well before I take the plunge to route a board that will fit inside of a Dunlop style wah enclosure.

I could use another set of eyes to verify the changes.

This is what I changed:

I replaced the 79L05 regulator with a LM317LZ, set for +7.3 volts.

I connected +V to VDD pins on all ICs and negative ground to all VSS pins.

Reoriented all electrolytics for correct polarity for negative ground.

Replaced P channel JFETs in limiter and gate sections with N channel JFETs.

Changed -VR to negative ground on components in limiter and gate sections.

Changed BBD VGG voltage divider for 14/15 VDD (+6.8V)

Connected R36 & 37 (at VCO, IC6) to ground.

Here's the schematic with negative ground changes.





Positive ground schematic for comparison:

[Fender3D]

R18 and 19 should go to GND
then what went to VR should still go to Vr (gate and limiter)
If you wanna use a NJFET D3 and D4 should be reversed

[armdnrdy]

Thanks Federico....as always,

I made the suggested changes and reposted the schematic. (Top image)

Changed the BBD output resistors to ground.
I inverted C4 and C6, and reversed D3 and D4.
Changed the grounds back to VR.

I really don't understand the limiter and gate sections....is the N channel JFET I changed from a P channel okay in the limiter section?

[Fender3D]

hmmm...
maybe C8 should go to GND and the gate switch to a resistor to +V...
and swapping FETs probably will require a new bias voltage

the limiter threshold is like bias trimmer in P90s, maybe Q2's emitter should go to GND with a NJFET, why don't try a PNP?

P2, D1 and D2 look strange, tied to GND without a capacitor from IC1b out

[armdnrdy]

I based the first changes I made to the limiter circuit off of this drawing of a limiter that I found on the net.



Looking at it now with a better understanding, The op amp is fed a bipolar supply, which would make T1 drain and T2 emitter connected to 0V or in the flanger circuit....VR.

The original circuit (second drawing above) uses P channel JFETs. I switched them to N channel assuming that they needed to be changed with the VR change from negative to positive. Wrong assumption?


P2, D1 and D2 look strange, tied to GND without a capacitor from IC1b out

That's how the circuit is originally. It was traced by Dirk Hendrick, as well as a different drawing...very hard to read, and I verified this circuit with quality, detailed pictures.

[armdnrdy]

I found a better example using a P channel JFET. It's a very similar circuit.....5V and gnd feeding the op amp, with all of the limiter components going to ground.

[bean]

R41 should connect b/w adj and out pins, I believe. I don't see why you would need to run the circuit a 7.3v, however. IMO, it would be better to run it off 9v, then use a resistor and 8.2v zener for the VDD supply. Reduces the part count and PCB footprint without that LM317. Then connect R16 to the new VDD to drop it to VGG voltage. Also, usually a small diode like 1n914 works well enough for that without using a voltage divider but either way (voltage divider is better of course). Your audio, LFO and clock all share the same power rail. I would create a second rail just for the LFO and clock/buffer section.

[armdnrdy]

Brian,

Good catch on R41! Corrected and reposted.

I contemplated a different power section but decided to build this design as close to the original as possible.

There is a limiter at the front of the circuit....so adding a volt or so of headroom really isn't going to make a difference.

During my extensive research on the Pedalflanger, I never came across a single hint of distortion or volume drop issues.
I think that great care was taken in the original design to offset these issues.

Adding a second rail for the LFO and clock/buffer section would increase the board size.
As it stands, I have to "fight" to work out a board layout with on board in/out and DC jacks that will fit inside a Dunlop style wah enclosure that has a switch installed in the heel section. 

My main goal is to convert the existing circuit from positive ground to negative ground.
I welcome all suggestions!

[Govmnt_Lacky]

As it stands, I have to "fight" to work out a board layout with on board in/out and DC jacks that will fit inside a Dunlop style wah enclosure that has a switch installed in the heel section. 

Dont forget that the gearing is reverse from that of a standard Crybaby. So, either the pot will need to be installed on the opposite side as a normal Crybaby or a work around will need to be figured out with the resistance. 

Or I could be completely WRONG... 

[Fender3D]

Hey Larry.

I'd breadboard it first, because some circuit block seems strange if not wrong:

ie.
Q3 won't stop wet signal when activated, it rather increases regeneration, connected as is...
I guess the node C9, C10, C11, R14 and Q3 (Source?) might be connected to Vr...
this way C11 will act as low pass, expecially since you have no LP filtering before BBD
and, with a NJFET, gate switch should go positive (maybe, at least, to Vr) for gate bypassing...
Again I'd connect Q3 (Drain?) to R10, Tr4, C17 node;
so you'll gate out the wet signal only
Then P2, D1, D2 force IC1b to GND if not DC decoupled...

[armdnrdy]

Federico,

I'll have to look this over a bit. I was thinking to breadboard the circuit sections that I'm not sure about to physically see what they're doing....and to learn what they're doing.

Greg,
Upon first, second, and third inspection....I thought that the gearing was reversed as well.
The pot is oriented 180 degrees from a "normal" wah but......the rack gear is oriented on the opposite side than a normal wah.
Which means...both configurations move the pot CCW in the toe down direction.

Problem solved before there was one! 

[armdnrdy]

Federico.....

You are an electronics super hero!

The two items that you kept pointing out were indeed incorrect! (The positive ground schematic above, is reposted with corrections)

The C9, C10, R14, Q3 node does connect to VR. The connection was a top board trace, under C10.

Dirks circuit trace drawing is incorrect. I originally connected it in that fashion, then changed it to match Dirks drawing. I figured that Dirk was tracing the actual pedal....I'm just working off of pictures.

The other issue....D1, D1, and P2 was incorrect on Dirks drawing as well...but I missed it.
Working back through the images of the circuit showed that there is a 10n cap in between R4 and the diodes to ground.

Well...I had to go back to the drawing board on the positive ground drawing, now back to the negative ground conversion.  

Thanks again Federico! You've been a great help as usual!

[armdnrdy]

I finally found the time to throw this circuit on the bread board.

It fired right up but the effect sounded subdued...very unremarkable.

I worked with the circuit and reviewed the trace and component images (double sided board) and found that Dirks circuit trace drawing had the "Intensity" control connected to the wrong part of the output. (positive ground schematic above revised)

I believe that's what Federico was pointing out.

The circuit sounds pretty cool!

There is a limiter that allows you to slam a chord hard without the threat of clipping the input while remaining very transparent.
(one may want to tuck this low parts count limiter away for other uses)

The noise gate actually gates the whole output instead of just the delayed signal. (I've checked the circuit many times to verify that it's correct) The gating happens at the very end of the audible signal, so not to interfere with the natural decay of the notes being played.
The reissues have a switch to cut the gate which I've verified as working. (Q3 gate switched to positive ground deactivates the gate)

Still a lot of work to do!

Since I have it on the bread board.........time to experiment for a negative ground version.

[Fender3D]

If this:

...I'd connect Q3 (Drain?) to R10, Tr4, C17 node;
so you'll gate out the wet signal only

maybe lowering R20 and rising TR4 will help better, decoupling wet and dry...

OT
does Geezer Butler use this pedal in "The sign of the southern cross"?

[armdnrdy]

If this:

...I'd connect Q3 (Drain?) to R10, Tr4, C17 node;
so you'll gate out the wet signal only

maybe lowering R20 and rising TR4 will help better, decoupling wet and dry...

OT
does Geezer Butler use this pedal in "The sign of the southern cross"?

I'll mess with it since it's on the breadboard. The signal is very low by the time the gate turns on....right at the point of not being audible.
With that being said....at such a low level, having the whole signal gated really doesn't sound bad.

I listened to The sign of the southern cross.

There is mention of Geezer using a Tycobrahe Parapedal and mention of Iommi using a Pedalflanger...so I wouldn't doubt it.

If he used it on Southern Cross...it almost sounds like there was a separate track of the Pedalflanger on top of a dry bass track.
I say this because if you listen closely to the bass track...it doesn't sound flanged. The airy, almost wind sound, frequency flange is riding up above the Bass track.

[pinkjimiphoton]

wow, nice work larry!!!!

[armdnrdy]

wow, nice work larry!!!!

Thanks Jimi.

I tried Federico's thought on moving the gate to only cut out the wet signal without changing the trimmer and R20 value.

It didn't change anything. The whole signal was cut at a very low audible level.

One thing that I've noticed while calibrating this circuit is the BBD output level trimmer is very interactive with the intensity control.

The user manual states that the Intensity control oscillates past the 3/4 position.

If I change the components and adjust the BBD output trimmer to make the gate work at the BBD output, that would probably affect the normal operation of the Intensity control.

The original flanger gates the whole signal and it sounds fine.... so.....I think I'll leave it as is.