Do you recognize this PCB?

[EBK]

Marc (with a 'c'),
If you find a post where Mark (with a 'k') said that he can't wait to build one of these, you win a lifetime supply of sausages.   

[Rob Strand]

I think duck's "pin 2 modulated" and the style (Rick's) gave it away.  It's a "Bigger Angel".

Cool, someone here had to know!

[Mark Hammer]

Mr. Hammer is obligated to etch and build this now, right? 

Right now, I'm just waiting for the board to cool down, so I can peel off the PnP and stick it in the etchant.

This is amazing.  What I can't understand is how I somehow decided to print it out in the first place without saving the actual file.  As Pink Jimi P can attest, I save a LOT of layouts and schematics.  Looking through all the main and subdirectories for various chorus circuits on my hard drive, I found neither the layout nor the circuit drawing.  Weird.

Thanks to all who donated their time and energy to this particular scavenger hunt.  It's not like I really need another chorus, but I had been curious to make a 2399-based unit just to hear what it was like, and this will let me do just that.  So thanks again.  I owe you all one.

[Rob Strand]

his is amazing.  What I can't understand is how I somehow decided to print it out in the first place without saving the actual file.  As Pink Jimi P can attest, I save a LOT of layouts and schematics.  Looking through all the main and subdirectories for various chorus circuits on my hard drive, I found neither the layout nor the circuit drawing.  Weird.

One trick I use to find related stuff is searching by dates.

[Mark Hammer]

A useful tip, though not useful in this instance.  All I had was the PnP transfer.  I also have a nasty habit of doing whatever I can to shrink file size, such that if an image is copped or reduced from grey or colour to black and white, the timestamp is rewritten.

In any event, the board is now etched and tinned.  And thankfully, I have the information and parts I need to make good use of it.

[duck_arse]

as antonis might say - :beers: :beers:

and Mark - my first thought was Hyperflange.

[Mark Hammer]

as antonis might say - :beers: :beers:

and Mark - my first thought was Hyperflange.

Mine too.    But today needs to entail getting my car license plate renewed.

BTW, the parts layout diagram shows an awful lot more parts than are listed below it.  Is there any parts listing that is more complete than what is shown here?

[bluebunny]

Good spot!  I wonder what happened to the rest?  Anyway, I couldn't find anything else in Fred's blog or my stash, so I cross-referenced the schematic to the layout and came up with this:

R1  10K R2  10K R3  10K R4  10K R5  10K R6  33K R7  4K7 R8  220K R9  220K R10 100K        R11 100K R12 4K7 R13 47R R14 68K R15 330K R16 330K R17 330K R18 4K7 R19 10K R20 10K R21 10K R22 10K R23 10K

R24 100K R25 100K R26 4K7 R27 47K R28 100K R29 47K R30 47K R31 47R R32 47R R33 100K R34 100K RPD 1M 1 Depth 470K Lin Speed 100K Lin

C1  10nF C2  10nF C3  1nF C4  1nF C5  100nF C6  100nF 2 C7  47uF C8  100uF C9  10uF C10 10uF C11 10nF C12 100nF C13 100nF C14 10uF C15 100nF C16 47uF C17 100nF C18 100nF C19 100nF C20 10uF C21 100nF C22 10uF C23 100nF

Q1  2N3904 D1  1N4148 IC1  TL074 IC2  PT2399 IC3  78L05

¹ RPD in schematic, but not in layout
² C6 in layout, but not schematic (extra power decoupling?)

I've not double-double-checked this, so all the usual "YMMV" caveats apply! 

[bluebunny]

BTW!!!  I also noticed that the resistor labelled R12 (bottom left) should be connected to ground on its LHS, not to the base of Q1.  (So the layout is a little wrong!)

[Mark Hammer]

Thanks, bun, for both observations.  Much appreciated.

[bluebunny]

My bad: obscured trace underneath that resistor, so ignore my "BTW" about R12...

Actually, it's the trace underneath R12 that's the problem: it shorts out C14!

I think the PnP should look like this:

[Mark Hammer]

Built.  Made the trace change before I fired it up,  Works.  Thanks.

[Mark Hammer]

Is it just me, or do 2399-based chorus circuits sound like a modulated short slapback delay?

[EBK]

They sort of have to, don't they?  Such designs use every trick possible to reduce the delay time, but it is still too long for a real chorus/flange effect, right?

I suppose you could try using a second chip to create a delayed "dry" signal to get closer.

[Rob Strand]

They sort of have to, don't they?  Such designs use every trick possible to reduce the delay time, but it is still too long for a real chorus/flange effect, right?

That's how I see it.

Anyone tried interleaving the samples to half the delay?  I seem to remember you couldn't manipulate the clocks but it's a while since Ive looked at it.

[Mark Hammer]

I wouldn't expect flanging capability, but I was sort of hoping for a thick Pat Metheny sort of chorus, like I used to be able to get on my old blue MXR Digital Delay.

[Rob Strand]

Notice on the PT2399 the clock frequency when the VCO resistor is made smaller but the clock levels off when the external resistance is zero.  There seems to be an internal resistor around 4kohm in series with the VCO pin.

What if the VCO pin was pulled below the ground rail?   My first expectation is we'd probably only get 0.7V below the the ground rail as it probably has protection diodes.   For an experiment it's probably best to add another resistor (say 1k) in series with the pin to prevent the protections diodes frying; at least until we know there a diode there or not.   But if it goes lower then perhaps we can squeeze more out of it.

Maybe squeeze a bit more by playing with the supply rail and shifting Vref.

[anotherjim]

I did diode test the 2399 pins. Some show diode to the rails, some don't. Those that do are probably parasitic that are there anyway rather than for protection. The chip is very easy to latch-up with careless handling.

We do know the VCO can be swung via the pin2 reference - what if you tweak that at the same time as minimising the pin6 control? The LA does this but keeps pin2 at average 2.5v with AC coupling from the LFO. You could try adding a DC offset to pin 2 to keep the delay range shorter?

[Rob Strand]

I did diode test the 2399 pins. Some show diode to the rails, some don't. Those that do are probably parasitic that are there anyway rather than for protection. The chip is very easy to latch-up with careless handling.

That was quick! Probably not the way to go then.

We do know the VCO can be swung via the pin2 reference - what if you tweak that at the same time as minimising the pin6 control? The LA does this but keeps pin2 at average 2.5v with AC coupling from the LFO. You could try adding a DC offset to pin 2 to keep the delay range shorter?

That's probably worth trying.   I suppose it's a matter of how far you can shift it.  Maybe a factor of 2?  which would put things in a better zone from a delay perspective.

My guess is you could permanently shift the Internal VREF.  However if you *only* did that then the DC level at the internal outputs would also drop and you would lose signal swing.   So that can be "undone" by connecting resistors to ground on the inverting input -  the same way the NE570/NE571/NE572's use a low Vref and tweak the DC level on the outputs of the opamps.

The internal points around the opamps are all referenced to Vref so it should work.  Also there's no *internal* points which have analog voltage swings.  The MOD/DEMOD blocks probably have current outputs so they would also work.    So from that I suspect Vref could be shifted without losing signal swing.

The cons relate to the DC shifting.  We can no longer DC couple the PT2399 opamp stages since the DC levels are at VREF on the input side and Vcc/2 on the output side.   That can be fixed by adding AC coupling caps.

[EBK]

Can you supply your own clock on pin 5?