noises with Sapphire Flanger (was add a killdry...)

[mr_doyle]

Hi all,

wondering if you could point me in the right direction.
I would like to add a "kill dry" function to a Morley Sapphire Flanger, so to get only the delayed signal and be able to connect it in parallel to the signal.

i'm looking at schematic here: http://www.morleypedals.com/sfbes.pdf

as far as i understand, could i simply add the switch after R21 to get the job done?

Thanks in avance!

D.

[Mark Hammer]

In principle, you are correct. The problem is that the overall bypass switch is set up so that if you hit bypass, both the delay AND the clean signal would be lifted from the output, and you wouldn't hear anything.

Seems to me that if you wanted to make that function feasible, you should probably install a 3PDT stompswitch, using 2 sets of contacts to switch/bypass the input and output, and the third to engage the status LED in the manner shown in the schematic.

Whether you NEED R27 before the bypass switch is something I can't answer, but its presence suggests it is there to complement R26 and achieve some degree of volume balance.  Personally, I'd suggest omitting R27 and replacing R26 with a 5k volume pot or trimpot.  That oughta do it.

[mr_doyle]

Thanks Mark!

I don't mind about the bypass, as the pedal would be switched in/out by a remote looper, so it would actually be always ON.

I'd install a switch to toggle the dry signal ON/OFF for the times when the pedal will be into a standard series chain.

Thanks for the resistors suggestion as well, i'll try removing them, see if i can get some more output out from the pedal...

D.

[mr_doyle]

works well, signal is hotter also!

now i only have to find a way to get rid of that metallic ring... :-O

D.

[Mark Hammer]

works well, signal is hotter also!

now i only have to find a way to get rid of that metallic ring... :-O

D.

Great!  You can propbably address the "metallic ring" (from higher regen settings, I imagine) by reducing the value of either C13 or C14 or both tolimit the low frequency content being regenerated.  It's that emphasis at room-like resonances that gets so irritating with many flangers.  Try chopping them both down (or simply one if its awkward) to .047uf and tell me if that moves more in the direction of what you want.

[mr_doyle]

Mark, you're da man :-)

i lowered C14 to .47uF and it's much sweeter now!
i'll replace C13 as soon as i can find another cap with that value (local shop in town is out of them)

Thank you so much!

D.

[Mark Hammer]

It's the culture and ghost of audio design invading our dreams and seeping in through the cracks my friend.  There are so many times and places where full bandwidth is NOT what the doctor ordered, yet somehow we slavishly pursue the goal of having as much bandwidth as can be achieved because audio design principles drag us there.  If we can't have it at the top end, we can at least have it at the bottom end.  Well that sounds plausible, in theory, but it works a whole lot better in the context of sound reproduction than in the context of music production.  This minor change to the regen loop that you did amply illustrates that.

By all means, read all the audio design stuff you can get your hands on.  Buy Audio Express, and back issues of Glass Audio or even Absolute Sound.  Save up and buy your own copy of Glen Ballou's "Audio Cyclopedia" for cryiong out loud.  But when it comes to what makes a pedal sound good, forget 50% of what you've read for the moment, and ask yourself "What do my ears need to hear for this to sound as conceived?  for this to sound *better*?"

[mr_doyle]

modding time again :-)

i'm fighting against a lil, bad noise.

It's a sort of zippering, and pulses at the same time with flanger speed (say one "zip" at the highrt point and another "zip" on the lower one).

"Zip" duration depends somewhat by the "travel" pot: at lower settings it's longer, and shorter on higher ones (become more as a fast "zap").

I tried to disconnect the LED removing R35 but it was not the cause.

Any input on this? Could this happen because of the grounding of the CD4007 or CD4047?

An now the second part: i've seen that the enclosure was originally not connected to the board at all.

Now i'm wondering if could i do anything as:

- connect the enclosure to the PCB with a wire to improve grounding
- connect the enclosure to the PCB with a cap () to improve grounding and RFI shielding (any sense?)
- separate audio grounds and power grounds: use the enclosure as shield and make the pedal balanced

As always, any help appreciated.

Tnx,
D.

[Mark Hammer]

Sounds a bit like clock noise.  It is possible that the routing of the wires for the dry-kill switch have come perilously close to clock-related segments of the circuit.  Those wires should probably be shielded or relocated.

Somewhat like the old A/DA Flanger, the Sapphire has more low pass filtering on the regen signal than it has on the delay signal per se.  Indeed, the only filtering I can see is provided by C9, which isn't a whole lot when you think of it.  Try tacking a second .001uf or maybe a .0015uf cap in parallel with C9 on the copper side and see if that makes any useful difference.

[mr_doyle]

Thanks Mark,

i'll try to add the cap and let you know.
Btw - the kill dry was achieved just removing a resistor, no wiring involved.

D.

[mr_doyle]

So, i tried to add the second cap (and also a third one) and, while there was a certain decrease in background hiss, the ticking noise was still there.
I used disc caps, if that matters.

I noticed that the noise takes a few seconds to begin once i power up the pedal and it's "almost" always there only if i plug a cable into the pedal input.

D.

[mr_doyle]

further note: setting the "travel" pot at minimum (or so) transforms the ticking "zip" into a constant sort of squeal...

also, regen has no effect on this noise (i'm keeping it down to 0)

D.

[mr_doyle]

ok, still trying to solve the problem...

after further tests, i discovered that this "zip" get worse/better... quite odd... so now it seems more a kind of hi frequency or RF noise... as soon as i touch some components on the PCB (specially caps) noise is worse.
I can also hear microfonics if i tap some of the electrolitics with a screwdriver... is this normal?

again, thanks to anyone who wants to help.
D.

[Mark Hammer]

That inserting a plug into the input enables the noise means that it is related to the audio input stage, since standard clock noise would be there with or without an input.  You will note from the schematic that the input jack shorts C1 to ground with nothing inserted into J1.

That it takes a little while to start up suggests that it is related to an electrolytic cap of some size.  For reasons that are mystifying, you will note that while C21 is rated at 50v (much higher than it needs to be), C20 is shown as rated at 16v.  Were it to have only encountered 9v batteries within its lifetime, that would likely provide all the margin of safety needed.  Might it have had something higher than that voltage accidentally plugged into the adaptor jack that could have led C20 to be off spec?  Possibly.  The fact that the Travel control setting alters the noise and that Travel simply taps a DC voltage from C20 also suggests that incomplete filtering may be the culprit.  Try replacing it with another cap of 220-470uf and see what happens.  BTW, are you currently using it with a battery or an adaptor?

[mr_doyle]

Thanks for your help Mark.

Noise is there also without anything plugged in the input jack, it just comes after a while.

It's also there with either battery or adapter, makes no difference. It could be that some wrong power supply has been used in past, got the pedal used so i can't tell.
A cap failure is an interesting possibility: as i wrote, there are a few microphonic caps, and also noise increases if i touch a few of them.
I'll do a better investigation and see if they are all "casually" rated at 16V, eventually replacing them and report back.

Thanks again,
Doyle