cool reverb apart from switch spdt 'pop' issue

[merlinb]

Pretty sure we tried at some point. I think there is a potential problem with phase cancellation (despite the delay).

The sounds unlikely to the point of impossible. I suspect you were experiencing some other effect.

EDIT: I found the data sheet, now I see that pins 5 and 6 are stereo outputs. For the stereo to serve any purpose they must be different in some way, so the phases from the two outputs must be different to some extent. You could experiment by taking the feedback either from pin 5 or pin 6.

Looking back at some of the comments, I think some of the difficulties encountered in this thread may be due to undiscovered construction errors, rather than design errors.

[samhay]

Pretty sure we tried at some point. I think there is a potential problem with phase cancellation (despite the delay).

The sounds unlikely to the point of impossible. There are so many delays inside the Belton that any sort of consistent phase cancellation is a highly dubious problem. I suspect you were experiencing some other effect.

I have to admit that I was a more than a little confused/surprised at the time.

It would make sense if there was dry signal in the feedback loop. It is certainly worth revisiting.

[deadastronaut]

morning,  where were we?..

the way i see it is the only problem with my breadboarded version is the fet switch not actually cutting off all the reverb in bypass...

ive tinkered with it to no avail..

clip.

reverb...then bypassed..
https://dl.dropboxusercontent.com/u/7464107/reverbfet.mp3

[samhay]

when the feedback is set to minimum, do you still get reverb in bypass?

[deadastronaut]

when the feedback is set to minimum, do you still get reverb in bypass?

no, not a hint.

[samhay]

Just to be double sure, if you set feedback to near maximum and wet/dry to fully wet, do you get any reverb in bypass?

[deadastronaut]

if i set feedback to max, and fully wet i get no sound at all in bypass

full wet reverb when on.

[samhay]

OK, so what you are hearing in bypass is the dry signal at the output being fed via the feeback loop to the op-amp  buffer/mixer in front of the brick.
While this might give you the best sounding feedback, it is just not going to work for a tails bypass.

The schem I posted yesterday lifts the dry signal from the feedback in bypass. This will cure this particulalry problem, but will quite possibly make the feedback sound very different in bypass, and mostly negate the point of having tails.

How badly do you want tails?

[deadastronaut]

How badly do you want tails?

hmmmm...i see. 

would be a really nice option though..

[therealfindo]

^ drop it to 10k then..should oscilate like a bugger..   (btw there is no 68k in my layout.)

which layout/pcb?..

This one: http://www.diystompboxes.com/smfforum/index.php?topic=101126.msg917324#msg917324

*cough*

Any ideas why I'm not getting the oscillations? 

[deadastronaut]

^ looking at R O trees schematic version it should be fine... the 33k swapped to 22k should osc like hell.

is your feedback pot ok?....is it doing anything?.

[deadastronaut]

as a blatant kludgey bodgey type solution i just tried this ultra stupidly simple (like me) approach......bear with me on this..

input...middle lug dpdt...

one throw to reverb in..

other straight  to out.

use the other side of dpdt for leds  red/green..

true bypassed, with tails...it works.

any forseeable problems?..

[deadastronaut]

well it works...33k then 22k osc..

https://dl.dropboxusercontent.com/u/7464107/tailsdpdt.mp3


only thing is when turning volume down on guitar it turns down reverb tails too...thought it was too simple..

anyhow..

valoosj came up with an interesting idea..

Dear Rob

At the risk of sounding stupid (lack of electronics knowledge  ): wouldn't it be possible to add a mixer to your reverb?
When reverb is not desired but tails are, send the input signal straight to mixer input A and the tails to mixer input B. When switching the reverb on, grounding mixer input A and sending your dry signal in the reverb and then to mixer input B.
And if you don't want tails, cut the sound going to B when turning off the reverb.

To my untrained mind, this seems logical. Though, if it were a good idea I'm sure you or Samhay or Merlin would have proposed this.
I hope you get it sorted out because the reverb sounds killer.

Cheers

Yorick

thanks yorick:  

[samhay]

I'm not sure if I quite understood correctly, but I think Yorick is suggesting something similar to my suggestion - lift the dry signal to the feedback loop/reverb in tails mode.

Rob - your idea of leaving the output always connected is not a new idea. It can work, but you have a voltage divider between the guitar pickup, the feedback loop and whatever the effect is plugged into. I guess this can get a little unpredictable.

[therealfindo]

^ looking at R O trees schematic version it should be fine... the 33k swapped to 22k should osc like hell.

is your feedback pot ok?....is it doing anything?.

Yep, the feedback pot is working.

I swapped the 68k in the pcb with the 22k..
oh, wait, just a thought... perhaps I've wired the spst switch backwards..

[deadastronaut]

I'm not sure if I quite understood correctly, but I think Yorick is suggesting something similar to my suggestion - lift the dry signal to the feedback loop/reverb in tails mode.

Rob - your idea of leaving the output always connected is not a new idea. It can work, but you have a voltage divider between the guitar pickup, the feedback loop and whatever the effect is plugged into. I guess this can get a little unpredictable.

yeah thought it was too simple... ..

@yorick: have you a schemo of what your idea is?....i'm a little confused now too.

[samhay]

I swapped the 68k in the pcb with the 22k..
oh, wait, just a thought... perhaps I've wired the spst switch backwards..

The TO Tiree schematic switches a 68k resistor parallel with a 33k resistor. 68k || 33k is about 22k. Using this layout, if you replace the 68k resistor, you will end up with a 13k equivalent resitor in the feedback and it will probably go nuts.
As I see it, you have 2 options, once it is wired up correctly:
1. put the 68k resistor back in
2. remove the switch and 68k resistor and replace the 33k resistor with a 22k resistor.  

[Valoosj]

I'm not sure if I quite understood correctly, but I think Yorick is suggesting something similar to my suggestion -

In order to visualise what I meant:

[slacker]

Yorick's idea makes perfect sense, it's what my earlier suggestion to "bypass the whole thing" amounts to.
This solves all your issues, you can have your wet/dry mix and feedback with dry setup. It just seems like admitting defeat to me, as we should be able to make it work without resorting to that.

Here's a schemo that basically does what you need.
http://www.diystompboxes.com/smfforum/index.php?topic=92036.0

You've already got the first buffer so you don't need that. This won't give you the option of true bypass, you'd need to use electronic switching in place of the switch if you still want that.

[samhay]

Am I missing something, or is there is no feedback in Yorick's schematic?