cool reverb apart from switch spdt 'pop' issue

[samhay]

I can hear why you want some feedback in tails mode.

Your new schem is missing the 10k resistor to C5. Is that still there? In the Echo Base, the feedback is mixed with the dry signal (which is lifted in bypass) just before the PT2399. This is similar to the latest schem I posted.

[slacker]

The "tails/direct/flashback mod" could be done if you want to go down the buffered bypass only route, no true bypass option. Or you could go with flashback/tails/true bypass.

For those playing along at home flashback is the opposite of tails, the input of the delay is always connected bypass disconnects the output. Direct disconnects both the input and output so the delay behaves in the same way as if you true bypassed it.

To be honest for a reverb I'm not sure how much use the flashback option is, I suppose for over the top ambient reverbs it might be good, you can kick it in and immediately get a big  wash of sound, for short reverbs I'm not sure you'd notice the effect.

[deadastronaut]

morning!..

@sam: corrected schemo..well spotted.




@ian: i'm beginning to think buffered might be the way to go then..as long as i can have the long oscillating tails ...

'' Or you could go with flashback/tails/true bypass. ''....hmmmm...

[samhay]

Let's think about where the signal goes in this circuit (x's are where you have FET switches; sorry about the crappy ascii art):

In -> buffer --------> mixer1 ---> out
           |                    /\      |
      x1-|                     |      |-x2
          \/                    |      \/
        mixer2 -> brick --|      |
           /\                           |
           |----------------------|

In your original tails mode, you killed dry signal going to mixer 2 at x1. However, when you had feedback up, the signal from mixer 1 was still being fed into mixer2 via the feedback loop.
If you lift the feedback loop in tails bypass (at x2) then you don't have any signal from mixer1 being fed to mixer2. This keeps dry signal out of the brick, but removes much of the 'tails' effect, as you don't have any feedback anymore. I am not sure how buffered bypass would help.

One options is like this (the latest schem I posted):

In -> buffer --------> mixer1 ---> out
           |                    /\
        x-|                     |
          \/                     |
        mixer2 -> brick --|
          /\                    \/
          |-----------------|

Now, we split the wet signal from the brick. One output goes to mixer 1, the other to the feedback loop and is mixed with dry signal at mixer2. In tails bypass, you kill signal from the input buffer only. The feedback loop is still active, but now is not mixed with dry at mixer2.
The feedback will sound a little different, as you are not mixing it with dry signal (via the dry and reverb pots) at mixer1. However, you might find this works better this way as the feedback will be more usable at low reverb settings.

[deadastronaut]

arghhhhhhhh my head hurts trying to think of/realize it... 

[samhay]

You mean my terrible diagrams didn't help?
It should be minimal work to try the last schem I posted on the breadboard.

[deadastronaut]

lol. yeah...

this one you mean?

http://www.diystompboxes.com/smfforum/index.php?topic=101126.msg939394#msg939394

[samhay]

yup - another one of my masterpieces.

[deadastronaut]

http://www.diystompboxes.com/smfforum/index.php?topic=101126.msg939394#msg939394

hi sam, just tried that ^ but it cuts off the decay very abruptly still...

[samhay]

Hmm, that's puzzling. What happens if you move R2 to in between the FET and C5?

[deadastronaut]

yeah thats the same,

( it has cut the reverb decay/feedback from being full when on..only short decay)


as a note i decreased the 33k feedback resistor right down but it will not go into oscillation like usual..

[samhay]

OK, need to have a think about this.

[slacker]

Can you do a quick experiment Rob. Using this schematic http://www.diystompboxes.com/smfforum/index.php?topic=101126.msg939391#msg939391 but without the Q3 switching so you aren't killing the feedback in bypass, if you set it to 100% wet on the wet/dry do you get good tails when you bypass it, obviously you'll get no dry signal in bypass but does the reverb tail off nicely?

If that works Ok the easiest thing to do might be to add a bypass around the whole thing, that way you keep all the functionality and sound of the pedal but you can have tails, just means another opamp or two.

Actually I've just had a cunning plan, I'll be back........

[deadastronaut]

ok baldrick , , i tried it fully wet, and yes i got excellent  trails with the fet switch off......(and no original signal as to be expected..)

[slacker]

Exellent. Here's the cunning plan, using that same schemo try this mod to the Q3 switching.

This adds a delay before the fet turns off when you bypass it allowing feedback to carry on for a while. Experiment with different values for the 100k for different delay times. Simulator shows some distortion but it might work ok.

[deadastronaut]

aha...nice one man. neat trick..

cool, that works, i used a 1M...and it gets the same decay/trail as when on/off...brilliant. 

[deadastronaut]

so now we have...

[deadastronaut]

one thing though...

when i have wet/dry set to dry ...and switch to tails, i have reverb on guitar (not as much as full reverb) while tails are going on....then when tails finally die off, the guitar signal becomes dry....

[slacker]

Cool, nice to hear it works. Yeah unfortunately you're going to get that because then your feedback signal includes the dry signal, the only way round that is to take your feedback from somewhere before you mix the dry back in. Or bypass the whole shabang, that will definitely work, it feels like a horrible bodge though.

Have you got a datasheet for the brick you're using, the one I can find online appears to be for a different version?

[deadastronaut]

Heres the datasheet for he btdr-2h...

http://www.mammothelectronics.com/v/vspfiles/assets/images/btdr.pdf