Idea: Delay with insert point?

[composition4]

Hi, had a thought the other night. Just wondering if it's already been done or whether it's not a very useful idea?

As you'd know, most (all?) delays pass the original signal and also a delayed signal.  What if a delay had insert points so that you could plug in a flanger/wah/octaver/whatever so that the original signal goes through unaffected, but only the repeats are flanged/etc?

If I'm not explaining myself, here's a really poor diagram:




Or maybe even a delay with insert points like above, but the insert points are in in between the repeats. Take for example an octave-up as an insert.  The first repeat is an octave up, then the second repeat is an octave up from the first, the third repeat runs through the octaver again and is ANOTHER octave up, etc. So the repeats get higher and higher in pitch, until the delays tail off. Or lower. Or more flanged, etc.

Sorry the last one is probably even more convoluted than the first.

Are either of these ideas better in my head than they would be in real life? Thoughts?

Jonathan

[sean k]

I just built a echobase with the PT2399 chip and it wasn't working and I did all sorts of worrying till as a last resort I started audio probing the circuit and learned that the thing did work. Not only that but I quite like the signal straight without lots of chopping down. I also learned why it's chopped is because the delay only does one repeat and multiples are because it keeps going around so it needs chopping of the highs to stop it getting overdone with chaf. Also this means I've learned something new and really need to view the datasheets to understand whats going on which is something I wouldn't have bothered with if the effect worked like I hoped it would.

So yes you can do what you ask and it is actually mentioned in the echobase thread.

What I also noticed was that I really liked the signal outta pin 12 of the PT2399 as opposed to the signal ued off of pin 14 and have started thinking about the phase of these two signals and why one is used and the other is cut back... Hmm, back to the datasheets, which I got from here http://sdiy.org/destrukto/vc-echo.html Oops, thats the datasheet for the PT2395 right there at the top but the datasheets pdf for the 2399 is further down on the page. Here it is http://sdiy.org/destrukto/datasheets/PT2399.pdf

Just to add, I did some awful things to my 2399 chip before I got it resting on 5V but it's still getting signal and delaying it, as far as I can tell, so it's a fairly hardy chip... hopefully 

[SISKO]

The first repeat is an octave up, then the second repeat is an octave up from the first, the third repeat runs through the octaver again and is ANOTHER octave up, etc.

I like  this. 
I think im gonna tray it. But before a need to build an octave up

[Mark Hammer]

The first repeat is an octave up, then the second repeat is an octave up from the first, the third repeat runs through the octaver again and is ANOTHER octave up, etc.

I like  this. 
I think im gonna tray it. But before a need to build an octave up

The problem here is that, unless you maintain unity gain across all repeats, you will run into problems of many types.  Consider that unless set for runaway feedback, the signal reachig the octave-up unit will receive a signal that continues to diminish in its "drive" capability with each repeat.  That will be true whether the insertion point is before or after the feedback control on the delay.  As the drive hitting the octave unit decreases, the capacity to actually produce an octave will decline.  Given that the actual octave is usually much lower than the fundamental, the odds of getting a 2-octave shft on the second go around are slim, and the odds of something on the 3rd go around negligible.  If it was a digital pitch shifter, then you'd have much better luck, but using an analog octave fuzz will prove to be frustrating.

Of course, if you set the delay unit for all-out feedback mayhem then you will need to keep a close watch on the fuzz output level.  The likelihood of overloading the delay chip is very high , and the amount of harmonic content being added to the signal on 2nd and 3rd iterations will turn the signal into mush very quickly.

At a theoretical level, it sounds like a great idea (and read Craig Anderton's review of similar setups in the A/DA floor model Harmnizer in DEVICE - http://hammer.ampage.org around page 11 for confirmation that you had a good idea), but it is really only feasible in the digital domain where pitch changs are not dependant on signal level.

[composition4]

Thanks for your replies, guys...

Mark, good point about the analogue octave-up. Would be interested to try this with a digital pitch-shift though, as you said.

I looked into it a bit further and it turns out my idea wasn't so original (and it's not just Craig Anderton with the same idea..).  The Keeley mods for the Boss DD-3 have this very mod, he seems to call it an effects loop within the delay. Which is probably a much better description in a few words than I managed for my whole post 

I came across some guy's  (no idea who he is) soundclick page where he plays a modded DD-3 with a flanger in the loop, like the idea I had -

http://www.soundclick.com/bands/default.cfm?bandID=190415&content=music

Listen to Gilmour's Island then 3rd Jam from the Sun. He doesn't really give the playing enough space for you to hear it properly but the start of the songs have enough space for it to be heard. I like the idea of the flanger.

Anyway I'm looking for a second-hand DD-3 and once I find one at the right price I'll have a tinker and try out the effects loop idea.

Jonathan