Rebote 2.5 delay with tails and modulation (another epic journey thread)

[deadastronaut]

ha ha, that 4.5 seconds was a joke clip....gotcha!...

anyway, i'm having problems putting the middle pt in there...and getting it working.

i have the dual delay working fine, but am obviously doing something wrong in connecting the middle one in series.

1: should i be 'duplicating' the first pt entirely?..

2: or is there an easier, less component way of squeezing it between..?

like i say i have the dual delay working on one breadboard, and on  another breadboard i setup another 5v reg, etc..grounds jumped over to first breadboard..

just for more space etc..so i can wire it in.

i'm a little lost...


this thread should be called '' 3 or 4 pt2399's in series''

to get a clean long delay...

[haveyouseenhim]

ha ha, that 4.5 seconds was a joke clip....gotcha!...

Man, you are a grade A bastard!

[deadastronaut]

ha ha...funny though..

but that's the aim of this...its about time someone bit the bullet and made a better delay out of these cheap ass chips in series.

tubeguys double is nice, much better/cleaner than the 1 pt.....but while its on breadboard i figured why not try it...

[samhay]

anyway, i'm having problems putting the middle pt in there...and getting it working.
i have the dual delay working fine, but am obviously doing something wrong in connecting the middle one in series.
1: should i be 'duplicating' the first pt entirely?..
2: or is there an easier, less component way of squeezing it between..?

I would go with something like Fig 3 here http://sound.westhost.com/project26a.htm for a minimum requirement for the middle PT2399(s).

If you want more than 2, then your options to control the delay time become quite limited - I would vote for a current miror like in the madbean design.

[deadastronaut]

cheers sam, i'l try that..

i duplicated the first pt, and it works now...but hissy...sio i'll give that a go..

the cleans are very clean though...really nice. 

[psychedelicfish]

If you've got any small value resistors, try connecting the digital grounds of each chip to the analog ground through a small resistor.

It may also be worthwhile putting a smaller cap, something in the region of 100nF, in parallel with the 47uF cap on pin 2 of each chip.

[Kipper4]

Ok Rob
I might just build the Chasm instead for now . until this comes to completion.
Have fun

[deadastronaut]

ok, i'll try those suggestions, , cheers ed and sam..

heres' what ian suggested by pm..

pt1: input from fig 2/1  but with output filter fig 3...

pt2: fig3

pt3: input filter fig3, but normal output filter from fig 2/1

like this....thoughts?.

[deadastronaut]

ok, i tried 3 pt's...and its pretty good.

i compared my standard rebote at max delay with this and the delay is far cleaner at the matched same speed..

the longer delays are much more like it and clean too...very cool.

now i'm going for '4' pt's  ...just for the extra max time clean headroom..

i know the tails works...done.

i know the lfo works..done..(sounds lush) 

this is going to be a bugger to make a layout for though....eventually..

anyway into the fray....

[psychedelicfish]

How's the noise with the latest breadboarding?

[deadastronaut]

still have original circuit on bread at the mo..

but the noise is really good as it is, very quiet, ...so far,  i'll bread up the above circuit as soon as i get time..

[deadastronaut]

right, ok , this is where i am with 3 pt's in series..(original circuit) before i bread up 4' i was  going to try 4 today, but ....probably wednesday now..

https://dl.dropboxusercontent.com/u/7464107/delay%203pt%27s%20test%201.mp3

as you can hear its pretty gritty at the 'very'  end, but still long and clean-ish..very useable anyway, and repeat is a little higher than initial note too ....4 should do it properly i think...

[psychedelicfish]

That sounds pretty good. My idea of a good delay is one that can comfortably do delays long enough to play the solo from Heaven and Hell.

My own PT2399 experiments of late have been to do with chorus. So far I have something that is absolutely dead quiet in terms of hiss and hum (even on the breadboard), that sounds IMO way better than the LA or One Chip Chorus, and that even gets slightly flangerish  . Did I mention that it also only uses 1 chip?   Voltage Regulator not included. I'm currently working on the LFO, because I find that the One Chip Chorus' LFO clips the waveform and doesn't sound very natural.

I'll follow your delay journey and probably build whatever you come up with in the end, and I'll also post whatever I come up with for this chorus.

[deadastronaut]

hi ed, cool, i'd lke to hear that,  isn't the delay to slow for chorus?.. i found trying to get the time short could cause problems, anything under 1k on pin6 will cause the pt to 'lock' (not delay) at power up sometimes..

i put a 1k on all 3 pt's and now it start up every time.

as a note, (to myself really ) i have a delay pot (trim) on each pt' but i set pt 1+2 to fastest time, and use just pot 3 to control the overall delay time..

works well,and is the cleanest option,  so in the end pt 1+2 (+3 when i add 4)  will just have a 1k to ground. and 4 will have time control..

i'll try and add the 4th later today...been busy.

[psychedelicfish]

hi ed, cool, i'd lke to hear that,  isn't the delay to slow for chorus?.. i found trying to get the time short could cause problems, anything under 1k on pin6 will cause the pt to 'lock' (not delay) at power up sometimes..

You just need a RC circuit whose time constant is calculated to be longer than the start up period of the PT2399, which you use to change the voltage on the base/gate of the transistor. The collector-emitter/drain-source of the transistor then slowly decreases in resistance as the capacitor on the base/gate charges, meaning you don't get lock up. I'm using a MOSFET which turns on after the start up of the PT2399

[deadastronaut]

ahhh righto...

back in a bit..

[deadastronaut]

aha...4 pt's in series..just played over heaven and hell ...does it comfortably..

i'll try the above 'simplified ' schemo later from scratch.....its a 2 breadboard nightmare, but will be worth it in the end i'm sure....


edit: i put 1k resistors from pin 6 of pt's 1/2/3 to ground...i just control the time with pot on pt 4..

[deadastronaut]

4 pt's...

plenty enough 'useable' delay time for me anyway...at least the 'normal' delays will be much cleaner too as well.

https://dl.dropboxusercontent.com/u/7464107/delay%204xpt2399%20in%20series%20test%201.mp3

[psychedelicfish]

^There's a wee bit of noise in the last part of the sample, but that sort of delay time is getting out of the normal playing zone and more into the experimental zone.

How are you controlling the delay times of the chips? If It were doing this I'd use transistors to control the delay time, varying the base voltage of all the transistors to simultaneously vary the time of all the chips. If you did that you could also have a capacitor to ground somewhere near the bases to prevent lockup, because with four chips in series your minimum delay time is going to be four times that of a single chip so you'd want as short a minimum delay time as possible.

[deadastronaut]

^ yeah gets a little crispy at the very end, i just let it roll to show how long it could get before 'crispy' kicks in...very acceptable range , like you say

it gets into the experimental realms after even a second really...so i'm happy with that so far. the way i'd be using it would be quite adequate for my needs...

the rebote is great, but just wasn't that little bit long enough, and clean enough at max times..


with the pt's time control i just put a 1k to ground to fix each pt's 'time' and just control delay from pin 6 pt4 only...(no lock up problems at all)

do you mean ''current mirroring'' as sam mentioned earlier ..?  (i also read that elsewhere.)

i'm not too concerned about minimum delay times being too long 'slapback' e.g rock n roll stuff...or faux fridge like reverb, well not at the mo anyway, i can tweak that later.

i just want to get a minimalist but very acceptable 4 pt delay going..


anyway, i'll be ripping it all out today and going with the 'simplified' version above, and have a tinker with the filtering to see if i can get away with less components etc....less is more