Stuck for "preset alternative" solution on delay build

[MrStab]

hi,

at first it seemed like someone asked me to build them a supercomputer delay with particle-collider technology, but i've since managed to round down their insanely-complicated request to a delay (in this case the Small Time) with switchable settings. i've spent the past coupla days researching all the options, and when i weigh up the practicality of the possibilities, i'm at a loss. so i thought i'd see what you guys think.

current plan: 3 rows of switchable pots (3 pots: Time, Repeats, Mix) and a rotary switch.
problem: it's so easy to accidentally turn one of the pots. presumably this guy wants it for live use, and it just seems a bit pointless to go to all that effort for something that he'll have to alter before each set (locking pots don't seem that widely-available). trim pots are an option - do you think the average "end user" would be cool with that? it might seem a bit daunting. i could drill holes for em on the back or something. 2 sets of trimpots, 1 set of regular pots.

i understand that digital pots etc. are probably out of the question, even though i considered learning PIC programming a few hours ago. i'm not being paid for this beyond parts, so i'd rather not go all-out on learning an entirely new skill.

tap-tempo was an alternative to presets that he liked, but the Taptation board is kinda above my budget (and probably space) limits.

i thought maybe sequential switching would at least make up a little functionality with the "using 3 sets of pots" dilemma, but i can't seem to work out how to adapt either of these to my needs as they seem to expect a single input): http://hammer.ampage.org/files/quadsequentialswitch.pdf or http://members.shaw.ca/roma/switching-2.html.

what do you guys think? the limitations are practicality vs. budget vs. my novice abilities. any ideas would be really appreciated. on the plus side, i think i've managed to talk them out of having an up/down octaver in the same box...

[Mark Hammer]

The Small Time uses a pleasingly simple variable resistance to set delay time.  You will probably confuse the chip by switch between pots, since there will be a moment, no matter how brief where there simply won't be any resistance between pins 4 and 6 of the PT2399 chip.

Under those circumstances, it is preferable to use one pot to set the maximum delay time, and use a footswitch, or rotary, to place other pots in parallel with that first one, to produce shorter delay times.  That does introduce a complication, in that the needed setting of any additional pots will depend on the current setting of the main one.  But I am confident it is all quite learnable.

So, to illustrate, we'll make the "master" delay time pot the 47k that is shown.  The alternate delay time might be a 100k-200k pot.  One end of that 2nd pot would be tied to the junction of the 1k resistor and 47k pot, and the other free end would be connected, via footswitch, to pin 6 when you needed the alternate preset.

Alternatively, since the whole thing depends on total resistance, one could have the 1k resistor, an 18k-22k fixed resistor, and a third fixed resistor to make up the full 48k.  Your presets consist of a pair of pots placed in parallel with that 3rd fixed resistor.  A DPDT selects between them, and another DPDT either bridges the 18-22k fixed resistor, or leaves it in the path.  That would give you two alternate short delays and two alternate long ones.  And if you wanted to go really nuts, you make that 18-22k fixed resistor a 20k pot and use the footswitch to bridge THAT.

[MrStab]

hmm... i'm still processing what you've told me, but i think i get it. using smaller building blocks of resistance, with periodic control points, as it were. if, as you say, each pot would depend on the others before it, then i'm unsure if that's the right route for me - things like that are fine for stuff i use personally live, because i like needless complication, but i can picture it being an issue for the person i'm making it for.

wouldn't a "make-before-break" type rotary switch avoid the zero-resistance issue?

i'd be up to other delay projects if they involved less-obscure, hard-to-find IC's. i'm mainly drawn to the PT2399 these days because it's all that's easily and cheaply obtainable - complexity is becoming less of an issue. slightly.

cheers! i'm unsure where to go from here - i'll try to visualise your last suggestion a bit more.

[moosapotamus]

I'd guess that a "shorting" switch (make-before-break) might work.

So, if I understand, you want to be able to switch all three pots at once (delay, repeats, mix)? With a 4P12T rotary switch? Looking at the schematic, it seems like the repeat and the mix pots each need two poles of the switch. Then you're left with no way to switch the delay pot. Or, at least one of the pots is odd man out... unless I didn't understand the original objective?

~ Charlie

[MrStab]

i was thinking a 4-pole, 3-way switch (extra pole for LEDs). i could be wrong and this probably won't work, but couldn't i flip solely between the wire either going to or coming back from each pot, as opposed to both to & fro? so there'd be a point where all the pot "options" converge on the board, but signal would only be going to/from one of em. unless there's some AC issue where stuff will flow backwards & wreak havoc. i'm probably thinking in simplistic, lazy-ass Tube Screamer terms or something - i'm just going with what feels like the logical approach to me, i'm unsure if there might be issues with this particular circuit.

[MrStab]

the switchable-pot thing is still on the cards, but i feel under (self-imposed) pressure to deliver, so there's a chance i may have to bite the bullet and go with the Taptation. i'm currently looking at a stripboard/vero layout for that - i figure i'd need to ground pin 11 (double-time input) to simulate a 50k pot, as per the app notes, but i'd probably wanna omit all the other features as well. which makes me wonder if i can justify the cost. all i know is you build a coupla FX for friends, and the next minute everyone wants you to make them a spaceship. lol
i need to go to bed. between this and another frustrating PT2399-related project (for myself), i've suddenly become disillusioned.

how would you guys respond if someone asked you to make something along these lines? basically with the goal of fast, easy adjustment between songs i guess.

[samhay]

The Small Time uses a pleasingly simple variable resistance to set delay time.  You will probably confuse the chip by switch between pots, since there will be a moment, no matter how brief where there simply won't be any resistance between pins 4 and 6 of the PT2399 chip.

I haven't tried this (as far as I know), but I suspect that having an open circuit here will not be too much of a problem. Current consumption of the pT2399 goes up with decreasing resistance from pin 6 to ground, so you MIGHT get away with using a break-before-make approach. Alternatively, as alternative to a master pot, you could try a large (100s k) fixed resistor instead, which will make your delay pots more intuitive. What sort of max delay is this fellow looking for?

If this works, then it should be fairly simple to use a 3P4T rotary switch to select between 3 sets (3 x 2 pots) of delay and repeats options.

Edit - this doesn't give you 3 mix pots too, but this is probably less of an issue.
I wonder if you really need seperate repeats and mix pots though, as these are pretty easy to set, and there is a fair bit of slop in finding their sweet spot(s).

[merlinb]

Could you be more specific about what presets he wants?
Does he want, like, three preset option for all three controls, making nine in all?
Or does he just want the delay time to be presettable to, say, three options, and no presets for the other controls?

[MrStab]

the trouble is, i don't think he's really thought it through and expect some sort of digital multi-FX-like solution. i've even mentioned that some things are just impractical without such a unit. all i could get out of him is that he wants either presets or tap tempo to be able to vary delay in a live set. so i get the goal, i just need to figure out a practical solution.

i assume he wants all 3 controls to be preset-able - that's how i'd go about it at least.

i did mention that this circuit has a 400ms limit, Sam, but it seemed lost on him - which leads me to think that'll be fine. he should've read the box. lol
from your site, Merlin: "With a 47k DELAY pot the delay time will go up to about 400ms but, even with all the filtering, digital noise starts to become audible around 250ms.".  i understand that the Echo Base can reach ~700ms, and the Rebote ~600, but the former would just complicate things even more and the latter - afaik- has no tails. plus i'm just more experienced with the Small Time.

i'm unsure what you mean about the 100k fixed resistor, Sam? i'm regaining confidence in the rotary switch method. i'm getting the impression that i'll absolutely have to put both ends of the pots through the switch as opposed to just one end of each. i could probably live without a mix pot. that'd leave no lug for status LEDs though, which could confuse things, but maybe not if it's a chicken-head pot.

locking pots seem to be pricy so i'm trying to consider whether certain types of trimpot would be durable enough. i think regular pots are out of the question, with their ease-of-accidental-turning. unless maybe im being paranoid about that?

it's a shame i don't have more time & resources, as i'd be up for learning PIC-programming (i've read the handful of threads on this), at least for experimentation's sake. though not a programmer, i've been a hardcore computer nerd for at least 20 years & i have dabbled.

cheers a lot, guys

[Paul Marossy]

From my vantage point:

Wouldn't be easier for him to just buy a Boss DD-20 or similar pedal? I hope he's paying you a decent amount of money or you'll lose your shirt in time spent on it.

[merlinb]

i did mention that this circuit has a 400ms limit, Sam, but it seemed lost on him - which leads me to think that'll be fine. he should've read the box. lol
from your site, Merlin: "With a 47k DELAY pot the delay time will go up to about 400ms but, even with all the filtering, digital noise starts to become audible around 250ms.".  i understand that the Echo Base can reach ~700ms, and the Rebote ~600, but the former would just complicate things even more and the latter - afaik- has no tails. plus i'm just more experienced with the Small Time.

All those circuits use exactly the same topology (basically straight out of the data sheet). If you want more delay, just keep increasing the delay pot! If this is for bass guitar then you could increase C7 to keep the noise down further at longer delays, since he won't need as much bandwidth as regular guitar. Just a thought.

i'm unsure what you mean about the 100k fixed resistor, Sam?

I think he means attach a 100k resistor to the delay pin, permanently. Then have three (or whatever) 100k pots in parallel with it, each one selectable with a switch. That way the delay resistance never exceeds 100k, even between switching. In practice the PT does not seem to mind having its delay pin left open. Still, some sort of definite resistance is good practice, even if it's 1M or something.

[midwayfair]

I see an extra issue: a preset in a PT2399 that involves the delay time is going to result in pitch sliding while the tempo stabilizes. And you aren't going to fix that no matter how you implement the "preset," even using the Taptation.

I'd seriously consider telling this guy that the sort of delay that can be built with presets isn't compatible with the resources at your disposal. Presumably he's not paying you what a Strymon Timeline costs, so he should expect that kind of functionality.

[Paul Marossy]

And FWIW, I've never had one working circuit from that "Laurier's Handy Dandy Circuits" site.

[MrStab]

Wouldn't be easier for him to just buy a Boss DD-20 or similar pedal?

i'm leaning very heavily towards telling him that, and have indirectly implied it - indirectly so i could see of it was at least feasible first. as this circuit doesn't come from a big evil corporation, i'd be uneasy about making profit.

this one's for guitar, Merlin. ahh, so the 100k is a "safety resistor" kinda. i'm having trouble visualising how i'd wire that parallel-y as opposed to in series, right now i'm thinking the outermost lug of delay pot 1 to the outermost lug of delay pot 3? excuse any blatant stupidity.

are higher delay times usable or just borderline at best?

Jon: do you mean like the kinda sound you get if you hit a note and turn the delay pot manually? as it's a rotary switch, i wouldn't expect him to be able to switch mid-song - it'd have to be by hand, so maybe i could get away with that.

this would be so much easier if i hadn't woken up to an empty coffee jar. lol

[moosapotamus]

I see an extra issue: a preset in a PT2399 that involves the delay time is going to result in pitch sliding while the tempo stabilizes. And you aren't going to fix that no matter how you implement the "preset," even using the Taptation.

I'd like to hear what that would sound like. Maybe it could be a "feature".

But, as long as the user didn't need to switch presets while playing in the middle of a song, prolly wouldn't ever hear it.

Someone suggested forgetting about including the mix pot in the presets. I think that's a good idea. Then you could get away with a standard 4-pole, 3-position rotary switch to have three different combinations of delay/repeats plus status LEDs. One pole for the delay pot ('cause it's just a variable resistor), two poles for the repeats pot (connect all the grounds together and switch the other two lugs), and the fourth pole for status LEDs.

But, like Jon pointed out, until you try it, I'm not sure what switching the delay time like that will sound like. My guess it that it will make a little "boing", like if you turn the delay knob quickly from one position to another. Could be cool. Could be weird. Could be a non-issue, maybe? But you might have to actually try it to find out.

~ Charlie

[MrStab]

One pole for the delay pot ('cause it's just a variable resistor), two poles for the repeats pot (connect all the grounds together and switch the other two lugs), and the fourth pole for status LEDs.

thanks, Charlie - that clears a lot up.

i have band practice tonight but i'll get to work afterwards (i do have my own Small Time, but it's in my graveyard of abandoned projects with stuff cannibalised from it, as i hadn't really needed one myself). i could just use a spare DPDT footswitch to test, maybe - not ideal, but beats ordering parts that might go to waste.

edit: I've just noticed a text message where the guy says he's unsure if he'll have the cash for another week or so. that's fine by me, less to worry about for now. lol

[midwayfair]

Jon: do you mean like the kinda sound you get if you hit a note and turn the delay pot manually? as it's a rotary switch, i wouldn't expect him to be able to switch mid-song - it'd have to be by hand, so maybe i could get away with that.

Yes, that sort of sound.

Wait, he wants a rotary switch for a preset? WTF is the point of a presets on a delay with only three knobs? How is that better than just twiddling the knobs to the new setting? You'll have to add that many more controls and it's just more things to knock out of their position while setting up. Why not just use two delays?!

Given that the guy is "unsure he'll have cash" make sure you get a deposit.

[MrStab]

this'll be one instance where i think i should ask for a deposit - i don't know them personally, it's a friend-of-friends type thing. i have a feeling there's an element of "wow, someone makes stuff! let's milk it" with no regard for either my limitations or sheer practicality. but then, on the other hand, maybe he's on a budget and wants something to fulfil his live needs. either way, after this, i'm going to be less open to custom requests.

[armdnrdy]

And FWIW, I've never had one working circuit from that "Laurier's Handy Dandy Circuits" site.

+1

I built a frequency counter with switchable frequency ranges. I found mistakes in his layout and schematic. I contacted him to verify my findings...he was pleasant enough...but it kills me that people post things on sites that will not work in the presented form!

[midwayfair]

this'll be one instance where i think i should ask for a deposit - i don't know them personally, it's a friend-of-friends type thing. i have a feeling there's an element of "wow, someone makes stuff! let's milk it" with no regard for either my limitations or sheer practicality.

hah, I typed something to that effect but decided not to make assumptions.