Experimental Delay & AmpMod from Noise Workshop

[moid]

Thanks very much for your great eyesight and effort! Wow, I wasn't sure if I'd get a reply so you've made my day

moid - hello and welcome to the forum. and this is certainly the best place for a question on this circuit at the moment.

your layout. you don't need to run all the earth wires to the same point. on the pots, for example, you can link across the 2 grounding pins, and then run a wire back to the veroboard, anywhere along the bottom row will do. a sleeve to sleeve to dc jack wire is good.

OK I've made all the grounds from the pots go to points on the ground row of the board. That makes sense, I can't think why I didn't think of it before!

I'll look closer to your board overnight, but: are we to assume there is a cut under the right most 1uF cap? and, I always preach '"double links" are are a bad idea, and to be avoided', especially seeing you have so much room on this board. me, myself, I'd put a shorting to the left of pins 3 and 4, then run a longer link to bottom row under the IC.

Yes, I forgot that one - the 1uF (whoops!). I've got rid of the double links and I think I've done what you suggested. Is there a simple explanation about why this is a bad idea (to use double links?). I've used them because I've seen them on other vero layouts and didn't realise they were problematic in some way.

nice use of colour, except you've used the same green to 4 different spots on the same side of the dia.

Fixed that! Sorry, I'm colour blind which means I really shouldn't play with electricity! I used red for power and black for ground which seems to be the convention, are there any other conventions for the colour use on cables? Green seems to be popular in other vero layouts, but I'm not sure if that signifies anything or it's just because it is the default colour in DIYLC.

and: your regulator. see if you can shift it right across to the IC pin1, and up 3 rows. (or, invert and shift across same level, about 3 or 4 rows from the IC.)

Thanks, that saved quite a lot of board and looks a lot more elegant than my maze layout.

if you shift the 220nF and 1k left 2, you'll save some more, but this is only a quick look, and you'll make some changes, so I'll look again when you show rev2.

I think I ended up moving other parts instead but I made the board a bit shorter which is great.

an update: the 1k from pin12 shows wrong. if you turn it 90 degrees, and cut under, it will be korrekt. there also needs to be a cut between the left hand 1k and pin8, if it is there, it's not visible.

I see no other errors.

I think I've done this correctly (the 1K) please could you see if I put it the right way around? I've added the cut to pin 8, thanks! I really appreciate your efforts. I probably won't be able to start this one until next weekend (too much work on at present) but I'll add some updates as soon as I get something built.

Hopefully the revised version is below: (EDIT: noticed problems with the two left most capacitors and fixed them)

[duck_arse]

okey dokey, it looks right now. but we might need some changes. allow me to mutter to myself here for a few minutes .....

double links cause confusion when people don't include the note, or people don't see the note. one wire for one hole, for mine.
stretching caps is another bad for me, especially with all those other holes there. if that right hand 10nF is a box or 3 spot cap, put it bottom row and 3 up. move the 1M right 1, the 22k right 1 and change to a 4 spot, and bring the 1k and 100nF down with as they are. (or turn the 1k back as in V1, and shift the other parts around it to correct.)
the 100uF, 100nF, 47uF should be nearer where they are needed. shift them right across close to the IC pins, and across 3 spots if they are that big, 2 spots across the reg makes them even closer. there is plenty of ground areas mid IC you can use, get creative, you can probably shrink that end by half.

as for your colours, it seems there are quite a few hereabouts with colour vision problems, something I didn't realise when splashing the pastels on my layouts. I'd be interested if you could tell/show me the best combinations of colours to use to prevent visual entanglements and cofusions.

I hope some of this is helpful.

[moid]

Yes thanks, it's all realy helpful! I'm not sure if I followed all of what you wrote exactly, but I've tried to understand the spirit of it and it helped me realise that I could shave five columns off the width of the board so that seems good to me Now I see the issue with double links, I'll try to avoid those in future. I hadn't thought of the stretching caps issue - you are right, I've got a few box caps that come with very short legs and I'd never be able to get them to stretch across such gaps. I think I've fixed those issues now.

Regarding colour usage for the colour blind, sadly there is no easy solution (beyond just using black, white and grey; the only colours that people with colour blindness can definitely see). There are three main types of colour blindness (broadly speaking) and each type has different colours that cause issues... so I don't think you can design (easily for the colour blind). I tend to use saturated bright colours because they are easier for me to distinguish from each other, and black, white and red are fine for me, but brown can be a disaster, as can purple, lilac, turquoise, many of the pastels (they often look the same)... but my colour deficiencies will be different to another colour blind person... my grandfather could only see in black and white and red - he had no understanding of blue or green at all. I think the best option is just to help someone if they ask because that's often the most useful thing. Luckily circuit diagrams are in fairly clear solid colours against plain backgrounds - the real world is much harder to understand becasue of the subtle hues and tones it contains. I am aware that I am never going to be an electrician, and I only work on guitar pedals because if using 9 volts even if I get totally confused I can't hurt anyone! (I haven't even managed to blow up any components yet). My wife always winces when I offer to look at the wiring in our home, luckily her brother is an electrician so he fixes that sort of work for me - I've tried watching him to understand what he does but the colour confusion is rather frightening and I don't really want to repeat my one experience of 240 volts when I did something wrong with a wall power socket and got zapped across the room (that hurt enough for me to give up on tinkering with mains electricity ever!)

I've put my latest version of the circuit below - I just wondered if I put a larger pot on the delay time would that give me longer repeats, or just very badly distorted noisy ones? I say this becaus I've got a 500Klin pot but no 250 at the moment and wondered if there would be a drastic change (it's no problem to buy a 250, I was just curious). I might add a clean blend knob to the circuit because I often like the delay to float behind my main playing and I'm not sure if this circuit can do that anywhere?

Thanks for your time and help

[duck_arse]

I recently decided to go with the brown/blue/green+yellow of the mains wiring scheme, although diylc doesn't have a striped colour. hmmm, I might just keep this new scheme anyway.

V3 is looking nearly there. nice work with the "folding" in the top right corner. it takes some people ages to work that out. and: if you move your left hand 220nF rightwards 2, 4 or 5 rows, you can then rotate, cut-under and lay flat the 10k, pointing straight off the edge to the pot.

but! you've lost 2 caps from the 5V line.

a larger pot will create longer delays, but you might not like what you hear. have you breadborded this, or heard knutolai's sound clip? I'm not too sure it will do sublte, mixed into background.

[moid]

Thanks for the compliments. I moved the 220nF as you said and then realised what you meant by the missing two capacitors! I had to make the board a bit wider after that, thanks for spotting those, I really thought I'd been careful. Is this likely to be good to build now? I haven;t breadboarded it (I'm not very good at that yet, I find things tend to work for me on veroboard).

I might try a larger pot when I build this just to see if I like the effect, but if it turns into total noise I'll get a 250 pot instead - it's easy to swap them. I like pedals that create more extreme sounds but if possible can be more normal if required... I've added a clean blend pot so I can mix the clean incoming guitar with the output of the board to decide if I want to a subtle effect or full on damaged repeats. I really liked Knutolai's sound clip, that's what made me want to try to build this (that and the fact that I just finished my first delay - the Clarinot and I have about 10 PT2399 chips hanging around after accidentally buying more than I needed )


one question (if you don't mind) why is there a green LED on the board? Does this show repeats and pulse? Or does it do something else?

[knutolai]

one question (if you don't mind) why is there a green LED on the board? Does this show repeats and pulse? Or does it do something else?

It's used to add more graceful clipping to the sound when setting the feedback pot to runaway feedback. It could be removed for a different runaway feedback character.

[duck_arse]

.... I just finished my first delay - the Clarinot and I have about 10 PT2399 chips hanging around after accidentally buying more than I needed )

it won't be long (dead astro's "chasm abductor") before you realise the error of this statement.

the board look ok, seems all the parts are there, somewhere. remaining component joggles can wait for a more advanced tutorial. also, a breadboard will help with your layouting, and save time and money and tears when it comes to what/whether to build.

[moid]

one question (if you don't mind) why is there a green LED on the board? Does this show repeats and pulse? Or does it do something else?

It's used to add more graceful clipping to the sound when setting the feedback pot to runaway feedback. It could be removed for a different runaway feedback character.

Thanks Knutolai, I'll keep it inside the enclosure then, I thought I'd ask in case it was for visual feedback of the sound.

[moid]

.... I just finished my first delay - the Clarinot and I have about 10 PT2399 chips hanging around after accidentally buying more than I needed )

it won't be long (dead astro's "chasm abductor") before you realise the error of this statement.

the board look ok, seems all the parts are there, somewhere. remaining component joggles can wait for a more advanced tutorial. also, a breadboard will help with your layouting, and save time and money and tears when it comes to what/whether to build.

Thanks, now all I need is some time to build this (not this weekend alas). I've got a breadboard and I do intend to learn more about using it, but I know I want to build this pedal

Just googled the Dead Astronaut Abductor - that looks pretty awesome, thanks for letting me know about it - a project for the Summer holidays though!

[knutolai]

Consider to socket the PT2399 as its quite a sensitive chip! The rest should be pretty simple to solder.

[moid]

Thanks, yes I've learned (the hard way) that I should always socket ICs and transistors! Now all I need is some free time (not this weekend alas).

[moid]

Thread resurrection! OK six months later but I finally had the time to build the Ekko delay... first version only passed clean sound (both with the 3PDT on and off), so I thought the PT2399 chip might be dodgy (I bought 10 of these for a ridiculously cheap price on EBay, so maybe that wasn't the best plan). I replaced the PT2399 with another chip from the same batch and now I can create constant oscillation noise, similar to the extreme settings in Knutolai's demo, but nothing else - turning the pots gives either nothing for 90 percent of the sweep or intense oscillation... could that new PT2399 also be dodgy? I socketed it so I could swap it out. It does sound like runaway feedback of the type you get when you set an analogue delay to have too much feedback/repeats. The Green LED also only turns on when the Delay is at maximum. The vero board isn't yet in an enclosure, its floating suspended on my table in parts, so perhaps if I boxed it that might help?

Anyway what I'm trying to say is help! Any suggestions? I followed the vero layout I posted and used all the components the same with no substitutions. Thanks

[moid]

OK close inspection of my PT2399 chips shows that I have two different 'models' of them. I have two from a reputable online store (Bitsbox) that have a different appearance to the the cheap ones I got from EBay... they have different text / font and the size shape of the indent marker to show the top of the chip is different... so I tried the Bitsbox PT2399 and I now have much less oscillation but nothing else. Also the volume pot seems to be loudest in the middle of its sweep and go quiet towards either extreme... which is bizarre. Thanks for reading.

[knutolai]

Hm sounds like your layout is faulty somehow. I would trust the bitsbox chips. Remember the 'Mod' pot has a reversed range (most intense at min.)

[moid]

Thanks for replying so quickly Knutolai - guess what, you are right and I'm stupid I've just realised I missed the right section of your schematic where there is 9v going to a 2n7000 transistor and a red LED. How does that part of the circuit connect to anything else? I can see there is a point labelled P4 which I assume connects to something else, but I can't see another P4 anywhere else to attach it to... would you mind telling me where that part should go please, I'd really appreciate it! (hopefully I can make it as a small extra vero board and connect it where required). Thanks

[knutolai]

The circuit with the Mosfet and LED does not have anything g to do with the audio circuit. Look up the LED indicator circuit for Valvewizards "engineer's thumb". The green LED connected to the PT2399 has no visual purpose. It's there to shape the repeats when the delay is cranked.

[duck_arse]

moid - can you post your pt2399 voltages please? and photos of the board, front and back? and the external parts so we can see the wiring?

[moid]

The circuit with the Mosfet and LED does not have anything g to do with the audio circuit. Look up the LED indicator circuit for Valvewizards "engineer's thumb". The green LED connected to the PT2399 has no visual purpose. It's there to shape the repeats when the delay is cranked.

Thanks - I looked at that link,  I assume that Mosfet section is something to do with the form of bypass switching for that pedal, but because I used true bypass on my circuit I didn't need it? I understand that the LED will affect the sound, but I wondered does it need to be lit to affect the sound, or is it still shaping the sound even when unlit? Perhaps the voltage going through it can be very low and still affect the tonal qualities of the music but not be powerful enough to make the LED light up?

moid - can you post your pt2399 voltages please? and photos of the board, front and back? and the external parts so we can see the wiring?

Here are some images (don't laugh, I'm no electrical engineer! ) the pedal isn't boxed yet so is a mass of cables and pots...


Close up


Back


For the voltages this is what I have (never done this before so I hope I did it right - set the DMM to DC, then stuck the ground probe into a spare aluminium enclosure, and held the positive probe against each pin using the pinout numbering of the chip (starting top left and going anti clockwise around the pedal - like this - )

I just had power going to the board, I didn't plug any guitar into it and the board sat on a sheet of white paper as seen in the photos.

1.    330mV but constantly increased very slowly to this value - I measured for over 2 minutes and it was still rising at a rate of 0.1 mV a second!
2.    160mV
3.    8mV
4.    fluctuates rapidly between -0.02mV and 8mV
5.    177mV
6.    148mV
7.    130mV
8.    9mV

9.    203mV
10.  210mV
11.  156mV
12.  147mV
13.  113mV
14.  108mV
15.  130mV but fluctuating wildly by up to 50mV in either direction
16.  174mV

I would've expected to see some numbers in whole volts (maybe I'm wrong in that?). Any opinions would be lovely, thanks for your time, both of you.

[duck_arse]

good news and bad. the good news first - your soldering looks pretty good. bad news - you need to pay some attention to your board. there is at least one bridged cut in the top row - this is a bad thing. go over all your cuts, and clean out the little crescents of copper, they can make shorts you can't see. also. run a sharp blade between the copper tracks, so you can pick off any other bridges there might be. (5 down, left hand side .....)

topside - if using 1/8W resistors, sit them flush on the board. bend the legs the right size, don't loop them, cause I can see a few loopy legs might be touching. same goes for caps, bend the legs so they will sit flat/flush to the board. and the links, you want them right-sized and STRAIGHT! that blue resistor top-right, should be hairpinned, stood on one end, neatly.

and, your voltages, looks like you've got it in one. the pin 1 is fed from a regulator, takes IN 9V, puts OUT 5V. but you need to check the datasheet for the part you have IN YOUR HAND, because there are some oddities around.

and see pin 3 and 4? they are tied together, very hard, and tied to ground, so there should be no voltage difference, no dither, and no voltage there. put your red lead against the black lead, and that reading is what should appear at all the grounded points.

so, rework that board, check the datasheeet, fire it up and get back to us.

[moid]

Thanks so much for the advice, I removed the bridges (I found a couple of other possible suspects as well, I guess that's what happens when you only get to make pedals late at night and your eyesight is somewhat less than eagle-like), straightened out the components and ladies and gentlemen (who am I kidding?) gentlemen and gentlemen, we have a winner! Crackly, grumbly, rumbling particles of noise echoing (ekkoing?) around like deranged moths at a light bulb! I am a very happy man Hopefully I'll get sometime tomorrow to drill a box for this.

I have a couple of other questions but you don't have to answer them if you're busy, you and Knutolai have been more helpful than I could ask already.

1. Is it possible to get an output from this circuit that would just contain only the delayed, repeated signal? (none of the original guitar at all) My Timebender pedal has a Dry Off setting which is great for generating strange noises that I record and then mix underneath cleaner guitar sounds later on and this pedal could be great for this if I could remove the original guitar. Feel free to say no this can't be done with this circuit; I have no idea how this could be achieved myself, I assume the Timebender does it digitally somehow.

2. The Blend knob I put in (to mix more of the original dry signal with the output of the ekko delay) works brilliantly at one end of the sweep (full ekko circuit sound), but at the other end of the sweep, all the delays are missing so I get clean guitar but over this is the constant crackle and hiss effect and I was expecting not to hear that. I rather like it, so I'm keeping it as an instant weird add lo-fi noise to the pedal chain thing but I'm wondering if I used the wrong pot type or did something else to prevent a clean blend from occurring?

3. My board components stand off a bit because I use crocodile clips attached to the metal legs next to each component to absorb some of the heat from my soldering iron so I don't cook the resistors / capacitors etc - I came up with this technique due to my first two pedal builds going appallingly wrong through me frying the parts with the heat from the iron, and since I've used crocodile clips as heat absorbers I haven't melted anything by accident. Presumably I should just try to work without them so the components can lay flush with the board where possible?