Can you play guitar through a smoke alarm?

[anotherjim]

I wonder if Wickes has an audio engineer in the business since they are the only DIY chain I know that stocks high-density mineral wool needed for bass traps, soundproofing and treatment as opposed to all the light fluffy stuff for loft insulation - which is useless for audio. Of course, that's the very opposite of reverberation, but if your device is too sensitive to ambient sound/vibration, you might want some to line a casing.

Anyhow, reverb springs are usually lightly tensioned and open coiled? Have you seen this concurrent thread?
https://www.diystompboxes.com/smfforum/index.php?topic=120207.0

[moid]

> the velocity of sound in aluminium is far faster than it is in steel...

Not "far".

stuff               Metres/second
Aluminium    6300
Beryllium        12900
Copper          4700
Glass(crown)    5300
Iron                  5900
Nylon, 6.6         2600
Oil (SAE 30)   1700
Plexiglas         1700
Steel, Mild    5920
Steel, Stainless    5800
http://www.classltd.com/sound_velocity_table.html

6300 vs 5900.... the available or convenient sizes will make more difference than that.

BTW: for a long-time small-size delay you want *slow* speed of sound.

> now I'm wondering what metal those biscuit / sweet tins are made from?

The only likely stuffs are plastics, Aluminum, and steel. ("Tin can" is 99.9% steel.) A magnet will sort all common steels from anything else. (Stainless Steel and Nickel are very weakly magnetic but too expensive for candy.)

Thanks PRR. I've checked any my tins are definitely steel (well they are magnetic at any rate).

I think I'm giving up on trying to create plate reverb with a tin now! You cannae break the laws of physics as a wise man once said A chat with my wife about buying large sheets of steel has not gone exactly to plan (she didn't object to buying one per se, but point out that our house doesn't have anywhere that one could store a 2mx1m steel sheet and frame... and my suggestion of hanging it flush with the ceiling in the hallway was not apparently her idea of interior decoration... I mean shouldn't every home have a plate reverb?). I'm going to move onto spring reverbs now! (I'm a glutton for punishment)

[moid]

I wonder if Wickes has an audio engineer in the business since they are the only DIY chain I know that stocks high-density mineral wool needed for bass traps, soundproofing and treatment as opposed to all the light fluffy stuff for loft insulation - which is useless for audio. Of course, that's the very opposite of reverberation, but if your device is too sensitive to ambient sound/vibration, you might want some to line a casing.

Anyhow, reverb springs are usually lightly tensioned and open coiled? Have you seen this concurrent thread?
https://www.diystompboxes.com/smfforum/index.php?topic=120207.0

Thanks anotherjim. That link did indeed inspire me and over the weekend I bought some piano wire because according to a piano string website I read you can make (you'll never guess this one) piano strings from it! (no shit, Sherlock!) aaaaaand.... (drum roll)... musical springs Yay! I immediately became as excited as a puppy with a new chew toy and bought 30 meters of 0.6mm piano wire for the very agreeable price of £3.40 from B&Q (Wickes sadly do not stock this item).

... and I spent the whole afternoon trying to turn a 12 meter length of it into a spring for a spring reverb... here's a tip: If your life is ever going too well and you feel like you'd like to be part of a stressful, pointless and slightly painful failure of a task, then I recommend trying to make springs out of piano wire. It will dissipate any good feelings you may have about the cosmos and enable you to have to explain some exciting new descriptive words to your son at the same time (feel free to substitute son for any other close family member or neighbours who are within shouting distance). So I managed to expand his vocabulary a bit, but now have about 2meters of piano wire coiled into a 1 meter length of cable that refuses to be any smaller. If it was flammable I would've taken it outside and burned it as a sacrifice to our Dark Lord of Oscillation.

So I've now ordered some copper wire and hope that that will be more conducive to spring making. And if that fails I'm going to have a good old fashioned cry in the corner. I'll report back when the copper arrives and I've made a spring.

[moid]

OK just for laughs I actually hooked up the weird piano spring to some piezo discs and played guitar through it... a piezo at the full 2m length of spring makes laser sounds, massively distorted fuzz, a slight hint of actual notes, and ear destroying volume if I accidentally graze the low E string... in fact the volume is enormously increased as I play from high to low, so the circuit desperately needs some sort of high pass filter to kill the bass volume. Should this come before the speaker, or after the piezo disc?

I also tried the piezo disc abut 50cm along the spring and got a rather lovely soft tone that is sort of reverby and I rather like, but it does accentuate the actual percussive touch of the fingers on the string far too much - perhaps a high pass filter would remove that? I rather like this sound, I just need to get rid of the percussive attack (Yes I tried playing with fingers -that's better, but it still sounds like someone hitting an acoustic guitar with the palm of their hand every time the note rings out.

I also tried putting the spring inside the metal tin I have which resulted in furious high pitch fuzzing like bees on metallophones with lasers going nuts. So the spring needs to be suspended away from surfaces somehow.

[PRR]

> the spring needs to be suspended away from surfaces somehow.

"Necklace reverb":
http://www.nshos.com/HAMMOND17.htm

[moid]

That's cool PRR I think I'm going to use some right angle metal brackets and suspend them from the lid of the box with nuts an bolts and then attach the cables of the piezos to them, and attach the speaker to another right angle bracket at the other end of the lid. It works in my head

Speaking of things that worked in my head up until a few minutes ago - my circuit ends with a ZVex SHO booster to make it louder. Then the output of the SHO is mixed into the dry output of the amplifier circuit I used. The SHO has one transistor and the amplifier has none, so that means the phase from the circuit will be inverted, and then mixed back with the dry audio... I suspect I may have an issue with phase cancellation. If so, I might leave it as an option (because sometimes that sounds cool), but if it sucks, can anyone recommend a simple phase inversion circuit so I can mix the two signals without losing too much volume? Thanks.

In case anyone's interested, the circuit currently has three springs - a short copper spring for short reverbs, a long copper spring for (hmm, guess what?) and the short steel spring / disaster object for lo fi metallic weirdness. All three go to separate piezo discs with separate mixer pots so I can blend between all of them for a variety of reverb tone and length. I'm going to try adding a feedback pot to mix some of the circuit's output back in before it hits the high pass section to see if that makes for a more interesting sound (If it just oscillates I'll probably remove it). I have ten pots and a switch so far... what could possibly go wrong?

[PRR]

A delay-line is *multiple* "phase shifts", varying with frequency.

There's no "right" phase.

[anotherjim]

Phase matters if there's direct non-delayed or very short delayed sound reaching the pickup discs. With this "design" there probably is some sound picked up directly from the drive speaker. By mixing the dry sound with an inverted wet, it ought to clean that direct "clutter" pickup out some. If you try to add feedback, the presence of direct in the delay line will cause positive feedback and very likely it'll oscillate - so you probably have to keep that inverted amplification relationship at all times.

The simplest way to add phase reversal if you want - is put a DPDT switch in the drive speaker voice coil connection? It can be done electronically, but with the amplifiers you have, it probably means adding an extra stage. The easiest way I can see is switching between + and - inputs of the 386, but switching the speaker coil means the least work?

[moid]

Thanks chaps I think the debate regarding phase is a bit of a moot point at the moment because I finally finished assembling this monster and it mostly doesn't work I've discovered that audio isn't going through one of the high pass filters (it is blocking all audio for some reason at the 12 nF cap... so perhaps that is dead, I'll try making that one again - if you combine two high pass filters do they do that sort of thing?). I do have both high pass filters controlled by a dual gang pot, and I've never used one of those before so maybe that is causing an issue?. I then bypassed the high pass filters and ran the amp circuit into the speaker - that works - the springs vibrate (as long as I hold the lid in the right places) and sound goes to the pots that blend in the different springs, but after that they don't seem to send the reverb sound onwards? I might be wrong it's late and I'm confused...and have to go back to work tomorrow eurghhh.  I definitely have some sort of grounding issue because if I touch the metal lid of the box in various places the sound will cut in and out randomly... near disaster as per usual Before you ask, no, I have no idea what I'm doing, I don't even understand what ground is in electronics, I'm like a monkey with a selection of electrical components ramming them into different holes and hoping that something good will happen - if I was allowed to spend an infinite time I would of course eventually create clones of all the greatest effects pedals in the world... but it would take an infinitely long time...

I probably need to draw a layout of what I've made because a photo of this Frankenstein creation will probably just make you laugh.. or cry... here you are (for entertainment purposes only, don't show to children please)

- Yes ladies and gentleman, a face only a mother could love... I'll put knobs on when the pots work!

Now, the real horror, avert your eyes!

[duck_arse]

I like the "Moffies" switch. those alligator clips - I'd have thought you were shorting the backing disk to the crystal face with those, and would result in no noises. [I'll stand for correction, mind.] ahh, unless you have something insulating the backing disk from the croc clip .....

[moid]

Thanks Duck, I was really hoping to find a switch made in Birmingham (they say uff there when they mean off) but sadly I didn't find any uff... much like real life sadly. Anyway, you may have been onto something with your idea about not shorting the piezo discs out - I thought, aha I need a strong gripping tool that is insulated, and like most of us my first thought was of course rubber coated nipple clamps! They have excellent gripping strength and the rubber coating prevents any short circuiting... sadly I was thwarted in this plan due to me needing three clamps (three piezos, see?) and I'm not Scaramanga, so that was out. However resourceful chap that I am, I decided all I had to do was cover the jaws of the crocodile clips with insulation tape and hey presto that would do the trick. So I've done that, but I still got problems... Sound is going through the effect now (with lo fi reverb) but it kept cutting out. Then I looked a bit more closely at the amplifier pots and realised the solder tags are touching the metal lid if it bends a bit (which it does whenever you plug anything into it, or turn the 3PDT switch on)... so I removed those and added more tape between them and the lid That got rid of a few more noises and it hasn't cut out  (much) since.

My high pass filter is still busted, so I need to work on that. The feedback pot doesn't work either (I think it always allows some dry signal through, and I'm not sure if it does what I thought it would do either. I'll make a drawing and then you can laugh at it and hopefully tell me what I need to fix that).

None of the pots that the piezos feed into seem to work properly - even when at what should be a sweep that turns them off they are still allowing a fair bit of music through. I was using 470K pots (why, because I got a large bag of them cheap when Maplins went bust) and I thought, audio signal into lug 3, audio out on lug 2, nothing on lug 1 - if I twist the pot all the way one way I should get 100% audio and all the way in the other direction and I should get silence... but I don't, I always get some audio. Should Lug 1 be connected to ground? Should I be using a 1M pot for this sort of work? I never have understood what resistance stops audio dead... does the volume effect how much resistance is require to stop the audio signal?

So progress... but it will be in short bursts from here on, the day is now packed with miserable emails and problems at work

[moid]

Hello everyone, it's been a dreadful week at work, but I finally managed to draw the contents of the box as a vero layout. It ain't pretty... but no matter how you look you won't find Wally in there, so it's got that going for it which is nice. Individual bits of the circuit work; the issue is fine tuning it so all of it works together



The sound comes in on top left, there is a dry out on top right that you can ignore, I just wanted to be able to get a dry extra line out of the box for running into a separate chain of pedals. The reverb/dry mix sound comes out of the box at the bottom of the image.

Problems - the second high pass filter vero board will not let any sound go through it. The first one is fine. Maybe I fried the capacitor or maybe you can't have a dual gang pot controlling two high pass filters? I don't know. I did wire the dual pot so that a separate cable goes from each high pass vero to each lug 3, but for the ground connection on that pot I did run one cable from lug 2 to the other lug 2 and then to the ground on hte power socket - maybe there should be two separate ground cables?

The three pots that should control how quiet or loud the amount of audio from each of the three springs is, do not appear to do their job properly - there is some reduction / gain in volume in the sweep of each of them, but they are always on fairly loudly and I can make the even louder by sweeping the pot, but I cannot get them to fade away to zero volume. Did I use the wrong pot? I used a lot of 470K pots because I had a bag of them and I'd previously made a circuit where I blended between two circuits using a 100K pot and it always allowed some sound of each circuit through no matter where it was set, so I thought a pot with a higher value should fix that... Maybe I was wrong? I'm now wondering if lug 1 of those pots should be connected to ground?

The pot marked feedback... well in my naivety, I thought I could use this to send the circuit signal back into itself to create a feedback loop that would be cool... but the more I look at this bit, the more dubious it looks to me, as if it's probably allowing dry audio to flood through the whole circuit instead of doing what I think it should...

I think that's all the problems - the speaker works and makes sound and the springs rattle away and sound does leave the circuit, but I want more control over which springs I can hear and to be able to get that high pass working!

I hope this makes some sort of sense to some of you - any offers / suggestions / threats / blasphemies gratefully received to the usual address

[anotherjim]

High pass filters paired like that ought to work, but...

Did you put them in line with a 4ohm speaker? Against 4ohm, the ganged pot isn't going to do much until it's close to 0 and is just shorting it out.

A better place for filters is at the amplifier input or pre-amp stage. That is between the volume pot wiper and the 386 input.

A better place to feedback to is the input to the 386. There's probably an elegant way to do that, but needs thinking about. The inelegant way is to connect the feedback pot to the input volume wiper.

The output from the SHO could also do with some highpass since the discs could still pick up some unwanted rumble. A smaller cap in series placed after the SHO output cap might be enough if you don't want to mod the SHO.

[moid]

Thanks AnotherJim. I had a go at what you suggested this evening but something's not gone quite to plan alas (AKA I have screwed up somewhere). I moved the high pass filter to where you suggested and it now works (yay), although it's a bit odd, doesn't do much until over halfway around and then it fades the bass down well (almost sounds lofi) but then kills all sound at the maximum extent of the sweep. I can live with that. The new feedback position can sometimes creative some interesting background oscillation and extra noise... some good, some bad... it would be cool if this faded out a bit at the end. I'll keep twiddling this! Part of me thinks it's cool, the other part doesn't want it to give me a constant high pitched drone sound in the background. The Bass boost control on the amplifier circuit now cuts all sound when set to being low (it might have done this before actually). The main problem is I'm not getting any sound from the piezos, so no reverb at all. Great fuzz sound though The speaker works, the springs vibrate... but nothing extra comes out. I've probably soldered something to the wrong something somewhere (why is is always last thing at night that I have time to do something complex). I will try to have another look later in the week and report back.

PS I didn't get round to adding the final high pass to the piezos - I'll try that once I've got reverb again!

[anotherjim]

Might be worth getting the hang of that high-pass filter.

Clearly, having a fixed capacitor is sensible, since variable caps are tricky and only work in small pF ranges too small for most audio. So that leaves the resistor value to vary. However, can you see that if the resistance value gets low, then it's pretty much a short circuit to ground?
The answer is to fit an additional fixed resistor in series with the variable at the grounded end. So if you turn the pot down to the minimum usable level, then if you then measure how much resistance the pot now is (out of circuit with power off), fit a fixed resistor in the ground connection equal to or a bit higher than that. Now you should be able to use the pot over its full range. As you want 2 stages of filtering, you need to fit an additional resistor in both pot grounds. I call this "padding out".
A pass filter takes out what it don't pass - duh! So yes, with a high-pass you hear reducing bass, not increasing highs.

The 386 has 2 inputs and one of them is unused and tied to ground. If you remove that ground, then you could put the feedback into that from a volume pot wiper off the SHO output. If it is too hot and easily gets unstable, you can pad it down with a resistor between the SHO output and the feedback pot. You might want a high pass to the feedback, The 386 input already has some "R" on the input to ground, so for simple, you only need to choose a cap in series with the chip input to make a single stage filter. It'll be in the nF ranges.

Going back to phase (or, really, polarity), try reversing the drive speaker connections. Feedback might be more usable one way or the other. It shouldn't make any difference to the "long" delayed sounds.
Even with the poshest delay, some feedback is going to reach a point of an unstable runaway. No avoiding that. All you can do is scale (pad) the controls for maximum usability as you see fit.

[ElectricDruid]

That's fantastic! I love the beautifully rendered carefully copied out version of the wired mayhem. That's how circuit diagrams should be - closer to the reality and less of this idealised "schematic" nonsense.

[thermionix]

Connecting 87 wires to one lug of a DC jack might be a little difficult.

[EBK]

Connecting 87 wires to one lug of a DC jack might be a little difficult.

That's not too far off from how I build things.  You just have to attach the wire end of a ring terminal to the jack and solder the 87 loose wires to the ring.  Cover the resulting mess with some gaffers tape, and no one will know how much you've obsessed over star grounding*.     (I think 7 wires is the most I've ever attached to one DC jack ground terminal in my version of reality.)

*Grounding technique (at least in guitar pedals) is more religion than science.

[thermionix]

You just have to attach the wire end of a ring terminal to the jack and solder the 87 loose wires to the ring.

But then your little ring makes a ground loop, and you're technically not star grounding anymore, and you have to lay awake at night wondering if your build can be just a little bit better than it is, and fear that someone will open it up years after your passing, and discover that your whole life was nothing but a fraud.

[EBK]

You just have to attach the wire end of a ring terminal to the jack and solder the 87 loose wires to the ring.

But then your little ring makes a ground loop, and you're technically not star grounding anymore, and you have to lay awake at night wondering if your build can be just a little bit better than it is, and fear that someone will open it up years after your passing, and discover that your whole life was nothing but a fraud.

Not a problem. 


I found a pic of my ring ground.  This was the only build I ever did this to, by the way.