String Reverb idea...

[logey]

Thats right, string...not spring. I want to make a device that I can send a signal to, have it induce the signal into a set of strings (4? 6?) that can be tuned or detuned, and then pick it up again and track it. The picking it up again part seems pretty straight forward...I've got tons of old humbuckers that I can employ for that. The part I cant quite suss is how to get the strings to vibrate in the first place. Some sort of coil maybe? I'm thinking about something similar to a spring unit. Can a guitar pickup be run backwards? How much gain would it take to get a typical guitar string to vibrate sympathetically? Any help figuring this out would be great.
Thanks,
Boyd

[RefD]

maybe drive the strings of a pedal or lap steel by using a speaker fed by a 100w amp suspended over or under the strings...kinda like the old piano reverb trick but with a steel instead of a piano.

the return would be coming off the pickup in the steel into a DI or something.

it'd be pretty neat to use the pedals and knee levers to tweak things as the strings ring in sympathy with the material coming from the speaker.

thinking out loud.

[ashcat_lt]

This thread is apparently 260 pages long now. The link I've provided is page 75 for no good reason whatsoever. Here's a shorter one that might give you more of the gist.

You can run a pickup backwards, but it takes a suprising amount of juice, and gets hot pretty fast. A coil with a lower Z works better, I think. So far they haven't found a good way to get the coil very close to the sensing pickup (like in the middle of a strat) without inducing horrible noise and/or magnetic feedback.

Fernandez and a couple others sell kits.

[Professor]

You would probably do alright to use a tactile tranducer of some variety, just like you would use for a spring or plate reverb. I don't know of the particular part makes and models, but you should be able to find them easily enough if you look up a DIY plate reverb thread. Or maybe just buy a cheap replacement spring reverb tank for a guitar amp and tear it open to take the parts you need - basically replacing the spring with your strings.

-Jeremy

[ubertar]

Sounds sort of like an ebow, except the pulse of the electromagnet is driven by an external signal, rather than from the instrument pickups. You need a coil with fewer windings of thicker wire than a pickup. The core should be iron, rather than a magnet. The signal needs to be amplified before hitting the electromagnet, because it needs a lot more juice to do its thing than what an audio signal is going to give it. But all in all, the concept is very similar to an ebow. I've seen the patent online, complete with drawings. I'm not sure where it is, but it's there to be found.

Replacing the spring in a spring 'verb with a string might be easier, but you'll probably learn more from trying it this way. Sometimes easier is better though.

[alf]

Nicolas Collins made a backwards guitar...

[logey]

hmm...lots of different directions to consider.
I think the first thing I am going to play around with it the "simulated string verb" otherwise known as an electric guitar placed in front of a cranked amp and recorded direct. I hope to be able to determine whether or not this is a sound/idea that I even want to pursue.
If it is, I think I will head down the path of using a speaker driver to somehow induct the signal to the strings and a guitar pickup to get it back. I have a 4ohm 6 inch driver that came with my Champion 600 that I would be willing to sacrifice for this project (I swapped it out for a Weber). Any thoughts on coupling the driver to the strings? Does it need to be physically coupled or will the EM field produced be enough to vibrate the strings?

[wenzel.hellgren]

Sounds sort of like an ebow, except the pulse of the electromagnet is driven by an external signal, rather than from the instrument pickups. You need a coil with fewer windings of thicker wire than a pickup. The core should be iron, rather than a magnet. The signal needs to be amplified before hitting the electromagnet, because it needs a lot more juice to do its thing than what an audio signal is going to give it. But all in all, the concept is very similar to an ebow. I've seen the patent online, complete with drawings. I'm not sure where it is, but it's there to be found.

Replacing the spring in a spring 'verb with a string might be easier, but you'll probably learn more from trying it this way. Sometimes easier is better though.

Check out the Sustainiac.

http://www.sustainiac.com/

The Stealth is pretty much what you were talking about up there Ubertar.

the model C is the "physical" way to do it.

[logey]

Sounds sort of like an ebow, except the pulse of the electromagnet is driven by an external signal, rather than from the instrument pickups. You need a coil with fewer windings of thicker wire than a pickup. The core should be iron, rather than a magnet. The signal needs to be amplified before hitting the electromagnet, because it needs a lot more juice to do its thing than what an audio signal is going to give it. But all in all, the concept is very similar to an ebow. I've seen the patent online, complete with drawings. I'm not sure where it is, but it's there to be found.

Replacing the spring in a spring 'verb with a string might be easier, but you'll probably learn more from trying it this way. Sometimes easier is better though.

Check out the Sustainiac.

http://www.sustainiac.com/

The Stealth is pretty much what you were talking about up there Ubertar.

the model C is the "physical" way to do it.

The Sustainiac seems to be more about feedback though...how would a system like this be used to process an external signal? Unless I'm misunderstanding I don't think it can.

[logey]

Sounds sort of like an ebow, except the pulse of the electromagnet is driven by an external signal, rather than from the instrument pickups. You need a coil with fewer windings of thicker wire than a pickup. The core should be iron, rather than a magnet. The signal needs to be amplified before hitting the electromagnet, because it needs a lot more juice to do its thing than what an audio signal is going to give it. But all in all, the concept is very similar to an ebow. I've seen the patent online, complete with drawings. I'm not sure where it is, but it's there to be found.

Replacing the spring in a spring 'verb with a string might be easier, but you'll probably learn more from trying it this way. Sometimes easier is better though.

Ubertar,
Could a guitar speaker driven by a guitar amp produce an em field that might work this way?

[newfuturevintage]

This thread is apparently 260 pages long now. The link I've provided is page 75 for no good reason whatsoever. Here's a shorter one that might give you more of the gist.

You can run a pickup backwards, but it takes a suprising amount of juice, and gets hot pretty fast. A coil with a lower Z works better, I think. So far they haven't found a good way to get the coil very close to the sensing pickup (like in the middle of a strat) without inducing horrible noise and/or magnetic feedback.

Fernandez and a couple others sell kits.

I've build one of these from the forum link above.

my effort:
http://projectguitar.ibforums.com/index ... t&p=358727

A couple things to note: yep, ashcat's totally right about about induced noise into the system when the coil and the driver are in close proximity. A very real phenomenon. In the OP's idea, this isn't a big deal: it's not a guitar, doesn't have frets, and the driver coil can be located away from the PU.

The amp also doesn't have to be self contained to a guitar, so it doesn't need to be battery powered, which is the biggest hurtle to the sustainer project.

In my testing, I found I could use the output from a guitar amp (I used a crate powerblock), and it'll drive the living shit out of guitar strings. It'll also arc voltage across to the strings if given the opportunity, so be careful.

Other general concers: the larger string diameter, the easier they are to move due to greater ferric content (ie: they're more magnetic). In sustainer systems, the High E string rarely moves, and as you get more to the Low E string, sometimes it moves too much and hammers against the driver coil.

The thing your missing logey, is that sustainers don't emit just any EM field. They're fed the output of the pickup of the guitar, so they emit something the strings are already sympathetically vibrating to. If this EM is in phase with the strings, you get a reinforcement of the fundamental. If it's out of phase, you get a cancellation of the fundamental, and a rise in harmonics, I believe in relation to the position of the driver coil.

In your application, you'd feed an echo send to the driver amp, on to the driver coil, then and pull the pickup's output back to the mix via a DI or a guitar amp. USE A HUMBUCKER. Single coils tend to pick up a lot of interference from the driver coil.

For an amp, I'd try using a headphone amp first, if that doesn't work, then a low-ish power stereo amp, ~ 10 watts, maybe. I think the little "T" amps would be great. For reference, the little LM386 chip I'm using in my guitar drives everything but the High E and B strings (it'll do the B with a fresh battery), and only has a 400mW output. It's main failing is when it distorts, the pickup gets a fizz induced from the square wave sent to the coil. With a more powerful amp, this won't be a problem.

I'd think something like a cheap lapsteel or a homebrewed lapsteel would work nicely. I'd also use 6 strings, all wound, maybe some from a bass, tuned to various notes in the key of the song your playing with.

If this doesn't work, get a buttkicker and bolt the f*cker to the plank.

Also put a felt damper on it to adjust the decay time!

[ubertar]

Ubertar,
Could a guitar speaker driven by a guitar amp produce an em field that might work this way?[/quote]

I doubt it. You're better off just coiling your own electromagnet. A speaker is like a dynamic mic in reverse-- feed it an electric current and the coil will move in relation to the magnet. That doesn't help you unless you attach it to the strings so it physically vibrates the strings. A speaker has a permanent magnet. To move the strings, you need an electromagnet-- a coil of wire around an iron core. The coil should be tight around the core, not mobile like a speaker coil. Use thick magnet wire. I'm not sure how many turns you'll need-- you'll have to experiment.

It may turn out that the effect isn't that different from hitting the back of the guitar with a soft mallet in time with the music... hopefully it's more interesting than that.

[ashcat_lt]

Honestly, I haven't read every post over there, but I'm pretty sure that the answer to any question you might come up with re: electromagnetic drivers can be found via one of those links I posted.

As newfuturevintage said, there are a number of design criteria which are a bit different thanks to the fact that nobody actually has to play this thing.

You could quite easily put the driver something like 20" away from the pickup, which should alleviate some of the EMI issue. You can get away with more "imposing" EM shielding/shaping measures as well.

On top of that, you can have some of the strings significantly shorter than what would be practical on a guitar. This would allow heavier guages, which are easier to excite.

I'm seeing something more like a zither. I'm sure that 6 strings won't be enough. No matter how you transduce the signal into the strings, they will only resonate worth a damn to inputs which are closely harmonically related to their fundamental frequency.

This guy builds spring reverbs. Seems like spring reverbs work by twisting the springs. Not saying it wouldn't do something interesting if you replaced the spring with a guitar string, but I don't think that it will be what you're looking for.

Acoustic coupling requires a lot more power and a lot more space. By the time you've built a resonant body for the strings, an amplifier to excite that body, and a resonably "soundproof" box to contain the whole thing, you might as well get a cheap piano. At least it would be good for other things.

I thought about the buttkicker thing, too. I'm afraid, though, that those things don't actually tranduce the input so much as shake (at their own frequency) when they recieve a signal. In that case it would likely be something like whacking the back of the guitar with a mallet (or your fist). That'll work to get feedback started in a situation where acoustic coupling can take over, but will likely get us a non-harmonic output.

If it was me, I'd just instantiate Pianoverb and be done with it...

[logey]

Honestly, I haven't read every post over there, but I'm pretty sure that the answer to any question you might come up with re: electromagnetic drivers can be found via one of those links I posted.

There is a lot of info in those threads...thanks for that. Unfortunately, as you pointed out later, most of it pertains to inducing feedback, rather than tranducing a signal to the strings.

As newfuturevintage said, there are a number of design criteria which are a bit different thanks to the fact that nobody actually has to play this thing.
You could quit easily put the driver something like 20" away from the pickup, which should alleviate some of the EMI issue. You can get away with more "imposing" EM shielding/shaping measures as well.

Good points.

On top of that, you can have some of the strings significantly shorter than what would be practical on a guitar. This would allow heavier guages, which are easier to excite.
I'm seeing something more like a zither. I'm sure that 6 strings won't be enough. No matter how you transduce the signal into the strings, they will only resonate worth a damn to inputs which are closely harmonically related to their fundamental frequency.

I actually am thinking of 6 to 10 strings...I think you're right in your assessment, but thats actually kind of the effect I'm looking for...a tuned resonant effect.

This guy builds spring reverbs. Seems like spring reverbs work by twisting the springs. Not saying it wouldn't do something interesting if you replaced the spring with a guitar string, but I don't think that it will be what you're looking for.

I also wonder about the feasibility of something like this as the strings will be under a great deal of tension...I doubt that the structure of most spring reverbs would be able to sustain that load.

Acoustic coupling requires a lot more power and a lot more space. By the time you've built a resonant body for the strings, an amplifier to excite that body, and a resonably "soundproof" box to contain the whole thing, you might as well get a cheap piano. At least it would be good for other things.

I do have a small grand, and yes it is good for other things...the problem with using it in this application is one of getting the strings to vibrate without creating so much sound pressure that it masks the effect produced?.

I thought about the buttkicker thing, too. I'm afraid, though, that those things don't actually tranduce the input so much as shake (at their own frequency) when they recieve a signal. In that case it would likely be something like whacking the back of the guitar with a mallet (or your fist). That'll work to get feedback started in a situation where acoustic coupling can take over, but will likely get us a non-harmonic output.

If it was me, I'd just instantiate Pianoverb and be done with it...

Yep, Pianoverb is awesome...maybe I should just play with it more and see if I can get it to do what I want. Pianoverb can be tuned, but only as a whole...no individual notes or scales can be emphasized. It is a great tool though.

[ashcat_lt]

I do have a small grand, and yes it is good for other things...the problem with using it in this application is one of getting the strings to vibrate without creating so much sound pressure that it masks the effect produced?.

Magnetic pickup? I've never tried it, but I'm pretty sure you could figure something out. Anything from somehow mounting a couple of guitar pickups in various areas to winding your own transducer. Might even be a commercial unit out there? Got a Rhodes you don't need? Piezo could work too.