Guitar Spring Reverb Driver

[jordancolburn]

Hey guys, love the forum so far. I've been lurking for a long time, but this is my first post. I'm an EE master's student with a have a ton of experience with theory and math, but not as much experience with practical designs and what works in the real world. I guess i'm asking for some help to learn general design rules or what practically works best from you knowledgeable folks here in the forum. Thanks in advance for any insight you can give with this project

My roommate was wanting a spring reverb for his guitar, so I designed a solid state driver circuit around This reverb tank, roughly based on The manufacturer's information. After some playing around on a breadboard with a bench supply, I made some minor modifications to come up with the circuit you see below.

It sounds great so far, but I can't seem to get it to work with power supply.

The quad op amp package I used was a TL074CN.

[Nigel Goodwin]

I suggest you have a look at some of the circuits used commercially, your's is vastly lacking in components, particularly capacitors.

[JJQuickenheim]

Hi Jordan,

This is my first post, too .

I agree with Nigel: not enough components. Try adding four or five capacitors and a couple of resistors. If you have an inductor laying around, it wouldn't hurt to throw one of those in, too. Or you can make your own inductor by wrapping some wire around a pencil. It's easy and fun.

Regards,
JJ

[jordancolburn]

Originally Posted by Nigel Goodwin

I suggest you have a look at some of the circuits used commercially, your's is vastly lacking in components, particularly capacitors.

Well, I'm adding a .1uF cap on the power supply pins to the op-amp. Also, most of the circuits I looked at used a lowpass filter around 150-300 hz on the spring reverb driver, I may or may not add that. I'm reluctant to add something just to add something, because It actually sounds very good as is and the sound is much smoother(more low frequency less high jangly ringing) than two other commercially available reverb tanks we've a/b'd it with.

If there are specific changes that follow some more conventional designs that someone can recommend, I'd love to try them out. I'm here to learn.

[Enzo]

What exactly do you mean it sounds great but won't work with power supply? How can you hear it if it won;t power up?

Accutronics has some application notes and suggested circuits. The Peavey folks have used the same circuit in oh just scads of models over the last 25 years, based on a single 4558 op amp. Fender tube amps generally use a tube driving a transformer. The transformer driven ones use a low impedance input on the reverb unit. The solid state stuff usually uses one of the higher impedance types. Fender solid state amps would also have successful designs in them. I agree with the above poster who suggested looking at some commercial designs. Not just to copy them, but to see what they did and learn a bit about why.

The suggestion to throw extra parts in there is only half a joke. The recovery stage needs to be fairly high in gain, so some caps to prevent HF oscillation would be good practice. And you already mentioned rolling off the bottom end fed to it. Bottom makes mud in reverb. OK for catfish, but bad for reverb. That was a 150Hz high pass filter, not low pass. No need to reverberate rumble. And nothing wrong with rolling off high end. Guitar amps generally play through speakers that are rolling off by 3kHz, and certainly not much over 5kHz.

[jordancolburn]

Originally Posted by Enzo

What exactly do you mean it sounds great but won't work with power supply? How can you hear it if it won;t power up?

Sorry, I should clarify, by power supply, I meant the unregulated split supply design in the schematic. It works with the bench supply I have on a simple prototyping board.

Accutronics has some application notes and suggested circuits. The Peavey folks have used the same circuit in oh just scads of models over the last 25 years, based on a single 4558 op amp.

I've seen both the accutronics and Peavey circuits, and based my initial design around the accutronics. My firsts tests wouldn't work so I ended up changing a lot all at once and came up with the circuit above. Now that I think about it I switched from my power supply design to the bench supply, so I might want to retry some of the accutronics stuff.

The suggestion to throw extra parts in there is only half a joke. The recovery stage needs to be fairly high in gain, so some caps to prevent HF oscillation would be good practice. And you already mentioned rolling off the bottom end fed to it. Bottom makes mud in reverb. OK for catfish, but bad for reverb. That was a 150Hz high pass filter, not low pass. No need to reverberate rumble. And nothing wrong with rolling off high end. Guitar amps generally play through speakers that are rolling off by 3kHz, and certainly not much over 5kHz.

Good, I thought so, haha (takes a while to get the "feel" of a new forum). The recovery stage is set to have a gain of 100 right now, and for caps to prevent high frequency oscillation, where would I put these in the circuit?

Yeah, I meant High pass filter for the 150hz to take the mud out although, it really doesn't sound bad at all as is. Considering making a simple adjustable filter so the cutoff can be varied into the tank as well. Thanks for the help.

[Enzo]

Oh. If it works on bench power, then what is different about your made up supply? Does it produce similar voltages, and are they refernced to ground properly? Is the power supply ground showing continuity to the reverb circuit grounds? When it doesn;t work, describe the failure mode. Does it sit ther silent? Of just hum? Or pass signal but sound like crap? It doesn;t really need to be regulated all that well.

Oscillation squashers? A small value cap in the feedback loop of the op amp will roll off the high freq response. If you are an EE, the calculate what needs to go in the op amp gain setting feedback loop to roll the stage's response off over any useful audio, say over 10kHz or something.

[jordancolburn]

I tested the supply first and everything gave great results with just the power supply. The ground was referenced right and the voltages were right on the output, but when everything was hooked up things were noisy and erratic.

Now that the reverb is troubleshooted with the bench supply, I hooked it up to the designed unregulated supply and it works great with even lower noise than the bench (it's a cheap little prototyping board multifunction board).

Oscillation squashers? A small value cap in the feedback loop of the op amp will roll off the high freq response. If you are an EE, the calculate what needs to go in the op amp gain setting feedback loop to roll the stage's response off over any useful audio, say over 10kHz or something.

Thanks, I wasn't sure exact which frequencies you were saying had oscillation issues(eg. audio, or much higher, in the 50kHz and up range). I did find a site that discusses this and many other practical issues that apply to audio op amp design(specific to the chu-moy, but applies to a lot of other applications) here:
Working with Cranky Op-Amps

Now all thats left to do is make the HP into the tank and implement the LP on the recovery circuit and put it in a nice enclosure.