Using Op-amps as reverb driver / recovery

[PositiveNegativeMan]

Does anyone here have much experience of this? I want to incorporate a spring reverb to an amp I did a while ago. My intention when I first built it was to keep it as small as reasonably possible. Now I want to add things in!! It's barely got space for the reverb so adding a tube driver is out of the question. There are a few things I can tweak to get a little extra space. So OP AMP seem the obvious choice - but I have no experience working with them. I know that I could build a stand alone reverb, but I don't want to. Building the smallest amp I could get away with was an interesting project and was almost as much fun as it was frustrating!!
Has anyone any experience?

[tedmich]

op amp driven reverbs are very common, PV typically used a dual 4558 and there are kits available (google amp reverb kit)

FYI the Belton BTDR-1H digital reverb module is tiny, cheap and VERY well reviewed, and an easy op amp circuit is available.

Reviews here:Belton BTDR-1H Digital Reverb Module

Belton DigiLog Reverb ?(neunaber technology - WET? Effects Pedals)?

kit:General Guitar Gadgets - Reverb Complete Kit


schematic:
http://www.generalguitargadgets.com/pdf/ggg_verb_sc.pdf

[Enzo]

If you go to the Accutronics web site, they have sample drive circuits.

But I agree with tedmich, the PV circuit is simple and has been used for decades. Steal it.

[Robert M. Martinelli]

Another design that is probably worth a look is the op-amp reverb driver/recovery section that can be found in Laney amps, simple, effective and easy to tweak IMHO.
JM2CW
Cheers
Bob

[PositiveNegativeMan]

Thanks!!

Thank you for the suggestions, I will begin my studies / planning. I think when this is finished the amp should be rechristened as The TARDIS...

[jrfrond]

The thing that you need to keep in mind is that, if you are using garden-variety opamps, you need to use a reverb pan with higher input impedance vs. a tube-driven circuit. The typical Fender tube Class A SE reverb driver uses a transformer to deliver the drive signal into an 8 ohm pan input, much like any Class A SE guitar amp. If you hung a speaker there, instead of a pan, you'd have audio output (a neat way to test these circuits BTW). An opamp circuit uses a pan that has a MUCH higher input impedance, typically 150 ohms, because the opamps cannot deliver much current.

I find that, to my ears, puny reverb drive current = puny reverb tone. If you want to go for the SS approach, you might want to enlist the help of small power amp IC's, e.g. TDA2614 (6W), LM380 (2W), LM384 (5W), etc. and drive a low impedance pan. It's really a SS version of the Fender circuit, but you don't need the transformer. The ability to deliver good drive current will improve the reverb tone tremendously.

Recovery into opamps is easy, because current drive isn't a factor.

If you are scratch-designing a circuit, keep in mind that you need to limit low-frequencies in the drive signal utilizing low-value coupling caps. Excessive lows will overdrive the springs and cause them to "boing". Not a pretty sound.

[Enzo]

I agree with John about the wimpy drive. Look at any number of SS designs and you will see a little push pull transistor drive stage after the op amp. The suggested circuits at Accutronics feature this approach. The appeal of an approach like the Peavey is its simplicity, but you can get fancier easy enough.

[Robert M. Martinelli]

Yep, I totally agree, that's one of the reasons why I stated the Laney reverb drive / recovery section was well worth a look, as, to increase the driver's output current capabilities, they "bridged" the two sections of a dual op-amp.
Cheers
Bob

[tedmich]

The thing that you need to keep in mind is that, if you are using garden-variety opamps, you need to use a reverb pan with higher input impedance vs. a tube-driven circuit. The typical Fender tube Class A SE reverb driver uses a transformer to deliver the drive signal into an 8 ohm pan input, much like any Class A SE guitar amp. If you hung a speaker there, instead of a pan, you'd have audio output (a neat way to test these circuits BTW). An opamp circuit uses a pan that has a MUCH higher input impedance, typically 150 ohms, because the opamps cannot deliver much current.

I find that, to my ears, puny reverb drive current = puny reverb tone. If you want to go for the SS approach, you might want to enlist the help of small power amp IC's, e.g. TDA2614 (6W), LM380 (2W), LM384 (5W), etc. and drive a low impedance pan. It's really a SS version of the Fender circuit, but you don't need the transformer. The ability to deliver good drive current will improve the reverb tone tremendously.

Recovery into opamps is easy, because current drive isn't a factor.

Sounds like you and Rod Elliott think alike


his design with a NE5532 and discrete transistor PP drive looks like it might sound great, like all the stuff I've bought from him has.

[jrfrond]

Sounds like you and Rod Elliott think alike

It sure seems that way!

[Koreth]

Mesa/Boogie did something similar in their Nomad line of amps, with a 5532 dual op-amp as the drive and recovery stages for the reverb.

[tedmich]

NE5532 are highly respected as good (and relatively cheap) dual audio Op amps, along with the more expensive OPA2134. John, Enzo, Robert and Rod Elliott agree that the section of the circuit into the reverb pan needs to move a good amount of current as opposed to the wide voltage swing, low current of a typical Op amp circuit. This is why John suggested, and Rod designed, a circuit using a beefier audio amplifier front end, with the standard op amp for the signal return after the pan. The theory makes sense and I think I'll try Rod's Elliott Sound Product kit once I get some extra $. Or maybe just go with a Belton D-verb and a nice NE5532 or OPA2134 driving it.

[jbefumo]

Heh -- great minds think alike: Little Dragon

Also wanted the smallest possible package, and now am thinking of adding reverb. Will probably use the Accutronics BTDR-2 digital reverb module.

Joe

[Wombaticus]

I've used the LM386 with good success as a reverb driver. There are a couple of tweaks that I would recommend to Elliot's design:

1. Add a 10uF 25V cap between pin seven and ground to increase PSSR/reduce hum.
2. Add an 18K resistance and 10uF (non-polar) cap in series between pins 1 and 5. This will knock back the voltage gain from 20 to just a bit above 9, which is the minimum closed loop gain where stability is guaranteed.
3. You'll probably need to tweak the input filter to make it suitable for the stage you're feeding it from. In my case I used a 500pF coupling cap with input pin 3 attached to the output of a fixed 1Meg / 56K divider (i.e. no dwell control)