OpAmps for DIY Bass & Guitar Pedals

[Jim Y]

Special low-noise for mic pre-amps maybe, but the rest?
Anything.
I don't see the need for fast amps in audio unless you really want them to pick up mobile phones (a source of interference that wasn't around when many classic circuits were designed).
For instruments - is a little noise a bad thing? - the classic tracks from the past must contain a lot of noise. I have a half-brained theory that a little noise makes a bed for the tone to sit on.

One of my favorite guitar sounds belongs to one Wilko Johnson (of Dr Feelgood) - a Telecaster straight into a H|H IC100 combo amp. This amp exclusively used uA741 chips.
Of course, these 741's were the original 14pin DIL packaged ones, and we all know they sound both warmer and brighter than the 8pin DIL or even the metal can ones.

[khanyz]

^ IMHO you are kind of right in that noise is an important component in Music but unless you're the J&MC, you need to control/tune it to your needs.

You can take a noisy circuit, filter out parts and make a great sound. This can also lead to unexpected results, both good and bad.

However, you can also take a very clean circuit, add noise/distortion (phase and amplitude) by adding components (such as diodes, capacitors etc.) and also get a great sound, but usually with some added flexibility. For this you need to start with a clean/linear circuit and this requires components which can operate well outside of the audio range, to provide headroom.

As far as interference goes, you always need to shield the circuits from external sources, but also to shield external sources from the cicuit. Distortion circuits, in particular, can operate at frequencies well above audio. This is done with a metal case (Faraday Cage) or similar, which is connected to ground or another potential reservior.

Confucios says there are many roads to the same destination, some have been trodden before, some are dirty but all are rock.

Cheers.

[Jim Y]

Well of course you can screen your build - but you can't always control whatever else is going to be connected. I've seen schematics of modern commercial designs with ferrite beads and suppression modules at every connector. The cell-phone can find it's way in anywhere. That said, your circuit doesn't have to be fast to demod any RF that get's in.
On top of that, with high speed parts you run the risk of parasitic oscillations which you won't see without a very high speed scope. This issue alone stops me wanting to to use fast slew-rate or high GBW parts. I usually only do instrument electronics myself and there really isn't much over 8Khz to consider.

[chroma]

can you still get 2604s? i thought they were out of production.

and... the tle2072 is another opamp that sounds better than an 072, is about half the price of the 2604/2227/etc. and is usually safe to replace in tl072 circuits... worth a shot at least...

[Jim Y]

I return to this thread since reading up on the subject of fast op-amps in audio.
Seems that a fast amp can actually improve RF suppression. What happens is that at the top end where gain runs out then the output impedance of the amp goes sky-high. This lets RF in via the output, then the feedback network back to the input, getting demodulated by any non-linearity and output as audible noise.

I have some Event monitors where the chip-amps do exactly this if a cell phone is anywhere near. (Must get some ferrites for the speaker wires which is the cure for this apparently)
Source is here...
http://sound.westhost.com/highspeed.htm#a1
An interesting site. Don't go near it if you believe green caps sound better than orange ones (or vice versa).

[tannoyman]

How about soldering in 8 pin IC sockets to accept a variety of op-amp models, so you can experiment with different ones in circuit without the hassle of de-soldering each time?

[pilgrim_heart]

Note too that speed pretty much always have to be traded for power consumption so if you could actually redesign the OP you'd probably want to reduce the speed to whatever you actually need. Of course that's not possible when buying them ready-made but it makes sense to not go for more GBW than needed. You can make a quick calculation of this by simply dividing the GBW by the max gain (linear, so if you need dB values just round up to the nearest multiple of 20dB, then 20dB->10x, 40dB->100x etc.) you want to get from the OP in question, the result is the bandwidth of the finished amplifier circuit. Video OPs etc. shouldn't seem very appealing when doing this unless you want massive amounts of gain.

A quick guess about stuff like tube screamers where one upsets the feedback loop all the time is that slew rate might make a noticeable difference. When an OP loop is upset it often results in slewing, if it does in this case the speed at which it happens might affect how harmonics are produced etc. Could possibly be at least part of the explanation for these differences in sound when using different OPs (assuming that they are not placebo).

[PurpleCatfishBettie]

on some opamp circuits, is it a given that +- are bridged with a capacitor and resistor? i saw a short video about building a 'fuzz face', and one of the last things the guy went over was putting in a cap and resistor (or was it 2 caps?) bridging +- on the breadboard. turns out it were 1 electrolytic cap, and 1 'regular' cap:



a number of mistakes in this quick video (the narration calls it out), but that part near the end with the caps bridging -+ had me wondering about circuits with opamps in general...

[PurpleCatfishBettie]

this is probably a bad example. it kind of looks like he's hooking 2 schematics together, first a fuzz face, then something with the op amp. i was just taken aback by the part about 'be sure to bridge power and ground with these caps', and wondered if that were 'a thing' with other op amp circuits.

[Cuauhtli]

Old thread but if you look at complete schematics you'll see caps connected across the power rails. They help remove ac ripple from the dc power rails and prevent possible (unwanted) oscillation from occurring in op amps. It's often recommended to put filtering caps as close to the op amp as possible. Usually with fairly basic audio circuits it's not critical but something to be aware of.

Putting a resistor across the power rails, on the other hand....

[kouts_m]

For most guitar/bass pedal stuff, just grab some TL072s. They're everywhere, they're cheap, they sound fine. Like 90% of pedals probably use 'em or something similar (TL071/074 are just single/quad versions). Low noise enough for most dirt boxes, buffers, whatever. Can't really go wrong.

If you wanna feel slightly fancier, maybe try an NE5532. Some folks like 'em better, maybe a bit lower noise in some spots. Also super common and cheap. Good for boosts or EQs maybe.

The only other one I'd really mention off the top is the RC4558 if you're building a Tube Screamer clone, 'cause that's the sound, ya know? Not the quietest chip, but it's got that vibe.

Pro tip: Socket the damn things! Seriously. Put sockets in your build. Then you can pop 'em in and out easy if you really wanna hear if swapping makes a difference. Sometimes it does, lots of times it doesn't make as much difference as you'd think compared to the rest of the circuit.