Author Topic: question about integrated circuits  (Read 2287 times)

9Lives

  • Solder Ninja
  • ****
  • Posts: 313
  • Grant Duffey
    • View Profile
question about integrated circuits
« on: May 24, 2012, 06:55:40 AM »
Good morning everyone. Everyone talks about swapping ICs for different flavors of sounds. I'm guilty my self. What I can't seem to find much info on this question. Lets say we have a typical tube screamerish circuit that uses the jrc4558. The schematic of that chip shows a bunch of transistors linked together. When you swap it with an op amp with fet input, what exactly happens? Is it  just luck when it works as a direct replacement. I can't help but love the opa2134 pair I have put in  my maxon od808 clone made by sabrotones layouts. But I can't help but to feel like I'm incorrectly using these chips. So is this acceptable? My ears like it.. But is the circuit really using these ICs full potential?

mgwhit

  • Electron Doctor
  • *****
  • Posts: 1071
  • Matt
    • View Profile
Re: question about integrated circuits
« Reply #1 on: May 24, 2012, 09:26:40 AM »
It's not "just luck", but since very few of us around here are Electrical Engineers, sometime I think it might as well be.  Obviously certain op-amps  became well known in pedal circles because they came stock in an original (or well-liked version of a) product.  Likewise, some got poor reputations.

The DIY community has been all over this for at least 20 years.  You've heard that if you give a billion monkeys a billion typewriters for a billion years you may just get Shakespeare?  There may not be a billion of us, but there's a small, finite number of available op-amps and ear-testing them in an 808 isn't writing Shakespeare.  Once you find the ones you like, you look for ones with similar specs and try them out.  Lather, rinse, repeat.

Even when a (modern, boutique) pedal builder designs an overdrive circuit around a specific chip, I think they're building around their perceptions of how it sounds (typically in an 808), not specifically around the electrical properties of a given op-amp. Maxon/Ibanez certainly weren't looking for the best op-amp for the Tube Screamer or they wouldn't have changed it to something worse down the line.  Maybe ProCo had intuitions about using the LM308 in the Rat.  Dunno.  That's not to diss any of the builders out there, but none of these op-amps were designed to sound awesome in overdrive circuits, and most Electrical Engineers are not building guitar pedals for a living.

R.G. Keen's article The Technology of the Tube Screamer tackles the op-amp, and basically says that the good ones are the ones that "recover from overdrive gracefully".  Good luck finding that in the data sheets, but there you have it.

culturejam

  • Electron Doctor
  • *****
  • Posts: 4633
  • Function f(x)
    • View Profile
Re: question about integrated circuits
« Reply #2 on: May 24, 2012, 12:14:38 PM »
You can swap them because they have the same function and the same pin assignments.

But they are somewhat different on the inside, and these difference may or may not be audible in effects circuits. FET input op amps, for example, have higher input impedance and lower output impedance than their BJT counterparts. But the way most circuits are set up, this difference is negligible.

The biggest difference, to me, is in how they distort when the gain is cranked up to make the output exceed the supply rail limit (ie - clipping). The most obviously audible example of this I can give is the Rat circuit. The difference between an OP07 and CA3140 is really quite stark.

And sometimes they produce a slightly different waveform in an LFO. Not sure what characteristics influence this, but it's something I've noticed.

9Lives

  • Solder Ninja
  • ****
  • Posts: 313
  • Grant Duffey
    • View Profile
Re: question about integrated circuits
« Reply #3 on: May 25, 2012, 07:34:37 PM »
ok, so let me ask the two of you this. I build the maxon od808 off of the sabrotone vero layout. I did the buffered bypass mod from amz project lab (cant remember which one exactly) so the first amplifier branches off to the bypass side of the dpdt w/ a 100k resistor to gnd with a 10uF +- cap. The circuit is similar to the tubescreamer except instead of the 2 transistor buffers it uses 2 dual op amps. Doing this was intresting to me bc one op amp is the input buffer(and bypass)/clipping portion  and the other is the tone stack and out put buffer. It also let a wide ope pole for the led. Very simple, and space efficient. I feel like I've build a very useful build and I use it alot.  So knowing what I know and reading the info. Orman says he perfers fet input buffers for High inputZ and low outputZ and BJT for output for driving capabilities. So I started playing with different op amps and I use opa2134 for input and clipping stage. 1st for the impedence and for the bypass buffer seemed like a good compromise. I also preferred the sound over the other op amps which I have a buttload. I used the LM833 and NE5532 for the 2nd. Ended up using the 833.. I battled back and forth and I felt like the 833 was a little more warm.. Maybe I'm wrong. Also tried LM4562.. I could almost hear the cleam signal along with it. To hi fi. So tell me fellas am I on to somthing here or way off track?