Best 4558 alternatives?

[Asymmetric]

I am building a few pedals at the moment, and all of them have 4558's or similar op amps.
I have modded a few to OPA2134's which I really like, but I'm sure that there's better options available with the same pin out.
What are the clearest, lowest distortion, hi fi like op amps that share the 4558 pinout? What are the most recent ones?

[GibsonGM]

Quick question, Asymmetric - why would you need hi-fi type audio?  Guitar signals aren't very 'hi fi', and the amps are not meant to reproduce hi fi, and in fact, much of why we like them is because of this    If there's a specific type of pedal you're interested in, and a reason why you'd need better, well - that's different... 

A great replacement for the 4558 is a TL072, which would be an 'improvement' of like 1,000 times in terms of noise level, "fidelity", slew rate (that we'll never use...).  Lovely high-Z JFET-input opamp.   Also almost The Standard around guitar effects.  NE5532 would be 'next gen' above the '072.

Yup, there are TONS better opamps than the 4558, but mostly, we don't need the qualities that make them great! Our signals are <20kHz, giga bandwidth product and slew rate just don't factor in!   And some SAY that these 'really clean opamps sound sterile'...so there could be some mojo at work, too! 

Welcome to the forum, A - I'm sure others will give you their opinion, too, as this can be a debatable subject, ha ha.

[PRR]

Essentially _ALL_ general-purpose Dual opamps have the same pinout.

Anybody who markets a new/better/cheaper opamp MUST use the standard pinout to have any hope of capturing business. Nobody is gonna revise PCBs to gain 0.01% THD or to save a penny.

That said-- some opamps are mellow like an old couch-Corgi, and some are high-strung like that over-revved Yorkie next door. Any opamp will squeal above the audio band if layout and bypassing isn't good. 4558 is fairly forgiving. OPA2134 less so. There are others where you have to read the spec-sheet carefully, and even then pray for luck.

The very common TL072 is, despite the penny-price, a very very good choice for nearly all guitar circuits and most studio-gear circuits except heavy line-drivers.

[Asymmetric]

I know that the TL072 is a good OP amp, but there is one big problem with it - it's been used a million times before. I'm after something new and I don't believe in mojo! When I say hifi, I mean crystal clear. I'm making a OS compressor and I want to clean the sound up. So basically, all general purpose op's use the same pin out, and I need to look for a high slew rate, low THD. When you say that I'd need to read the spec sheet carefully, what should I be looking out for?

Thanks for the recommendation of the NE5532 I'll check it out.

What's a high-Z JFET-input opamp? How is it different to a normal one?

[karbomusic]

I'm making a OS compressor and I want to clean the sound up.

I use OPA2134s in my OS comps; you'll like 'em. However, it's an OS and the noise it creates by being a compressor, is likely going to mask most differences. I used it more so because mine has a blender circuit and like what they say about the chip whether I can hear and demonstrate it or not; haven't take the time to compare to be honest. It's also a FET opamp, 5532 is BJT IIRC hence the 2134 working well in blend situations removing the need for a buffer in front, same for the TL072 IIRC.

[PRR]

> making a OS compressor and I want to clean the sound up.

Lipstick on a pig.

Any compressor "distorts". Elaborate ones may not sound bad, by using the ear's "faults" to mask the necessary signal mangling. The Orange Squeezer(?) isn't in that class.

Feedback optos can be fairly low distortion at levels below 0.1V across the cell; but they are funky.

MUCH depends on the sidechain, its faults, and overload margins.

The cop-out would be a buffer into one of the very excellent compressors on the market. The "RNC" is well regarded among affordable compressors. Prices go from there to the sky. More bucks is not always "cleaner"; some exotic old-school boxes are prized for their "sound".

[GGBB]

I built a modified OS I called the Orange Smoothie. It is very quiet IMO. I tested 8 different opamps in circuit looking at background noise using a Cubase plugin. I wasn't being terribly scientific, but it was a fair comparison. There was a small but notable difference. I tested the RC4558, RC4580, TL072, NE5532, MC1458S, OP275, OPA2604, and OPA2134. 4 opamps showed lower noise floor than the other 4, and within each group, the levels were very close. The winners were the RC4558, RC4580, TL072 and OPA2134. I ended up using the TL072 because it was as quiet as any of them and of the four quietest had the most neutral sound. (Yes - there were notable sonic flavours to the opamps and the OPA2134 had a "hi-fi" character I felt wasn't neutral.) All of these differences were subtle but I could hear and measure them.

I'd suspect something went wrong during the test if the change in noise level between groups happened at one point (i.e. tested 4 showing one noise level, then tested 4 more at different noise level) but that's not what happened. So I suspect possibly something about the design of the chip itself interacting with the circuit made for the differences. I really don't know why.

I believe that regardless of the opamp, the stock OS can be made quieter simply by using better power supply decoupling and filtering which is what I tried to do with the smoothie. The opamp itself will probably make far less difference than simple noise prevention design considerations.

[samhay]

>I know that the TL072 is a good OP amp, but there is one big problem with it - it's been used a million times before. I'm after something new and I don't believe in mojo!

On it's own, an op-amp dooesn't sound like anything (it won't make any noise) - it is the stuff that you connect to the op-amp that you should be concerning yourself with.

[PRR]

> it is the stuff that you connect to the op-amp

Too true.

GGBB's tests show this. In some ways, the NE5532 has lower hiss voltage than the '741 family like 4558. But in this specific circuit (I am guessing) the '5532's high hiss Current shows badly.

[Asymmetric]

Thanks for the advice, I'm starting to get it now. I really love the OPA2134. The only thing now that confuses me is the pricing of OP amps.

Take a look at this:
http://uk.rs-online.com/web/p/operational-amplifiers/6605544/

It has only 1Mhz and a slew rate of 0.8V/µs for £3.81


While this one:
http://uk.rs-online.com/web/p/operational-amplifiers/5343762/

has 125Mhz and a slew rate of 950V/µs for £1.78!


What's this all about? Is there something else I'm not seeing?

[merlinb]

I know that the TL072 is a good OP amp, but

The TLE2072 is an upgraded TL072. Better slew rate, noise, input range, and THD into heavy (600 ohm) loads.  Not that you'll notice the difference in a guitar application! 

[Mark Hammer]

Here's a nice comparative chart, scanned from the 1989 edition of Horowitz and Hill's outstanding classic "The art of electronics".  It doesn't have everything currently available, but allows you to compare many if not most of the standards.

http://hammer.ampage.org/files/op-amp_chart.zip

[Hatredman]

NE5532 would be 'next gen' above the '072.

I really love the 5532 above all, it is a studio-grade OpAmp with enough power to drive a small speaker , but you have to watch out for pitfalls. It is pinout-compatible with the 4558, but unlike the OPAs, the LMs and the TLs and many others, it is not a direct replacement because it has compensation and bias requirements that need to be addressed. In other words, the circuit around the 5532 is a bit different from the common circuits we are used to with the 4558 and the others.

It`s a fine OpAmp, and I like it very much, but you just CAN`T compare it in the same  circuit with the others.

That said, I agree with everybody else that we're talking guitar here, so HiFi, steep slew rate, MHz bandwidth and other stuff are of no importance. In fact, opamps with limited specs may sound better exactly because the limiting factors are crucial for the tone. I know a Brazilian overdrive maker who use OPAs and swear by them, and another friend is crazy about the Mojo of old LM308s or JRC4558s or old russian germanium transistors and when he tries to show me the differences I just can`t hear them. But for me, on guitar applications,  in my opinion the only thing that`s better than a TL072 is a TL072 that you got for free. The Hi-Zi alone is a problem solver by itself.

[Hatredman]

http://uk.rs-online.com/web/p/operational-amplifiers/6605544/
It has only 1Mhz and a slew rate of 0.8V/µs for £3.81

http://uk.rs-online.com/web/p/operational-amplifiers/5343762/
has 125Mhz and a slew rate of 950V/µs for £1.78!

What's this all about? Is there something else I'm not seeing?

The LM7171 is a video driver. Just because the bandwidth is wide it doesn't mean that it will perform accordingly in the audio range. It may, it may not.

The difference on price is based not on the specs, but availability and supply/demand. Video drivers are used everywhere on TVs, DVD players and so on. Huge market worldwide. The OPAs are HiFi components with limited use, as consumer audio products use cheaper options or dedicated audio chips (sometimes everything is digital, the only analog part is the output transistors of the Class D power amplifier). Low demand, higher price.

[Asymmetric]

Thanks everyone. That's cleared a lot up. Just downloaded the thing from Mark which outlines a lot.

At the moment I can't decide between the OPA2604 and the TLE2072AIP! I'll try others as time goes on, but my next purchase will be one of these two. What do you all think?

[Hatredman]

I also recommend you watch this video 'till the end and paying attention.

You'll see that the OpAmp doesn't matter on a guitar application. It's the circuit around the OpAmp that matters, specially the caps.

[GibsonGM]

If an opamp can 'slew' incredibly fast....meaning, adjust/react to a very fast frequency coming in, perhaps in the MHz...it doesn't matter at all for what we use them for.  They won't be "better".   Our frequencies of interest lie below 20kHz...way far below.  Those awesome abilities won't have any part in what we're doing.   NOISE tends to be the biggest issue to consider, and as a few have already said, good design practices and 'the things you surround the opamp with' are going to determine the quality of your build...we can debate this for a long time, I can only say that for ME, the TL07X works for everything I do, very well, with no wish for any improvement

[Hatredman]

...the TL07X works for everything I do, very well, with no wish for any improvement

Amen, Brother!

[Asymmetric]

So I guess the best thing to do would be to re create an op amp with full size resistors, caps, transistors, and diodes!

Check out page 4! http://pdf1.alldatasheet.com/datasheet-pdf/view/123848/TI/TLE2072AIP.html

[PRR]

> Is there something else I'm not seeing?

There's a 42-page datasheet. Read it. Especially the first page where they boast about the stuff they can boast about (and are mum about the non-boastable features).

The OPA277 is "precision". In particular, the input DC error voltage is 10 MICRO-Volts. A thousand times smaller than many opamps. DC accuracy is superb.

It sure has uses. If you are cloning DNA you may need temperature controlled to a small part of a degree. This might be just a few microVolts off your thermocouple.

What is the "sound" of super-accurate DC? None at all.

Jet engines are very fast, but awkward for shifting loads in a cramped warehouse. An X-acto is good for small cuts. If you are cutting mortises for a timber-frame barn you want a framing chisel. For production timber work you want a power morticer; that would be wrong for spinal surgery.