Best OP-amp for EQ circuit?

[neskor]

I'm trying to do upgrade of OP in my JBL 5235 crossover.
OP-amp are NE5532 and TL074.
Any suggestions for best sounding OP-amp in EQ circuit?

 

[Speedskater]

First two questions:
1] Do you have a good oscilloscope?
2] Do you know how to use it?

 

[mickeymoose]

First off: Without the schematic no qualified answer can be given (don't ask me (us) searching for it!)
Second: What is it that you would like to improve? Those opamps, even after all these years, have high ratings. E

 

[neskor]

Here is the schematic.
I don't have oscilloscope 🙁
I would like ti improve transparency

 

[AndrewT]

Each low pass, high pass, buffer and other use opamp will need to have it's own important parameters identified. Using that information, selected opamps to suit each duty can be inserted and checked for correct operation.

This is not a swap and guess tweak.

 

[Mooly]

It is important to check for stability when swapping opamps but I think the OPA2134 and OPA4134 would be a safe swap here. It all looks as though it is fairly well compensated already.

 

[jcx]

all of the Z are in the range you expect to see fet input op amps - I'd guess that the 5532 was used on the output for its better line driving ability compared to the TL074/2

today there are fet input "audio" op amps with much better outputs - OPA4132/4 and 2132/4 are esy to find in DIP

most recent "audio" op amps are only offered in surface mount - also watch the supply V rating - newer processes often don't handle the +/-18 V supply used on that schematic

 

[Simon B]

I'm very dubious about there being any value in changing the op-amps without having a clearly identified reason for doing so. Improving "transparency" is a bit of a nebulous starting point for making circuit changes.

Mooly refers to some newer devices as being a "safe swap". They may well work fine in this circuit, but that isn't the same thing as working better.

JCX talks about more modern op-amps with "much better outputs" - from the context I think he means they can source more current, but is it needed?

There's also the small matter of changing them over without making a pig's ear of the pcb. Funnily enough, I'd say the point at which my soldering started to get good was around about the time I got my first scope. And by good I mean able to solder and unsolder devices to a pcb repeatedly without killing it.

I think neskor, that you'd be much better advised to spend time on the things that might well have suffered some degradation over time - switches, pots, contacts of all kinds.

Andrew T has it exactly:

...This is not a swap and guess tweak.

As does mickeymoose:

...What is it that you would like to improve? Those opamps, even after all these years, have high ratings.

Fix the bits that might actually need fixing. Not the ones that almost certainly don't.

 

[jcx]

I just took the question as whether there were better speced op amps for that circuit

I agree that there would be little reason to suspect the swap would improve performance audibly

Substituting the OPAx134 would perhaps lower the noise floor by few dB at the input

its a toss up whether the 5532 current noise is worse than the OPA V-noise with those source/feedback Z - unlikely to be heard with any reasonable system gain structure


I have directly observed both the TL07x and NE5532/4 outputs behaving poorly by eyeball on my oscilloscope – the early generation monolithic processes couldn’t make decent PNP

With buffered output the TL07x can provide remarkably low distortion but you really don’t want them driving any load/sinking even a few mA

The 5534 output is better over audio frequencies, distortion with load holding up decently but the sinking/sourcing output dynamic Z is still vastly different at higher frequency, the pull down is way slower than the NPN follower in the output


The shown circuit doesn’t give the conditions that show the output weakness to full effect but the worst case ~ 4 kOhm load for the TL074 would probably be visible in distortion analysis if the unit uses pro signal levels


so replacing the TL074 with OPA4134 would likely result in measurable lower noise, distortion - which admittedly may not be "audible" - maybe less justification for changing the 5532 if sticking with DIP op amps - both choices can be improved on by using smt adapter boards and even more recent chips like the OPA164x

 

[Simon B]

I just took the question as whether there were better speced op amps for that circuit

I agree that there would be little reason to suspect the swap would improve performance audibly

You're quite right jcx, the OP's question was just about op-amps.

I have strayed from that in my answer, though justifiably, I think. I'd certainly want to know what it was I was trying to fix ("imporving transparency" just isn't good enough for me) before setting to swapping active devices. It concerns me perhaps more than it should that folk will think of swapping op-amps before they clean/replace pots and switches.

 

[AndrewT]

Does that make the 5534/2 a good candidate for the ClassA biasing modification?

Or does that indicate we should be taking the output from PIN5?

 

[simon7000]

I would change out the 25/25 electrolytic caps in the HF section first. I don't know if you can fit a film cap in there. C11/111 doesn't need to be 25uf. Try one side first to see if you can hear a difference.

 

[skysound]

Use OPA2132 or SMD to DIP adapters and OPA827.

Personally, I would not use the 2134 or it's quad version. The 827's are twice the price of even the 0132 series chips, but they are superlative. To me, the 0132 series and 827 are upgrades that the trained ear can hear readily. Very low distortion; quite transparent. With such low distortion IC's, excellent and very sharp imaging are possible with stereo program when near field or a minimal reverberant field is possible.