Replace OP amps. Completely pointless, or not?

[Sokel]

You listen to music with half of its power at 6 kHz or above? Or you are just interested in measurement for measurement's sake?

There is a valid argument for those of us who like measurements and that comes for the very class D amp's companies as in their evolution they state higher and higher "useful" bandwidth (last Purifi goes to 80kHz if I'm not mistaken) .

So,isn't it sane to be measured against their claims?Or else what good those 80kHz are? (measurements wise always,we don't confuse that with audibility,can't have it both ways if we care and knowing that the true limits are vastly lower)

 

[Roland68]

Sorry for the bump, but I thought that it was already determined that the LM4562 (and OPA2134) are stable in the A07?

The post is already well over a year old, I never wrote anywhere that the LM4562 in the A07 oscillates and the rest doesn't change.
If someone has noticed this, you can certainly post the measurement protocol or refer to it.

 

[cgallery]

The post is already well over a year old, I never wrote anywhere that the LM4562 in the A07 oscillates and the rest doesn't change.
If someone has noticed this, you can certainly post the measurement protocol or refer to it.

I was just confused because the very first page seemed to indicate that the LM4562 is stable on the A07, it seemed like you may be disagreeing (the emphasis added was yours, not mine).

 

[DanielT]

Sorry for the bump, but I thought that it was already determined that the LM4562 (and OPA2134) are stable in the A07?

My tip is to skip changing the op amp in your A07. At best you hear no difference, other than imaginary ones, at worst you plug in an op amp that is not suitable for the amplifier.

Maybe you are knowledgeable in the technical aspects of amplifier design but there are a lot of people who don't seem to be, and DESPITE this lack of knowledge, they still change op amps wildly at random in their small class d amplifiers. Why?
To draw a parallel. I mean, who without technical knowledge randomly starts tinkering with a turbo in a car engine by screwing here and there, changing turbo positions/pressure in order to get more power? In the worst case, it can cause a collapse.Having said that, a professional setting of turbo pressure and so on can lead to increased power that is really felt when you press the gas. A real feeling and not some imagined better sound via op amp feeling change.

BUT, since the A07 is load dependent, it is safe to say that you should try to have suitable speakers together with that amplifier because a wrong match can create an FR that actually can be heard, see here:

Cheap amp without load dependency.

Hey there guys. Some might know I've been dealing with a load dependency issue with some cheaper amplifiers I use and my speakers (Cnotes) that results in an excessively bright speaker. I tried borrowed some cheap-ish class a/b amplifiers to test and the issue went away with those. I was...

audiosciencereview.com

Or do as I do. Get rid of your A07 and get an amp that is not load dependent and you won't have to worry.
It should be added that I think there was something wrong with my A07. It lost power the longer it was on, but when I turned it off and let it cool down and then turned it on, I felt it was more powerful again. Weird.

 

[cgallery]

My tip is to skip changing the op amp in your A07. At best you hear no difference, other than imaginary ones, at worst you plug in an op amp that is not suitable for the amplifier.

Maybe you are knowledgeable in the technical aspects of amplifier design but there are a lot of people who don't seem to be, and DESPITE this lack of knowledge, they still change op amps wildly at random in their small class d amplifiers. Why?
To draw a parallel. I mean, who without technical knowledge randomly starts tinkering with a turbo in a car engine by screwing here and there, changing turbo positions/pressure in order to get more power? In the worst case, it can cause a collapse.Having said that, a professional setting of turbo pressure and so on can lead to increased power that is really felt when you press the gas. A real feeling and not some imagined better sound via op amp feeling change.

BUT, since the A07 is load dependent, it is safe to say that you should try to have suitable speakers together with that amplifier because a wrong match can create an FR that actually can be heard, see here:

Cheap amp without load dependency.

Hey there guys. Some might know I've been dealing with a load dependency issue with some cheaper amplifiers I use and my speakers (Cnotes) that results in an excessively bright speaker. I tried borrowed some cheap-ish class a/b amplifiers to test and the issue went away with those. I was...

audiosciencereview.com

Or do as I do. Get rid of your A07 and get an amp that is not load dependent and you won't have to worry.
It should be added that I think there was something wrong with my A07. It lost power the longer it was on, but when I turned it off and let it cool down and then turned it on, I felt it was more powerful again. Weird.

I use two A07 amps (I'm biamped, they're fed by a MiniDSP). I've had these amps for several months and always felt some cymbals sounded unexpectedly harsh whether I was listening to digital or analog (vinyl) sources.

It is as if an extra dose of harshness was added to already harshly recorded cymbals. Subtle and profound at the same time is one way I've described it.

This was apparent in both the digital and vinyl copy of Change a Pace (Duke Jordan) (tons of cymbal energy). Also on Forget About It (track on Alison Krauss + Union Station live album) three cymbal hits about a minute in. Countless others, too. I listen to a lot of jazz trio recordings, don't get me started on how this impacted anything with Blakey as drummer, who always brings some extra cymbals energy.

Anyways I did finally swap to LM4562 chips on my HF amp and those cymbals are now more balanced, less tiring.

So my next question was why, how do I measure/quantify this. This is how I ended up at this thread.

Other thoughts that occured to me were whether the 5532 chips I replaced were garbage.

So I'm afraid I'm adding my name to a now rather long list of people that noticed a difference for the better.

At least, though, I'm trying to determine a way to capture this difference.

It is entirely possible I'll give up and switch to another amp before that happens though.

 

[antcollinet]

I use two A07 amps (I'm biamped, they're fed by a MiniDSP). I've had these amps for several months and always felt some cymbals sounded unexpectedly harsh whether I was listening to digital or analog (vinyl) sources.

It is as if an extra dose of harshness was added to already harshly recorded cymbals. Subtle and profound at the same time is one way I've described it.

This was apparent in both the digital and vinyl copy of Change a Pace (Duke Jordan) (tons of cymbal energy). Also on Forget About It (track on Alison Krauss + Union Station live album) three cymbal hits about a minute in. Countless others, too. I listen to a lot of jazz trio recordings, don't get me started on how this impacted anything with Blakey as drummer, who always brings some extra cymbals energy.

Anyways I did finally swap to LM4562 chips on my HF amp and those cymbals are now more balanced, less tiring.

So my next question was why, how do I measure/quantify this. This is how I ended up at this thread.

Other thoughts that occured to me were whether the 5532 chips I replaced were garbage.

So I'm afraid I'm adding my name to a now rather long list of people that noticed a difference for the better.

At least, though, I'm trying to determine a way to capture this difference.

It is entirely possible I'll give up and switch to another amp before that happens though.

Your first step in vallidating is to confirm if you can really hear a difference in controlled conditions (accurately level matched and blind etc)

Until you've done that, you've not confirmed there is anything really there, other than your subjective - perceptive bias prone - impression.

Audio Blind Testing - You Are Doing It Wrong! (Video)

Just uploaded a video on the basics of performing controlled tests in audio. It was motivated by saving myself text in having to write all of this down when telling someone who to do these tests right. And as a counter to a few online personalities to keep saying they do "blind" tests yet when...

www.audiosciencereview.com

 

[egellings]

That's how it is but..

Interesting though. No audible difference. Test via Youtube with its limitations, but the result would certainly, most likely, be the same if the test was carried out with a top notch audio reproduction solution.

With the large open loop gains of op amps allowing for large amounts of corrective feedback, I would expect to hear no difference in S.Q. with different models of op amp in circuit, so long as they are operating properly in that circuit; i.e. not oscillating or being overloaded and have specs that are adequate for the application.

 

[DanielT]

I use two A07 amps (I'm biamped, they're fed by a MiniDSP). I've had these amps for several months and always felt some cymbals sounded unexpectedly harsh whether I was listening to digital or analog (vinyl) sources.

It is as if an extra dose of harshness was added to already harshly recorded cymbals. Subtle and profound at the same time is one way I've described it.

This was apparent in both the digital and vinyl copy of Change a Pace (Duke Jordan) (tons of cymbal energy). Also on Forget About It (track on Alison Krauss + Union Station live album) three cymbal hits about a minute in. Countless others, too. I listen to a lot of jazz trio recordings, don't get me started on how this impacted anything with Blakey as drummer, who always brings some extra cymbals energy.

Anyways I did finally swap to LM4562 chips on my HF amp and those cymbals are now more balanced, less tiring.

So my next question was why, how do I measure/quantify this. This is how I ended up at this thread.

Other thoughts that occured to me were whether the 5532 chips I replaced were garbage.

So I'm afraid I'm adding my name to a now rather long list of people that noticed a difference for the better.

At least, though, I'm trying to determine a way to capture this difference.

It is entirely possible I'll give up and switch to another amp before that happens though.

Because you have two A07 amps you have a golden opportunity to blind test.
Ask a friend to randomly pop in the op amps in the amps.It can be different, it can be the same op amps, so double set of op amps is needed for your test.
Then only your friend knows which amp has which op amps. Do a number of blind tests. Same procedure again have your friend randomly pop in the op amps and you test again. If you can spot sound differences then you can, otherwise you are just imagining things.

Don't have your friend with you when you do your blind test because he or she can (unknowingly) signal to you which op amp is in which amp.

 

[cgallery]

Because you have two A07 amps you have a golden opportunity to blind test. Ask a friend to randomly pop in the op amps in the amps. Then only your friend knows which amp has which op amps. Do a number of blind tests. Same procedure again have your friend randomly pop in the op amps and you test again. If you can spot sound differences then you can, otherwise you are just imagining things.

Only one amp has the LM4562 chips. So the amps could just be switched from LF to HF use as an initial test. That spares us the PITA it is to swap the op amps in these things (there is little clearance).

If I can hear a difference swapping the entire amp (again, each amp having different op amps), then I can proceed to isolating only the op amps by swapping those. If I don't hear a difference, I'll abort there.

I'm thinking some gift wrapping ribbon installed under the op amp may make pulling them a little easier, if it gets to that.

 

[antcollinet]

Only one amp has the LM4562 chips. So the amps could just be switched from LF to HF use as an initial test. That spares us the PITA it is to swap the op amps in these things (there is little clearance).

If I can hear a difference swapping the entire amp (again, each amp having different op amps), then I can proceed to isolating only the op amps by swapping those. If I don't hear a difference, I'll abort there.

I'm thinking some gift wrapping ribbon installed under the op amp may make pulling them a little easier, if it gets to that.

Don't forget the controls - accurate level matching using a voltmeter - not knowing which amp is in use - repeated tests and getting a significant number correct (9/10 is typically required for significance, and restarting the test if you get a couple wrong is not valid)

These things are not that easy to do right.

 

[cgallery]

Don't forget the controls - accurate level matching using a voltmeter - not knowing which amp is in use - repeated tests and getting a significant number correct (9/10 is typically required for significance, and restarting the test if you get a couple wrong is not valid)

These things are not that easy to do right.

Well during phase one I'm just going to see if I can hear a difference between the amps w/ different op amps. Level matching will just involve keeping the volume at max as I set levels via the MiniDSP.

If I can make it through that round, then round two will be more disciplined.

 

[solderdude]

level matching won't help when you know what op-amp is in there.

 

[antcollinet]

Level matching will just involve keeping the volume at max as I set levels via the MiniDSP.

That might not be accurate enough - if component tolerance means there is as little as 0.2dB gain difference between the two amps.

 

[cgallery]

level matching won't help when you know what op-amp is in there.

I will have no idea which amp is being used for HF or LF duty.

 

[cgallery]

That might not be accurate enough - if component tolerance means there is as little as 0.2dB gain difference between the two amps.

Well again phase one only serves to determine whether I should proceed to the more rigorous phase II.

No sense in having to open and swap op amps if I can't pass phase one.

 

[antcollinet]

Well again phase one only serves to determine whether I should proceed to the more rigorous phase II.

No sense in having to open and swap op amps if I can't pass phase one.

True enough.

Personally though, I'd not bother. There is sufficient data to tell us that op amp rolling makes no audible difference unless they result in an unstable amp - which is definitely not an improvement.

 

[cgallery]

True enough.

Personally though, I'd not bother. There is sufficient data to tell us that op amp rolling makes no audible difference unless they result in an unstable amp - which is definitely not an improvement.

That is an interesting link however two issues. One, I'm suspicious of the stock op amps. Two, I'm pretty sure I've got a tell.

I also do wonder how often this has actually been tested. If op amps were in ZIP sockets, likely this would get tested more often. They (the ICs) can be a royal PITA to swap around.

 

[antcollinet]

If op amps were in ZIP sockets, likely this would get tested more often.

One of the reasons it is not, is that engineers, who understand the designs, and understand how op amps work in them, also understand that swapping these things into circuits make no difference that is audible.

It is like someone who doesn't understand mechanical engineering swapping the front wheel of a bike to one from a different manufacturer - but where the wheel size, weight, rotational inertia, tyre interfacing geometry (etc - all important characteristics) are identical to measurable tolerances. Then expecting to be able to cycle faster.

Then they claim it has worked because it "feels faster" even though the numbers on the cycle speed display have not changed.

It makes no sense.

 

[AnalogSteph]

If you're suspicious of the stock opamps (and I mean, there have been sightings of purported 5532s which were behaving more like 4558s), the better approach would be testing those on a protoboard out of circuit. Current consumption alone would be quite telling (a real 5532 should take 8-9 mA, a 4558 about 3.5 mA, a TL072 about 2.8 mA), but you could go all Samuel Groner on the thing as well.

 

[cgallery]

One of the reasons it is not, is that engineers, who understand the designs, and understand how op amps work in them, also understand that swapping these things into circuits make no difference that is audible.

It is like someone who doesn't understand mechanical engineering swapping the front wheel of a bike to one from a different manufacturer - but where the wheel size, weight, rotational inertia, tyre interfacing geometry (etc - all important characteristics) are identical to measurable tolerances. Then expecting to be able to cycle faster.

Then they claim it has worked because it "feels faster" even though the numbers on the cycle speed display have not changed.

It makes no sense.

There are engineers (one in this thread) that would apparently disagree to some extent. There are also those that eschew op amps for discrete even though at least some important performance metrics suffer.

We have a pretty decent protocol for phase one. Saturday is the day. I just hope the connectors (RCA and especially banana) are up to this, the connectors are a bit on the flimsy side.