[Rroff]

LM4562 should not be running hot - it is a touch more susceptible to power supply noise than many other opamps but usually that manifests itself as a distinct grainy or robotic effect on the sound. Do you know if the power supply rails are bypassed to ground with 0.1uf or similar high speed capacitors near the opamp(s)? a common cause of opamps running hot is a poor connection somewhere - something I've encountered more than once when rolling opamps are weakened a socket or soldering.

Oh, I get now what you meant about “clinical.” Plan on get it some LM4562 but it’s funny, you know, I’m not sure... we especially older people who remember the ‘sound’of the analogue world prefer its imperfections and ‘musicality’ and reject the ‘modern’ sound of digital etc.

That is why I like the OPA2228 it bridges the gap between the two worlds.

adding 1nF film caps in parallel to feedback resistors and those between the inverted input and to the ground may fix it.

Bandwidth limiting capacitors can definitely be a help but ideally they should be calculated with the resistors, etc. to make sure they are actually doing something useful and far enough away from the audio frequencies to minimise any distortion.

 

[HeyManslowdown97]

That is why I like the OPA2228 it bridges the gap between the two worlds.

Make Perfect sense and VERY practical as a “stock” dap. I love the idea of contrasting the digital and analogue world competing in the hardware.

My guess is that the op amp in the circuitry in Aune A1 isn’t compatible to LM4562NA in this application or in the stage it’s being used (which is rare since it’s very universal and not fussy based on what I’ve read. Anytime you add something in electronics some other part is losing “love” so power supply could be the problem but thats a byproduct if it wasn’t a previous problem. You really have to know exactly what the cause and effect of op amp is doing to itself and what it needs. Its also important to remember that adding resistance lowers the voltage.

“Voltage varies directly with current. "R" is the constant of proportionality telling how much it varies. If I add in a resistor to a circuit, the voltage decreases. If you have a resistor in a circuit, with a current flowing through it, there will be a voltage dropped across the resistor (as given by Ohm's law)”

Depends on how forgiving daps are but generally I have found you can’t just op amp drop in some daps like nobodies business when it’s not designed to be developmental board but a fixed dap with very little leeway unless you know EXACTLY what you are doing and know how to hack the PCB with custom modding. In the end you gotta ask yourself is this worth it?

 

[raoultrifan]

@HeyManslowdown97, @Rroff I'm glad you are referring to the "unspoken" name of the O2 builder (BTW, your links and few words will probably get removed soon by an admin and you'll get warned).

There are basically YES and NO on both worlds: engineers that tell that two opamps having similar specs should sound the same, audiophooles that says that 100 USD opamps are sounding much better. Sometimes they are both right!

On the first crowd, the engineering world (objective people), they are probably right on most scenarios, but there could be so many scenarios when some opamps are simply sounding a bit differently, mostly because it all depends on the surrounding passive electronic parts and the PCB layout, so some opamps might oscillate a bit or simply pick-up some EMI/RFI noises and act differently in the audio chain. E.g.: pre or post ringing, but also a 0.5dB gain or loss of amplification will definitely change how the soundstage will be perceived.

On the 2nd crown, the audiophools perhaps (subjective people), that are swapping dozens of opamps and are getting a different soundprint after every change. Hmmm...I've heard people swapping LM4562 with LME49720 and telling the world that is a complete "game change", like a getting from Moon to the March. Given the fact that the above two opamps are completely identical in terms of specs and manufacturer design, I would definitely say that it's most likely due to psycho-acoustics, but it might be also related to the fact that volume match was not done by a scope or REW + microphone, or listener position has been changed (or headphones were moved around the ears with 0.5-1 inch).

What it matters in the end is how everyone of us is liking the output sound and, of course, the opamps we used to not oscillate. Ideally would be to have the setup measured somehow (RMAA, ARTA, REW etc.), just to be sure amplifier specs become better after swapping the opamps:
- THD (not higher than 0.001% if we really want the best out of our audio equipment)
- background noise (or THD+N for both items, which should be not higher than 0.01%, which represents only -80% of distortions + noise)
- DC output (less than 1mV for DACs and no more than 5mV to headamps, also 10mV should be fine for planars or power hungry cans)
- volume match (ideally less than 0.1dB between the channels, but usually less than 0.5dB is acceptable)

Some specs explanations could be found here: https://www.cambridgeaudio.com/en/blog/cd-specs.
A correlation between % and dB could be found here: http://www.sengpielaudio.com/calculator-thd.htm

What I'm trying to say is if the designer of the DAC chip, for example PCM1717, states a THD+N of 90db which translates in a THD+N of 0.003%, after choosing the correct decoupling caps and opamps I should not deviate much from DAC chip's specs. If an opamp will audio-orgasm my ears, but THD+N is getting worse with 10dB or even 20dB, then I should simply swap back the opamp till I'll get the best figure possible.

The effects of LowPassFilter to the output THD+N (source: http://www.ti.com/lit/an/sbaa055/sbaa055.pdf)

​

 

[CoiL]

LM4562 should not be running hot - it is a touch more susceptible to power supply noise than many other opamps but usually that manifests itself as a distinct grainy or robotic effect on the sound. Do you know if the power supply rails are bypassed to ground with 0.1uf or similar high speed capacitors near the opamp(s)? a common cause of opamps running hot is a poor connection somewhere - something I've encountered more than once when rolling opamps are weakened a socket or soldering.

First of all, I don`t hear any "grainy or robotic" sound and secondly, can it be what @Ivan TT said?

Most likely it is oscillating (and most likely in MHz range, so will not be audible).
Interesting side-effect of oscillating opamp is that it effectively biases it in class-A operation, giving intensity of sound (and power consumption/heat).

...because other opamps work ok and I think there are no issues with soldering, otherwise OPA2228 wouldn`t also work as it should I belive?

 

[HeyManslowdown97]

On the first crowd, the engineering world (objective people), they are probably right on most scenarios, but there could be so many scenarios when some opamps are simply sounding a bit differently, mostly because it all depends on the surrounding passive electronic parts and the PCB layout, so some opamps might oscillate a bit or simply pick-up some EMI/RFI noises and act differently in the audio chain. E.g.: pre or post ringing, but also a 0.5dB gain or loss of amplification will definitely change how the soundstage will be perceived.

I agree but again it comes down to knowledge of knowing both theoretical knowledge and applying it to application and you can’t know all variables. Being a hobbyist like the majority of DIYERS I cannot say with absolute certainty lacking a electronic engineering I know everything on these threads (nor can most who regurgitate falsity).

As for removing chosen “words” it’s annoying but that’s the type of censorship you get from digital tribalism. FYI: they don’t like the usage of “audiofools” in these parts either.

Lastly, when it comes to sound reproduction its an artifact so it’s a representation of an event NOT the event itself so it’s bothersome when people say things like it’s sounds “artificial” or “natural” (an opinion, not a fact) in objective terms. One mustn’t take these myths seriously.

There’s been a lot written about the “airy” myth being natural. Heck there are digital filters they do it (like the photo analogy I used)

I once had a guy tell me that he doesn’t use silver because it’s too “bright” and I pointed out that most (if not all) parts on a dedicated dac uses some silver alloy or its also used in soldering paste or soldering. He became indigent and stuck to his dogma.

As for the enthusiasm of op amp rolling I’ve been there, where I thought I have these Eureka moments which I’ve looked back on with cringe-worthy regrets.

A simple solution to adding gain to do so in post production with a limiter or with an app in real time on an actual track, no need to ticker with the hardware resulting in hurting yourself or your equipment as we don’t live in a 1996 world and like you said there are a variety of ways and solutions such as firmware, digital filters, software, apps, etc that will get you a desired sound or “soundstage to be perceived” other than op amps which have a finite in what they can do to a perceived soundstage.

In short even audio engineer today uses software, algorithms and don’t require actual physical space of say a mixing board only a Quality sound card with a op amp with good performance that functions. It’s when you start tinkering that you play with variates. If Op Amp Rolling requires fine tuning then it’s best to know what you are do or what the variables are.

The “unknown” author wasn’t deliberate I didn’t frequent in these threads until recently. I’m not an “insider” and not sure if I want to be part of a bubble... you are honest and it’s always a pleasure to hear your feedback

 

[CoiL]

But what happen when You put "objective engineering ppl" in blind test and ask which setup sounds "better" and more "natural" without telling him anything about specs of opamps.
I`m guessing we would have different results but if that person already has heard "electrically pure" system, then his psychoacoustic perception of sound has already "coded" to it "being better & truer". But what if that "engineer" somehow hasn`t heard such system he "believes in due to specs" and presented with different (or exactly same) setups? I bet we would get different results in that case too. To me, we all are "coded" and "coding" ourself in some way and when it comes to "natural sound reproduction" I put ppl into 2 sides - ppl who belive only facts, calculations, specs etc. and ppl who just grasp music with their senses and stay true to what they personally hear to be right for their own ears. I place myself somewhere inbetween.
To me is just illogical to think that gear with "purest technical specs and build" sounds same way "natural" to everybody! That is thought insane in itself! We are humans (biological and intellectual) not specs on paper.

 

[raoultrifan]

But what happen when You put "objective engineering ppl" in blind test and ask which setup sounds "better" and more "natural" without telling him anything about specs of opamps.[...]

Did that myself for few times; I remember very well when I had two identical ASUS Essence One MKii side-by-side and one had 4xMUSES01 in I/V and the other one had 4xNE5532 in I/V. So two people were 10 out of 10 inclined that the Essence One having the MUSES01 in I/V was having a bigger soundstage and somehow the sound was also more detailed (it was a blind test, we had no clue which one has the MUSES and which one the 5532). We were both right, but on the scope and RMAA and ARTA I realised the MUSES01 is causing a bit of post-ringing and also 2nd and 3rd harmonics are higher (about 20dB difference). In this particular case an worse THD and a bit of pre-ringing is causing the audio equipment to sound more...euphonic.

More details here: https://www.head-fi.org/threads/bur...sion-and-reviews.854912/page-18#post-14799047, but basically with MUSES01 I had a THD+N of 0.067% (-83.5dB) and with NE5532P I had a THD+N of 0.0024% (-92.4dB). So, in my case, NE5532P was able to provide a more accurate sound (very close to CD-Audio specs), but less euphonic than MUSES01 was able to offer.

 

[HeyManslowdown97]

But what happen when You put "objective engineering ppl" in blind test and ask which setup sounds "better" and more "natural" without telling him anything about specs of opamps.
I`m guessing we would have different results but if that person already has heard "electrically pure" system, then his psychoacoustic perception of sound has already "coded" to it "being better & truer". But what if that "engineer" somehow hasn`t heard such system he "believes in due to specs" and presented with different (or exactly same) setups? I bet we would get different results in that case too. To me, we all are "coded" and "coding" ourself in some way and when it comes to "natural sound reproduction" I put ppl into 2 sides - ppl who belive only facts, calculations, specs etc. and ppl who just grasp music with their senses and stay true to what they personally hear to be right for their own ears. I place myself somewhere inbetween.
To me is just illogical to think that gear with "purest technical specs and build" sounds same way "natural" to everybody! That is thought insane in itself! We are humans (biological and intellectual) not specs on paper.

Then you are asking their personal preferences as in subjectivity. That’s why datasheet only give performance results NOT what is a perceived sound quality. Rustling leaves in a movie aren’t “natural” they are added as effect during post. Cameras and microphones are placed in strategic locations to get the more intimate portrait to the viewership who subscribed to their favorite football clubs channel. Even stereo recordings are different then if you saw the performance live in their arrangements in what position you in, in the club or room, etc.

The airy sound is an effect so is the “warmer” sound etc. and is how the op amp is tuned; it’s a personal preference, nothing more, nothing less.

Op amp rolling is also used by guitar in FX Pedels that give you certain sounds preferences but again they are also being replaced by software that can get the same effects without changing ANY hardware!

 

[CoiL]

Btw, I totally forgot about V5i-D as I can also swap it against OPA2134. In output stage it didn`t work but maybe as dac differential it will?
Will report back soon.

Today tried out Burson V5i-D instead OPA2134 @ DAC differential opamp stage (+ OPA2228 @ output stage) - and it works!
Initial impressions from first moment was - very clean sound, like pitch-black clean. A smidge "borderline" highs but not sibilant or harsh.
Left it playing overnight and will do comparison against OPA1692IDR.

 

[nesty]

Today tried out Burson V5i-D instead OPA2134 @ DAC differential opamp stage (+ OPA2228 @ output stage) - and it works!
Initial impressions from first moment was - very clean sound, like pitch-black clean. A smidge "borderline" highs but not sibilant or harsh.
Left it playing overnight and will do comparison against OPA1692IDR.

@CoiL what dap are you using to experiment with those opamps? Thanks

 

[CoiL]

@CoiL what dap are you using to experiment with those opamps? Thanks

Just read littlebit backwards... no DAP being used, it`s my desktop gear - Aune T1. Stock DAC differential opamp is OPA2134 and stock headphone amp board opamp is NE5532.

 

[HeyManslowdown97]

More details here: https://www.head-fi.org/threads/bur...sion-and-reviews.854912/page-18#post-14799047, but basically with MUSES01 I had a THD+N of 0.067% (-83.5dB) and with NE5532P I had a THD+N of 0.0024% (-92.4dB). So, in my case, NE5532P was able to provide a more accurate sound (very close to CD-Audio specs), but less euphonic than MUSES01 was able to offer.

This! (My guess) was Most likely a byproduct of MUSES01’s better transistors and overall construction than how it performed in benchmarks which in turn allowed the sound card overall “health” to sound better.

This is why Op Amp Rolling isn’t an exact science and why many manufacturers have already gone through the painstaking task of task what sounds the best for their products. Moreover, why would they pay 46 dollars for a MUSES01 vs NE5532 to get increasing overhead cost when the results aren’t that night and day!

Those that do will often Jack up prices for the prosumer to make up the cost differential. In this case the sound card manufacturer assumes NO liability if the prosumer decides op amp roll and folk over the 46 dollars for the MUSE01.

This is why Evaluation boards are the way to go if you really want to see how a op amp would truly sound to full potential (in theory at least).

In reading the unknown maker of O2 he had some interesting things to say about the OPA627 being a “Supercar” like a Ferrari that be can ONLY be used really on a racetrack.

I’d STILL would want to test drive “Supercar” in its true element if given the chance but I have yet seen a DAP that’s specifically designed for the OPA627 or modded for it to hit all of its benchmarks and prevent it from oscillating!

 

[CoiL]

Left it playing overnight and will do comparison against OPA1692IDR.

In pairing with OPA2228PA (HO stage) - OPA1692IDR > V5i-D.
In this configuration (as DAC differential opamp) V5i-D has very slight mid-bass/lower-mids "gloom", which I`m very sensitive to and it ruined some HQ live recordings for me.
And I think the "pitch-black" clarity I mentioned before about V5i-D came actually from slight imaging "congestion" and slight lack or "air/space" as there are some ambient details subdued/lost in HQ live recordings.
Currently sticking with OPA1692IDR + OPA2228PA until my new OPA2134UA chip arrives for comparison.

 

[Ivan TT]

Currently sticking with OPA1692IDR + OPA2228PA until my new OPA2134UA chip arrives for comparison.

You really have to try more of opamps from AD stables.
TI are always and inevitably coloured (except for maybe 827 that I don't have but someone mentioned here as AD-sounding).
AD8620 is pretty much end game, class of its own (unless opamp is compensating for something else in the chain, of course).

 

[CoiL]

You really have to try more of opamps from AD stables.
TI are always and inevitably coloured (except for maybe 827 that I don't have but someone mentioned here as AD-sounding).
AD8620 is pretty much end game, class of its own (unless opamp is compensating for something else in the chain, of course).

Yeah, I have awsome result with ADA4841-2YRZ (DAC differential) + ADA4897-1ARZ (HO stage) in my modified DX50.

You meant AD8620 in DAC differential opamp stage or in HO stage?
Anyway, not sure if it is compatible since it requires almost double operating supply current and is rated max +-13V compared to OPA2134 +-18V ...if You meant dac stage.
Also it is 20€, which is quite a price for thing that I`m not sure about "improvement". I mean V5i-D isn`t also cheapest and 1) didn`t work in HO stage 2) ain`t providing me better SQ in dac stage also.

I really like how OPA2228PA sounds in HO stage and probably will stick with it. Now just need to figure out if OPA1692 provides me any improvement over stock OPA2134 in dac stage or not, latter I ordered new one because I cut off pins when removing from PCB. Maybe stock chip in dac stage turns out to be still slightly better and then my little curiosity wasn`t justified but anyway got some experience.

Currently I think most sound improvement came from HO stage opamp upgrade, which is weakest part of Aune T1. Maybe next step would be upgrading some resistors and SMD capacitors but not sure it provides my ears with noticeable improvement. I already have done primary and secondary stage power supply filtering capacitors upgrades, which gave clearly noticeable upgrade in SQ.