Rolling (replacing) Op-amps in Topping D10 DAC

[gorg]

I’m using the optical input, not the analog inputs. Does it still matter about the 600mV?

If you using opt input in your edifiers, op-amp rolling in topping d10 is pointless. Check if another connections (coax, analog) distorting signal.

 

[andreasmaaan]

I’m using the optical input, not the analog inputs. Does it still matter about the 600mV?

I see! Had missed that. No, that's a totally different question then.

As @gorg says, if you're using the optical output from the Topping, none of the analogue stage (i.e. opamps) are involved. The problem is somewhere in the digital domain, or with the speakers themselves.

One thing I'd try would be to output via the RCA out from the Topping to the analogue inputs on the speakers. If that fixes the problem, you know it's something in the digital domain.

In any case, this is probably what I'd do if I were you anyway. The DAC in the Topping is likely to be far superior to that in the speakers, so it makes sense generally to do the DA conversion in the Topping and then send the speakers an analogue signal (unless the speakers do a subsequent AD/DA conversion from the analogue inputs, which is also possible).

Good luck sorting it out

 

[Jimmy]

Sorry, I didn't notice that you were using the D10 as a USB to toslink converter. As andreasmaan says it's likely that you'll get the best possible sound using the D10 rca outputs.

The distortion that you are hearing maybe is due to the way the D10 constructs the digital signal, for example I have an old DAC that badly distorts if I leave the output data format on auto, only when I set it to 24 bits on 32 bit packets it works as it should (SPDIF).

 

[HionHiFi]

Got it. Ok, I've changed the routing of the signal per your recommendations @andreasmaaan and @gorg. It is now: HP Laptop (USB, 32bit/192kHz) --> Topping D10 (analog) --> Edifier S2000 Pro ---| No distortion now. Thank you.

The Edifiers do, do ADC, but I suspect the perceptual loss of quality is minimal at best.

I see! Had missed that. No, that's a totally different question then.

As @gorg says, if you're using the optical output from the Topping, none of the analogue stage (i.e. opamps) are involved. The problem is somewhere in the digital domain, or with the speakers themselves.

One thing I'd try would be to output via the RCA out from the Topping to the analogue inputs on the speakers. If that fixes the problem, you know it's something in the digital domain.

In any case, this is probably what I'd do if I were you anyway. The DAC in the Topping is likely to be far superior to that in the speakers, so it makes sense generally to do the DA conversion in the Topping and then send the speakers an analogue signal (unless the speakers do a subsequent AD/DA conversion from the analogue inputs, which is also possible).

Good luck sorting it out

 

[andreasmaaan]

Got it. Ok, I've changed the routing of the signal per your recommendations @andreasmaaan and @gorg. It is now: HP Laptop (USB, 32bit/192kHz) --> Topping D10 (analog) --> Edifier S2000 Pro ---| No distortion now. Thank you.

The Edifiers do, do ADC, but I suspect the perceptual loss of quality is minimal at best.

Ok, great to know you've fixed the problem / isolated it to something in the digital domain.

No doubt there is a measurable loss of fidelity now due to the additional DA/AD conversion. As you you say, this is likely to not make a significant audible difference though.

One last thing you could try is the RCA/SPDIF from the Topping to the speakers perhaps?

 

[HionHiFi]

Right on. I’ll checkout the coax out into the speakers to see if that works. Kinda of sucks no that so know something may be wrong with the digital path in the DAC making it a bit less versatile. Which is part of the reason I bought it in the first place. Meh!!

Ok, great to know you've fixed the problem / isolated it to something in the digital domain.

No doubt there is a measurable loss of fidelity now due to the additional DA/AD conversion. As you you say, this is likely to not make a significant audible difference though.

One last thing you could try is the RCA/SPDIF from the Topping to the speakers perhaps?

 

[sssn]

Is that socketed? If so, yes.

Hey amir, do you still haven O2 to roll opamps with? Would be a nice change of pace after a slew of DAC measurements again

 

[amirm]

Hey amir, do you still haven O2 to roll opamps with? Would be a nice change of pace after a slew of DAC measurements again

I haven't sold anything so if I had it before, I still have it now. It will have to wait for some dead time later in the year.

 

[Eiffel]

Hi guys. I've read this forum for few months and now I saw this topic.
I've ordered yesterday a D10 to play with it, so in a week or two I will post my impressions in the other topic.

Now, I saw the measurements made while changing the buffer OP-AMP. I've played a lot with OP-AMP rolling in the last few years - I own an Asus STX ( first version ) and a Burson Play who have socketed OP-AMP's
It was something expected because the changeable OP-AMP is the buffer and act also as a LPF filter. And OPA2134 is probably one of the best for a LPF. Now, changing the buffer does not alter too much the overall sound. The I/V stage is the most important - this give about 90% of the sound.
The buffer/LPF just filter noise. It may alter at maximum 10% of the sound, and usually the soundstage is altered.

Now, why 2134 is probably the best here ?

Well, first it has a laid back sound. That compensate the sound of the ESS Dac, which is neutral with a tint of bright and dryness. So it would be the first choice for the I/V.

Second, OPA 2134 is just a 2132 who failed the frequency test at the factory, but it's still able to go up to 100kHz. That's what is TI "Audio" range OP-AMP - they are industrial OP-AMP's who are suitable also for audio and who cannot run at more than 100kHz. They are rebranded and sold for Audio.
That's what makes 2134 even better in a Low Pass Filter.

You may get slight better results, probably, using a 2132 ( from what I tested on the STX it may give a 0.5dB better SNR ), or a 2604.
A good buffer/LPF Op-AMP may be also Muses02, but this has power rejection issues - you must provide perfect clean power to do it's best.

I intend however to try the V6 from Burson in his place in D10 as soon as I get it.

The funny think with the measurements is that they are unable to say anything about how the sound "sounds". OP-AMPS with similar/identic measurements may give a different sound. Of course, I speak about OP-AMPS that can fit in the circuit, because those who don't, can be quite messy.

 

[Deleted member 2944]

Subjective evaluation is fine and dandy, but it doesn't have any definable reference so it's irrelevant to anybody but you. (There's no science in it.)
I'm not sure how you arrived at your 90/10 percentage priority, but I suppose that is subjective also?

I don't find the sound of the OPA2134 "laid back" at all....at least in the applications I've listened to.

Your post is highly speculative.

Dave.

 

[Thomas savage]

Put the screw driver down , just have a couple of beers .. everything will sound WAYYY better, hell folks even look more attractive and no op amp will do that for you.

 

[HeadphoneFan]

Put the screw driver down , just have a couple of beers .. folks look more attractive and no op amp with do that for you.

Which type of beer would you recommend?

 

[Jimster480]

Hi guys. I've read this forum for few months and now I saw this topic.
I've ordered yesterday a D10 to play with it, so in a week or two I will post my impressions in the other topic.

Now, I saw the measurements made while changing the buffer OP-AMP. I've played a lot with OP-AMP rolling in the last few years - I own an Asus STX ( first version ) and a Burson Play who have socketed OP-AMP's
It was something expected because the changeable OP-AMP is the buffer and act also as a LPF filter. And OPA2134 is probably one of the best for a LPF. Now, changing the buffer does not alter too much the overall sound. The I/V stage is the most important - this give about 90% of the sound.
The buffer/LPF just filter noise. It may alter at maximum 10% of the sound, and usually the soundstage is altered.

Now, why 2134 is probably the best here ?

Well, first it has a laid back sound. That compensate the sound of the ESS Dac, which is neutral with a tint of bright and dryness. So it would be the first choice for the I/V.

Second, OPA 2134 is just a 2132 who failed the frequency test at the factory, but it's still able to go up to 100kHz. That's what is TI "Audio" range OP-AMP - they are industrial OP-AMP's who are suitable also for audio and who cannot run at more than 100kHz. They are rebranded and sold for Audio.
That's what makes 2134 even better in a Low Pass Filter.

You may get slight better results, probably, using a 2132 ( from what I tested on the STX it may give a 0.5dB better SNR ), or a 2604.
A good buffer/LPF Op-AMP may be also Muses02, but this has power rejection issues - you must provide perfect clean power to do it's best.

I intend however to try the V6 from Burson in his place in D10 as soon as I get it.

The funny think with the measurements is that they are unable to say anything about how the sound "sounds". OP-AMPS with similar/identic measurements may give a different sound. Of course, I speak about OP-AMPS that can fit in the circuit, because those who don't, can be quite messy.

I didn't find this to be the case at all.
I tested a few different Op Amps in my A30 and only found that it makes the sound cleaner or dirtier based on slew rate.

 

[Wombat]

I didn't find this to be the case at all.
I tested a few different Op Amps in my A30 and only found that it makes the sound cleaner or dirtier based on slew rate.

Imagine a poker machine with different op-amps shown on the drums. That is what the unknowledgeable are playing.

 

[Wombat]

Which type of beer would you recommend?

It must have alcohol in it unless one is sensitive to suggestion.

 

[Eiffel]

Slew rate is important too - however if an OP-AMP is between 12-20 it should be stable for Audio and fast enough. Higher slew rate in some configurations may gives you oscillations.
Also very important is the OP-AMP input - what kind of input it has.

I'm not sure how you arrived at your 90/10 percentage priority, but I suppose that is subjective also?
I don't find the sound of the OPA2134 "laid back" at all....at least in the applications I've listened.

Yes it's subjective, and depending on OP-AMP's you use it may vary. More fair would be to say "approximative".
But yes, changing I/V does change a lot in the sound. Changing Buffer/LPF gives only very slight changes. Sometimes may give nothing at all.

OPA2134 has a laid back sound, quite warm - everybody on the internet can confirm this. Now it also depends on the DAC used. With the ESS it seems they compensate each other.

And an OP-AMP will react different depending on other components. Will not gives you the same sound in another DAC.
For example - the Muses01 in I/V on my STX sounds very natural and pleasant, slight warm sound but still quite neutral.
On Burson Play with them the sound is cold. Still neutral, but not very pleasant.

 

[Wombat]

Slew rate is important too - however if an OP-AMP is between 12-20 it should be stable for Audio and fast enough. Higher slew rate in some configurations may gives you oscillations.
Also very important is the OP-AMP input - what kind of input it has.


Yes it's subjective, and depending on OP-AMP's you use it may vary. More fair would be to say "approximative".
But yes, changing I/V does change a lot in the sound. Changing Buffer/LPF gives only very slight changes. Sometimes may give nothing at all.

OPA2134 has a laid back sound, quite warm - everybody on the internet can confirm this. Now it also depends on the DAC used. With the ESS it seems they compensate each other.

And an OP-AMP will react different depending on other components. Will not gives you the same sound in another DAC.
For example - the Muses01 in I/V on my STX sounds very natural and pleasant, slight warm sound but still quite neutral.
On Burson Play with them the sound is cold. Still neutral, but not very pleasant.

'everybody on the internet can confirm this'. You have to do better on ASR.

 

[Jimster480]

Slew rate is important too - however if an OP-AMP is between 12-20 it should be stable for Audio and fast enough. Higher slew rate in some configurations may gives you oscillations.
Also very important is the OP-AMP input - what kind of input it has.


Yes it's subjective, and depending on OP-AMP's you use it may vary. More fair would be to say "approximative".
But yes, changing I/V does change a lot in the sound. Changing Buffer/LPF gives only very slight changes. Sometimes may give nothing at all.

OPA2134 has a laid back sound, quite warm - everybody on the internet can confirm this. Now it also depends on the DAC used. With the ESS it seems they compensate each other.

And an OP-AMP will react different depending on other components. Will not gives you the same sound in another DAC.
For example - the Muses01 in I/V on my STX sounds very natural and pleasant, slight warm sound but still quite neutral.
On Burson Play with them the sound is cold. Still neutral, but not very pleasant.

Warm?
Its not at all warm, its perfectly suited for music, high bandwidth at that too.
It can do 16+ bits no problem.
I tried OPA2227, NE5532, LM4562, NJM4562.
I only found that the OPA2227 is dirtier.
The NE5532 is definitely much better, and I really tried hard to find differences past the NE5532.
I did find one part in one song with a specific background sound that was cleaner on the LM4562 vs the NE5532. Otherwise I couldn't tell a difference and this is with me going back and forth in specifically technical parts of high resolution test tracks literally listening for minute differences....
On the OPA2134 (the default Op Amp) I couldn't tell a difference between that an the LM4562.... although there was once again one part in a different track that I thought (subjective) sounded cleaner on the LM4562. So I kept the LM4562 installed.

Otherwise there was no actual differences, definitely not any of this "warm" or night and day type of claims you are making here...
Without me listening specifically for differences, I wouldn't have been able to tell any difference in the 3.
And if I had a device shipped with the OPA2227 and I hadn't listened to the same tracks with the OPA2134, its likely I wouldn't be the wiser to even know the difference there.. Although it is actually audibly flatter once you do notice it in comparison.

 

[Eiffel]

You tried them on a D10 ? In the buffer/LPF ?
Then it's normal what you said.
You cannot change the sound changing the buffer. You may get a better or worse background or a different soundstage.
NE5532 is similar with 2134 in sound. I think they were designed in the same period and were competitors.
LM4562 should have a bigger soundstage than 2134. But also will be slight noisier.
BTW - LM4562 is also LME49720 if someone would be interested to try.

 

[SchwarzeWolke]

In Science, those who claim something have to deliver the proof of it.
In this case double blind test