Op Amp Replaced: Fosi V3 Monoblocks LM4562NA/NOPB $8x3

[jmdesignz2]

The acoustical output of speakers is solely determined by the applied voltage (output signal of the amp).

The amp in question is load sensitive so when you would record the electrical output with the speakers connected it should be O.K.

The Solo is not good enough for this experiment when the goal is to prove the difference in electrical output.
It is good enough for the TPA3255 and to record 'changes in sound' but not good enough as a measurement device for the op-amps.

you don't see a problem in the amp voltage burning out the Solo line input? You don't see any value in the speaker interaction with the amp?

I agree with your last sentence

 

[jmdesignz2]

The acoustical output of speakers is solely determined by the applied voltage (output signal of the amp).

The amp in question is load sensitive so when you would record the electrical output with the speakers connected it should be O.K.

The Solo is not good enough for this experiment when the goal is to prove the difference in electrical output.
It is good enough for the TPA3255 and to record 'changes in sound' but not good enough as a measurement device for the op-amps.

"amp in question is load sensitive"

what do you mean by this? The V3 Mono is supposed to be fairly load independent per Amir's measurements. I thought that was one of the strengths of this amp and the PFFB

 

[solderdude]

you don't see a problem in the amp voltage burning out the Solo line input? You don't see any value in the speaker interaction with the amp?

I agree with your last sentence

No problem whatsoever as long as you don't turn up the volume too high !.
The +22dBU (9Vrms) limit is pretty high but to keep distortion to a minimum would recommend not to exceed 5V (which is already pretty loud).

That said... I would recommend this one:

Cosmos ADC | index

e1dashz.wixsite.com

You need to be very careful when trying to measure/record balanced amplifiers.

But would not buy it only to please the ASR crowd.

All you need to convince is yourself and you already done that. Such recordings are not going to help you make your point.

You are in the perfect position to blind test YOUR own perception but will need to have a helper and follow some rules. Takes a lot of time as well.
And then, in the end, you will have spent a lot of time only to find out you cannot fully trust your own hearing.

Some of 'us' have already gone through this sobering experience.

"amp in question is load sensitive"

what do you mean by this? The V3 Mono is supposed to be fairly load independent per Amir's measurements. I thought that was one of the strengths of this amp and the PFFB

That the amp is load sensitive, not as bad as the regular V3 because the mono is PFFB. The V3 is worse in this aspect as that one also is load dependent in the audible range but it still is load dependent.

What's impressive is the size/weight to power ratio and price/power ratio.
Perfectly fine amp to use b.t.w.

 

[jmdesignz2]

The acoustical output of speakers is solely determined by the applied voltage (output signal of the amp).

The amp in question is load sensitive so when you would record the electrical output with the speakers connected it should be O.K.

The Solo is not good enough for this experiment when the goal is to prove the difference in electrical output.
It is good enough for the TPA3255 and to record 'changes in sound' but not good enough as a measurement device for the op-amps.

I agree- the Solo is not good enough

The Scarlett Solo's SNR of ~110.5 dB is a limiting factor. The Fosi V3 mono block's SNR of ≥123dB is well above what the Scarlett Solo can accurately measure. The Scarlett Solo is not capable of capturing the full range of we are trying to measure.

The Scarlett Solo has a higher noise floor and more distortion than the amplifier we are trying to measure. When we use the Scarlett Solo to measure the Fosi V3, we are not measuring the performance of the amplifier or the op-amps.

 

[voodooless]

Fosi V3 mono block's SNR of ≥123dB is well above what the Scarlett Solo can accurately measure.

Nope

 

[Sokel]

Even if this doesn’t burn up the solo this how does this account for the speakers actual acoustic output?

It also introduces another device with its associated characteristics

Have you actually done this yourself?

There are a number of things to consider doing that.
First of all you have to know your gear.

If for example the amp you're using is a bridged design with balanced OUTPUTS, no ground connection is allowed to it, ever.
You have to make sure, no kidding, interface has to be isolated for example.

To protect the measuring interface just fix a nice voltage divider and your done.

 

[Salt]

Again, page #5, appreciate all yours efforts, but is it worth the time (priest?)... just questioning (all the given informations are widely accessible within this forum)

 

[solderdude]

Nope
View attachment 476581

The V3 mono is not the shown above V3

acc. to Fosi, measured by Amir.

SINAD: XLR: 101dB ; RCA: 93dB 100dB ; RCA: 93dB
SNR: ≥123dB 119dB
THD: ＜0.006% 0.002% (5W)
Rated Power Output: 48V/5A--240W@4Ω 243W@4ohm (1%)

For the price and size these are excellent numbers.
Will it get 'better' with an LM4562 instead of the NE5532 ? Nope as the performance is limited by the TPA3255, both LM4562 and NE5532 (in low gain) are both already superior to the TPA.

 

[jmdesignz2]

The V3 mono is not the shown above V3

acc. to Fosi, measured by Amir.

SINAD: XLR: 101dB ; RCA: 93dB 100dB ; RCA: 93dB
SNR: ≥123dB 119dB
THD: ＜0.006% 0.002% (5W)
Rated Power Output: 48V/5A--240W@4Ω 243W@4ohm (1%)

For the price and size these are excellent numbers.

appreciate the updated numbers

I am using the 46v 10a so perhaps marginally higher power output?

 

[jmdesignz2]

Nope
View attachment 476581

Wrong Amp though

this is the correct one- vastly better specs

 

[solderdude]

appreciate the updated numbers

I am using the 46v 10a so perhaps marginally higher power output?

No... this is the same as even with the 5A power supply (which has headroom for peaks up to 7A) is not reached so the output is voltage limited (31Vrms) and not current limited. The 48V is regulated.
With 2ohm loads there may be slightly more output voltage available with the 10A supply but not in 4-16 ohm loads.

The below numbers were reached using the stock 48V/5A (240W continuous) power supply in 8, 4 and 2 ohm

 

[Multicore]

You suggested “Connect the speaker signal to (Scarlett solo) line input”

Won’t this burn up the Scarlett solo?

Not if you are careful. Scarlett Solo spec is +22 dBu for full scale on the line inputs. Hence I assume it doesn't burn up at that Voltage. +22 dBu is 27.583 Volts peak-to-peak swing or 9.752 V RMS of a sine wave. On 4 Ohm load (speaker) that's about 14 Watts and 22 on 8 Ohm. For most speakers that's very, very loud. Hence the amp output Voltage is right in the range for that infertace.

So I think it's safe to start with the volume knob low and carefully wind it up until either the volume is as you like to listen or the meters on the interface approach clipping, whichever comes first.

The key insight to why it won’t burn up the Scarlett Solo is Ohm's law. The internal resistance behind the Scarlett Solo's line input is 60 kOhm, 4 orders of magnitude higher than that of the speakers. So when it's hooked up to the speaker in parallel about 4 order of magnitude less current flows through it than through the speaker.

 

[Multicore]

The Solo is not good enough for this experiment when the goal is to prove the difference in electrical output.
It is good enough for the TPA3255 and to record 'changes in sound' but not good enough as a measurement device for the op-amps.

That's an interesting thought. I think you're right but I was thinking along a different line. I hadn't been thinking about the op amps or anything else inside. I'd been thinking about how someone who can hear a difference between this and that through their loudspeakers might set about measuring. For that purpose I had assumed that the budget line-level converters such as Solo are able, with appropriate input signal level, to convert with more accuracy than human hearing. It should therefore work for this purpose. But I grant that for a another purpose, to capture the difference in the product's performance with different op amps, as Amir set out to do, you're probably right.

 

[solderdude]

I agree with that sentiment and it is obvious that if it is audible it is recordable and already very measurable (in at least one or 2 basic measurements).
However, this would not satisfy OP. In fact nothing will convince OP the perceived differences is just human nature instead of electrical origin.
I suspect that even when he starts to test blind properly and statistically valid and (what I expect) the perceived differences would not be provable this way this won't convince OP.
Even if he got Amir as far as to redo all measurements on the V3 mono with LM4562 and NE5532 and it would show no difference (which it will as in practice the LM4562 and NE5532 in such a circuit are equivalent instead of 'much better' OP will still not be convinced.
Simply because he perceives it and many others do too so it cannot possibly be anything other than real. The rationale will be: you can't measure everything with some basic measurements and the ears are much more discriminative.

When differences would be audible then the solo would have recorded that. There would be some noise added and distortion.
But .... as the specs are 'worse' than that of the (op)amp it would not convince OP.
I reckon OP (as well as many, many others) believe the 'subtleties' are very small. In reality they are somewhere in the -30dB to -60dB range so recordable.
And because well measuring gear all differs in the 'low level performance' gear that shows 'improvements' in the -100 to -130dB range thus may be the reason for the difference in sound they perceive while measuring fine in other aspects.

The problem here is that these people never experiment with sudden attenuations or mix 'signals' of -70 to -100dB with music only to find out they can't hear any of that.

So even when OP bought TOTL ADC and records the differences and were not able to AB those files it would be nagging in his mind... what if the ADC does not use LM4562 ?
What if there are NE5532 in there that 'filters' out the fine nuances ?

Its a no-win situation. OP needs to learn to test with the least possible bias. There is no fun in that.
It is time consuming, you need a helper, it is extremely tiring as it requires a lot of attention, the differences in actual output of the amp (and thus the speaker) are non-existent.
Also the results don't match with any subjective findings (as there aren't any actual differences) which is very disappointing.
People who actually done this kind of testing either see the light and realize one should use the ears to enjoy music or, out of frustration, revert back to subjective listening as they can hear things they can't objectively.

 

[jmdesignz2]

I agree with that sentiment and it is obvious that if it is audible it is recordable and already very measurable (in at least one or 2 basic measurements).
However, this would not satisfy OP. In fact nothing will convince OP the perceived differences is just human nature instead of electrical origin.
I suspect that even when he starts to test blind properly and statistically valid and (what I expect) the perceived differences would not be provable this way this won't convince OP.
Even if he got Amir as far as to redo all measurements on the V3 mono with LM4562 and NE5532 and it would show no difference (which it will as in practice the LM4562 and NE5532 in such a circuit are equivalent instead of 'much better' OP will still not be convinced.
Simply because he perceives it and many others do too so it cannot possibly be anything other than real. The rationale will be: you can't measure everything with some basic measurements and the ears are much more discriminative.

When differences would be audible then the solo would have recorded that. There would be some noise added and distortion.
But .... as the specs are 'worse' than that of the (op)amp it would not convince OP.
I reckon OP (as well as many, many others) believe the 'subtleties' are very small. In reality they are somewhere in the -30dB to -60dB range so recordable.
And because well measuring gear all differs in the 'low level performance' gear that shows 'improvements' in the -100 to -130dB range thus may be the reason for the difference in sound they perceive while measuring fine in other aspects.

The problem here is that these people never experiment with sudden attenuations or mix 'signals' of -70 to -100dB with music only to find out they can't hear any of that.

So even when OP bought TOTL ADC and records the differences and were not able to AB those files it would be nagging in his mind... what if the ADC does not use LM4562 ?
What if there are NE5532 in there that 'filters' out the fine nuances ?

Its a no-win situation. OP needs to learn to test with the least possible bias. There is no fun in that.

You bring up some valid points

But you don't seem to be seeking a genuine debate. You are using a rhetorical strategy of preemptive refutation to close off any possibility of a productive discussion by delegitimizing my position from the outset.

You propose to know what I am thinking and will think in the future. You assert that nothing will convince the me, thereby invalidating any future points I might make
Paradoxically you suggest that the burden of proof is on me, while simultaneously arguing that any evidence I might find (e.g., proper blind testing, professional measurements) would be insufficient to convince me.

I appreciate your technical expertise but would like to request you refrain from those types of debate as it's a bit unfair

What I will restate is - my experience was subjective and I expended approx $30 inclusive of shipping and tax to experiment with this by replacing my op amps with technically provably superior/ certified parts.
Which I think was most likely to produce a better result than expending over $120-$300 on niche parts with questionable data sheets

Scientifically, I remain curious instead of simply / dogmatically shutting the door on any improvements. Alas, I don't have the resources, time or patience to investigate this further.

I will be happy knowing I have done no damage to my equipment and replaced the stock with provably superior parts at (AFAIK) the best price/performance ratio

 

[voodooless]

what if the ADC does not use LM4562 ?

What if the mastering equipment used TL072

 

[solderdude]

You bring up some valid points

But you don't seem to be seeking a genuine debate.

There is nothing to debate as there is no middle ground.
There is perception and electronics (incl. measurements).
There is some relation of course otherwise there would not be a point in measuring.
The problem is audibility thresholds and functionality of the brain which is highly variable and inconsistent as well as inaccurate.
The brain 'analyzes' in a totally different way than measurements.

You are using a rhetorical strategy of preemptive refutation to close off any possibility of a productive discussion by delegitimizing my position from the outset.

Correct... consider I have been where you are now ... but about 30 years ago.

Paradoxically you suggest that the burden of proof is on me, while simultaneously arguing that any evidence I might find (e.g., proper blind testing, professional measurements) would be insufficient to convince me.

The one making the claim bares the burden of proof.
Evidence of signal fidelity has been posted numerous times.

There is a discrepancy between perception and signal fidelity, there always has been and always will be. The reason is the human brain and all the input it gets where measurements only look at an electrical signal that changes in amplitude over time. Both amplitude and time can be measured with accuracy extremely far exceeding human hearing.

My advise to you is ... do some testing yourself but in a different way.
It is tiring, difficult to do, requires controls and a helper that is willing to assist during the entire testing. That might take a few days due to the virtually non existent actual differences.
The main part of this test method is removal of the 'knowing what is playing' part and avoiding 'tells'.

Amir redoing measurements with LM4562 won't convince you simply because there will be NO differences between these op-amps which does not jive with your perception.

I appreciate your technical expertise but would like to request you refrain from those types of debate as it's a bit unfair

It might seem unfair but ... it is the actual crux of the matter.
The differences you perceive simply can not come from an actual different wave-form from the amplifier.
It shows you are human.
Use your ears to enjoy music. If changing op-amps helps you (for whatever reason) then that is fine, a bonus ...
There will be no confirmation from the electrical side that changing the op-amps in this particular design actually changes the output.

 

[MAB]

But you don't seem to be seeking a genuine debate. You are using a rhetorical strategy of preemptive refutation to close off any possibility of a productive discussion by delegitimizing my position from the outset.

You don't seem to be seeking genuine information.
I provided you with the actual performance of the LM4562, and you either didn't understand, or just wanted to debate rather than understand. Debating audio isn't as fun as you seem to make it out to be. You prefer to spend time arguing and debating. Instead, you seem to misunderstand what is going on with your electronics, and clearly misunderstand what people have explained to you, as demonstrated in the rest of your message:

You propose to know what I am thinking and will think in the future. You assert that nothing will convince the me, thereby invalidating any future points I might make
Paradoxically you suggest that the burden of proof is on me, while simultaneously arguing that any evidence I might find (e.g., proper blind testing, professional measurements) would be insufficient to convince me.

I appreciate your technical expertise but would like to request you refrain from those types of debate as it's a bit unfair

What I will restate is - my experience was subjective and I expended approx $30 inclusive of shipping and tax to experiment with this by replacing my op amps with technically provably superior/ certified parts.
Which I think was most likely to produce a better result than expending over $120-$300 on niche parts with questionable data sheets

Scientifically, I remain curious instead of simply / dogmatically shutting the door on any improvements. Alas, I don't have the resources, time or patience to investigate this further.

Not really. You have performed one of the most pointless things in audio, rolling OpAmps, and came here to have a debate about it. Fallen in to the same trap hordes of audiophiles fall into, yet here you are at ASR where you could actually find out what changed, and how you might determine difference. Odd you don't have time or resources or patience to learn, yet you waste so much of it debating.

 

[Salt]

And if you feed the Troll, you feed the Troll.
We are at page #5 for no show of perception ...

 

[dcolak]

Why not record the speakers playing the song once with old OpAmp once with new OpAmp and compare the two, see if there's a difference.

I bet there is