UPGRADE: FOSI V3 MonoBlocks OpAmp Rolling That Actually Might Work?

[joeg26er]

Evidently, there is a Texas Instruments version of the opamp used in the Fosi Mono V3 amps that actually has scientifically better specs??
but it does cost over 89X the original ones

Is it worth trying? If I buy 10 I am risking $81 including shipping...I will need two in each amp

According to AI
Summary: LM4562NA vs NE5532P in the Fosi V3 Mono

Summary: LM4562NA vs NE5532P in the Fosi V3 Mono​

• Lower Noise
  • NE5532P: 5 nV/√Hz
  • LM4562NA: 2.7 nV/√Hz
  • Approximate improvement: 5.3 dB lower input voltage noise
    Result: Cleaner background, reduced audible hiss in sensitive or high-gain systems.
• Faster Transient Response
  • NE5532P: 9 V/µs slew rate
  • LM4562NA: 20 V/µs slew rate
    Result: Sharper transient reproduction, improved clarity and detail retrieval.
• Higher Bandwidth
  • NE5532P: 10 MHz gain-bandwidth product
  • LM4562NA: 55 MHz gain-bandwidth product
    Result: Better high-frequency extension and more linear performance across the audio band.
• Lower Harmonic Distortion
  • NE5532P: 0.0005% THD+N (typical)
  • LM4562NA: 0.00003% THD+N (typical)
  • Approximate improvement: 24 dB lower distortion
    Result: Greater transparency, reduced coloration, and improved stereo separation.
• Thermal Characteristics
  • NE5532P: 97 °C/W (PDIP)
  • LM4562NA: 102 °C/W (PDIP)
    Result: Thermal performance is effectively equivalent; both are safe for use in the Fosi V3 Mono without added cooling.
• Physical and Electrical Compatibility
  • Package: Both are DIP-8
  • Pinout: Identical
  • Supply Range: ±2.5 V to ±17 V for both
    Result: LM4562NA is a direct, drop-in replacement with no modifications required.

---

Conclusion​

The LM4562NA can be installed directly into the existing DIP-8 sockets without conflict or additional modification. It is a scientifically justified upgrade for applications where improved resolution, lower noise, and greater fidelity are desired.

 

[jmdesignz2]

I ordered the ten of the LM4562NA - should be here tomorrow. I'll have a couple extra if anyone is interested to test these.

 

[DVDdoug]

that actually has scientifically better specs??

It's NOT going to improve the sound. It might not even improve the amplifier's measurements since the op-amp is one of the least critical parts of the amplifier circuit.

The "magic" of op-amps is that they have super-high open-loop gain. Negative feedback ("corrective feedback") brings the gain down to something useable and it improves the performance (lower noise, lower distortion, and flatter frequency response).

In higher gain applications like microphone preamps or phono preamps, lower noise can be an advantage (if the noise is audible to begin with) and there is still lots of feedback to make everything else "perfect" (audibly perfect).

...And there's always SOME risk opening something up and messing with it.

 

[TurtlePaul]

It is not going to do much of anything measurable. The TPA3255 chip and the output filter are the main distortion generators. The input opamps have no control over those.

 

[MAB]

It is not going to do much of anything measurable. The TPA3255 chip and the output filter are the main distortion generators. The input opamps have no control over those.

Yeah, agreed. It's a nothing-burger, line-on-line the same as NE5532, OPA213x, etc.

 

[JSmith]

Is it worth trying?

No...

'Op-amp Rolling Using Sparkos on Fosi V3 Mono'

Apr 2, 2025

This is a study of measured differences if one swaps up the TI NE5532p dual IC op-amp in a Fosi V3 Mono amplifier with a "discrete" opamp from Sparkos. It is a test requested in the youtube comments of my video test of the same using Douk A5 and Sonic Imagery op-amp. The Sparkos SS3602 retails for US $79.80 but I found a discount coupon that brought the cost down to $74.33 with shipping included.

Per Fosi documentation, there are three op-amps. The one at the bottom is only for unbalanced/balanced conversion. That leaves two op-amps to swap. Company says if you...

• amirm
• Replies: 348
• Forum: Stereo and Multichannel Amplifier Reviews

• 

'Sparkos SS3602 Opamp Rolling In Fosi P4'

Apr 3, 2025

This is a detailed measurements of effect of swapping the stock NE5532P op-amp in Fosi P4 preamplifier with Sparkos SS3602 "discrete" opamp. The SS3602 costs US $79.80 and the NE5532P, US $0.47.

Swapping opamps in P4 is more challenging due to back connectors/switches that need to be disconnected. I am not sure how successful I will be putting it back together as some of the tiny screws are almost stripped now. I looked and there was no obvious mention of swapping opamps in P4 from Fosi so I suggest not trying it regardless of what you are about to see.

So far, we have...

• amirm
• Replies: 347
• Forum: Stereo and Multichannel Amplifier Reviews

• 

'Does Op-amp Rolling Work?'

Mar 16, 2025

When I post my last video on review of Douk A5 amplifier, I mentioned that rolling (changing) op-amp ICs in there is fruitless. I got have a dozen comments under that video in youtube asking why so I thought I experiment again with the A5. Note that I have done the same testing with both DACs and Power Amplifiers and found the change to not make a difference. But let's see if the results are different this time.

The Douk A5 makes this job both easy and hard. It is easy because you can just lift the plexiglass and replace the op-amps. The difficulty was...

• amirm
• Replies: 824
• Forum: Stereo and Multichannel Amplifier Reviews

• 

JSmith

 

[solderdude]

Evidently, there is a Texas Instruments version of the opamp used in the Fosi Mono V3 amps that actually has scientifically better specs??
but it does cost over 89X the original ones

Is it worth trying? If I buy 10 I am risking $81 including shipping...I will need two in each amp

According to AI
Summary: LM4562NA vs NE5532P in the Fosi V3 Mono

Summary: LM4562NA vs NE5532P in the Fosi V3 Mono​

• Lower Noise
  • NE5532P: 5 nV/√Hz
  • LM4562NA: 2.7 nV/√Hz
  • Approximate improvement: 5.3 dB lower input voltage noise
    Result: Cleaner background, reduced audible hiss in sensitive or high-gain systems.

Alas... the noise of the TPA3255 is dominant by a factor 100.

• Faster Transient Response
  • NE5532P: 9 V/µs slew rate
  • LM4562NA: 20 V/µs slew rate
    Result: Sharper transient reproduction, improved clarity and detail retrieval.

Alas the transient reproduction is lowered considerably by the TPA3255 that comes after the op-amp and above all its output filter far below that of the 'worst' op-amp here. So the slew-rate of the used op-amp is of no importance here.

• Higher Bandwidth
  • NE5532P: 10 MHz gain-bandwidth product
  • LM4562NA: 55 MHz gain-bandwidth product
    Result: Better high-frequency extension and more linear performance across the audio band.

Alas the frequency extension is limited by the TPA3255 and above all its (load dependent) output filter to 40kHz (-3dB) which is about 30x lower than what the 5532 will do in a circuit with a few x gain. So ... again ... the GBW of the used op-amp is of no importance at all in this amp.

• Lower Harmonic Distortion
  • NE5532P: 0.0005% THD+N (typical)
  • LM4562NA: 0.00003% THD+N (typical)
  • Approximate improvement: 24 dB lower distortion
    Result: Greater transparency, reduced coloration, and improved stereo separation.

That's great ... the TPA3255 is 0.006% and increases to 0.01% near the max. output. Soooo... the op-amps being 'magnitudes better' is of no consequence in this particular amp.

• Thermal Characteristics
  • NE5532P: 97 °C/W (PDIP)
  • LM4562NA: 102 °C/W (PDIP)
    Result: Thermal performance is effectively equivalent; both are safe for use in the Fosi V3 Mono without added cooling.

Yep, the output filters will be heating up the amp internally much more than the op-amps. The thermal resistance of the op-amps is of no importance.

• Physical and Electrical Compatibility
  • Package: Both are DIP-8

The TPA3255 = HTSSOP (44 pin) just FYI

• • Pinout: Identical
  • Supply Range: ±2.5 V to ±17 V for both
    Result: LM4562NA is a direct, drop-in replacement with no modifications required.

That's correct.

---

Conclusion​

The LM4562NA can be installed directly into the existing DIP-8 sockets without conflict or additional modification. It is a scientifically justified upgrade for applications where improved resolution, lower noise, and greater fidelity are desired.

This conclusion is correct. The LM4562 can be installed directly into the DIP-8 sockets without conflict or additional modification.
In this particular amp, however, improved resolution, lower noise, and greater fidelity are not achieved because of... you might have guessed it ... the TPA3255 performance that comes after the op-amp.

Is it useful in this amp ? No, totally not.

Should you 'invest' in buying and trying op-amps ? No, you can use the money and above all time for much more useful things.

 

[DualTriode]

Alas... the noise of the TPA3255 is dominant by a factor 100.

Alas the transient reproduction is lowered considerably by the TPA3255 that comes after the op-amp and above all its output filter far below that of the 'worst' op-amp here. So the slew-rate of the used op-amp is of no importance here.

Alas the frequency extension is limited by the TPA3255 and above all its (load dependent) output filter to 40kHz (-3dB) which is about 30x lower than what the 5532 will do in a circuit with a few x gain. So ... again ... the GBW of the used op-amp is of no importance at all in this amp.

That's great ... the TPA3255 is 0.006% and increases to 0.01% near the max. output. Soooo... the op-amps being 'magnitudes better' is of no consequence in this particular amp.


Yep, the output filters will be heating up the amp internally much more than the op-amps. The thermal resistance of the op-amps is of no importance.


The TPA3255 = HTSSOP (44 pin) just FYI


This conclusion is correct. The LM4562 can be installed directly into the DIP-8 sockets without conflict or additional modification.
In this particular amp, however, improved resolution, lower noise, and greater fidelity are not achieved because of... you might have guessed it ... the TPA3255 performance that comes after the op-amp.

Is it useful in this amp ? No, totally not.

Should you 'invest' in buying and trying op-amps ? No, you can use the money and above all time for much more useful things.

Complete agreement with solderdude.

Solderdude has it right.

The difference is not so much measured distortion. To me the most likely most notable measured difference is noise. By the time the program material comes out of the speakers who knows what the weakest link of the chain is? The refrigerator compressor, the, the Phono RIAA pre-amplifier, the pre-amplifier, power-amplifier. If you can hear distortion something is broken or it is speaker caused distortion.

Where I care about the noise the most is what comes out of the speaker into the room between music tracks when everything should be silent.

Power line hum and buzz is the worst.

Hiss out of the mid and tweeter is next on my list.

The last place to place fault is the op-amp.

The exception is tube amps, in that world 5% distortion is Hi-FI, sometimes on purpose with feeling.

Thanks DT

 

[jmdesignz2]

Complete agreement with solderdude.

Solderdude has it right.

The difference is not so much measured distortion. To me the most likely most notable measured difference is noise. By the time the program material comes out of the speakers who knows what the weakest link of the chain is? The refrigerator compressor, the, the Phono RIAA pre-amplifier, the pre-amplifier, power-amplifier. If you can hear distortion something is broken or it is speaker caused distortion.

Where I care about the noise the most is what comes out of the speaker into the room between music tracks when everything should be silent.

Power line hum and buzz is the worst.

Hiss out of the mid and tweeter is next on my list.

The last place to place fault is the op-amp.

The exception is tube amps, in that world 5% distortion is Hi-FI, sometimes on purpose with feeling.

Thanks DT

What about increasing the 2200mF capacitors to double the value or more and adding a small network of bypass 10mF, .10, .01 capacitors to further eliminate any of the high frequency noise from the chain? Won’t that have a larger impact on performance ?better dynamics, transients , Lower IMD?

 

[solderdude]

It uses a switched mode power supply.
That 'replenishes' the DC faster than the 100Hz pulses of a linear power supply.
This means 2200uF of 10mF won't make any difference.

The 'noise' from SMPS that can be problematic is not solved by adding decoupling caps... as that noise is common mode and not differential mode.
Besides .... decoupling caps should not be near the reservoir caps but at the pins of the switching components. That's where they are effective not in parallel to reservoir caps.
They are also utterly pointless in parallel to coupling caps.

Also... most SMPS don't like to see a high capacitance without additional inductors in the DC path.

 

[jmdesignz2]

Also... most SMPS don't like to see a high capacitance without additional inductors in the DC path."

Fact-check: True.
Adding excessive capacitance to the output of an SMPS can cause problems.

• Start-up inrush current: Large capacitors draw a significant inrush current when the power supply is first turned on. Some SMPS designs may trigger an overcurrent protection (OCP) event and shut down if this current is too high.
• Stability issues: The SMPS control loop is designed for a specific output impedance. Adding a large external capacitance changes the dynamics of the feedback loop, potentially introducing a phase shift that causes the SMPS to become unstable and oscillate.
• Solution: As the statement alludes to, additional components like inductors can be added to create a filter that isolates the SMPS from the bulk capacitance, ensuring stable operation. However, this is a consideration for the overall system design, not an argument against using adequate bulk capacitance with the TPA3255.

 

[jmdesignz2]

Is the following correct?

.

1. SMPS ripple: An SMPS creates a stable DC output by rapidly switching its internal power components. The output filter capacitors smooth this switching to produce the DC voltage, but they cannot eliminate all of it. A small, high-frequency voltage ripple (differential-mode noise) remains on the output voltage.
2. Load-dependent noise: As a load draws current, it causes a voltage drop across the power supply's output resistance and inductance. This effect is especially pronounced during transient events, such as when the amplifier needs to supply a sudden burst of current for an audio peak. This creates more differential-mode voltage noise that is directly related to the current drawn.

 

[solderdude]

1: Yes, this is caused by feedback principle.
Thankfully the circuit that is fed by a power supply has a certain PSRR (PSSR) so little of that enters the signal path.
Some designs have a low PSRR where others could have a PSRR well over 100dB.
Of course when a design has a low PSRR the power supply must be as 'clean' as possible.

2: yes, dynamic loads are always a challenge both for SMPS as well as linear regulated and unregulated power supplies.
Thankfully the circuit that is fed by a power supply has a PSRR so little of that enters the signal.

I could not find any data on PSRR for TPA3255 though but stands to reason it is high.

A series inductor followed by a capacitor will help with continuous loads in lowering differential 'noise' but when relative high dynamic loads are present the voltage will still sag a little on high currents.

 

[DualTriode]

What about increasing the 2200mF capacitors to double the value or more and adding a small network of bypass 10mF, .10, .01 capacitors to further eliminate any of the high frequency noise from the chain? Won’t that have a larger impact on performance ?better dynamics, transients , Lower IMD?

From the op-amp point of view the switching power supply noise is of small importance, the op-amp PSRR will attenuate the power supply noise. In the world of op-amps the primary sources of noise are voltage noise, current noise and resistor noise from the related support circuit. There are good models, do the math.
DT

 

[NTK]

I could not find any data on PSRR for TPA3255 though but stands to reason it is high.

 

[solderdude]

On page 1 even... completely glanced over it.

This means the voltage ripple (no load, no signal) should be <150mV (A-weighted) = 400mVpp (A-weighted)
A 48V/5A Meanwell GT280A/360A has 200mVpp ripple. (not all SMPS are created equal).

Load regulation = +/- 2% = +/-1V for 48V power supply (so 2Vpp = 0.7Vrms non weighted) = 0.4mV will make it into the signal with a transient near max output power (30Vrms).
Below any audible threshold I would say because of masking this only happens when music is playing loud.

For those that wonder about the contribution of the ''sound quality determining' op-amp in this aspect...
The NE5532 opamp the SVR (=PSRR) is about 40dB better (100x lower in PSRR) and on top of that is fed via a regulator so is not a contributing factor at all.
No need to worry about the 'power supply quality' for the op-amps.

 

[jmdesignz2]

From the op-amp point of view the switching power supply noise is of small importance, the op-amp PSRR will attenuate the power supply noise. In the world of op-amps the primary sources of noise are voltage noise, current noise and resistor noise from the related support circuit. There are good models, do the math.
DT

Adding more capacitors was exclusively for the benefit of the op amps. It was for the overall performance of the amp

of course it’s not practical to use enough capacitance to result in zero voltage sag. However less sag, less frequently should mean better sound

 

[solderdude]

Adding more capacitors was exclusively for the benefit of the op amps. It was for the overall performance of the amp.

It won't do anything for the op-amps ... simply because of PSSR/PSRR/SVR which means ripple on the power supply op-amp simply does not make it in the audio path. They are an effective isolator for power rails issues. Besides... due to the limited max. supply rail voltage in power amps they will always be fed through some kind of regulator which also lowers/removes power supply 'ripples/noise' in power amps.

of course it’s not practical to use enough capacitance to result in zero voltage sag. However less sag, less frequently should mean better sound

For power amps that sag only becomes an 'issue' when the output stage is clipping and the output device(s) of the actually are connected directly to the voltage rail.
As most people never use amps that way (clipping level) it is never an issue.
So no ... more capacitance does not result in 'better sound'.

I know subjectively this helps, just like making compound capacitors, even with coupling caps but that is not resulting in better measured performance nor in signal fidelity but has another reason entirely

 

[jmdesignz2]

It won't do anything for the op-amps ... simply because of PSSR/PSRR/SVR which means ripple on the power supply op-amp simply does not make it in the audio path. They are an effective isolator for power rails issues. Besides... due to the limited max. supply rail voltage in power amps they will always be fed through some kind of regulator which also lowers/removes power supply 'ripples/noise' in power amps.

For power amps that sag only becomes an 'issue' when the output stage is clipping and the output device(s) of the actually are connected directly to the voltage rail.
As most people never use amps that way (clipping level) it is never an issue.
So no ... more capacitance does not result in 'better sound'.

I know subjectively this helps, just like making compound capacitors, even with coupling caps but that is not resulting in better measured performance nor in signal fidelity but has another reason entirely

You’re likely right about the op amps subject

But you are wrong about extra capacitance having no effect on the measurement or signal or amps would have even smaller capacitors.

There’s absolutely a benefit in moderate listening and definitely for transient peaks especially for example Als Sprach Zarathustra can definitely drive this amp well into clipping if the listening position is at typical distance 10-12ft away with a 86dB speaker that dips into 2-3ohm territory

 

[jmdesignz2]

Also. Improving the slew rate of the op amp will help with especially transients because that’s the bottleneck in the performance of the amp

It’s A "garbage in, garbage out" scenario: A Class-D amplifier like the TPA3255 can only amplify the signal it is fed. It doesn't matter that the TPA3255 has a very high switching speed; it cannot "fix" a distorted signal that it receives from a slower op amp.