Topping PA5 II Stereo Amplifier Review

[onethreetwowaitthatswrong]

So let's say one would like to replace his old and big Teac amp and its faulty DAC that sits between a DX3Pro+ and 90 dB sensibility bookshelves speakers in order to receive and transmit digital flux in a small room.
Doesn't the PA5 II seems like the perfect solution ?
B-b-but there seems to be so many issues with them. So if one was to aim for the best scoring gear thats been measured around here, wouldn't one be more reassured, with a similar result, if he replaced it with a power amp like the Audiophonics MPA-S125NC, little bro of the 250NC tested here and using the DX3+ as preamp ? Or remove the topping as well in favor of a all in one solution like the SMSL AL200 ? Or decide that money should be kept and be on his way with the sufficient (?) power of a simple Fosi Audio V3 ?

 

[Power Pop 23]

So let's say one would like to replace his old and big Teac amp and its faulty DAC that sits between a DX3Pro+ and 90 dB sensibility bookshelves speakers in order to receive and transmit digital flux in a small room.
Doesn't the PA5 II seems like the perfect solution ?
B-b-but there seems to be so many issues with them. So if one was to aim for the best scoring gear thats been measured around here, wouldn't one be more reassured, with a similar result, if he replaced it with a power amp like the Audiophonics MPA-S125NC, little bro of the 250NC tested here and using the DX3+ as preamp ? Or remove the topping as well in favor of a all in one solution like the SMSL AL200 ? Or decide that money should be kept and be on his way with the sufficient (?) power of a simple Fosi Audio V3 ?

I vote for the Fosi Audio V3 - it may be all you need. Be advised Fosi is developing a similar amplifier with unbalanced RCA inputs and differential (balanced) inputs. It may not cost much more than a V3 and will allow the option of amplifying balanced gear - in case you run into hum from an AC power ground loop.

FOSI AUDIO ZA3 Stereo / Mono Amplifier

This small Fosi Audio ZA3 integrated amplifier has the particularity of being able to operate in both stereo and mono mode. It incorporates two TI TPA3255 amplifier chips enabling it to deliver up to 2x150W or 1x240W @ 4Ω.

www.audiophonics.fr

The above listing is sort of a placeholder. I am not certain of the actual release date.

 

[Julf]

It may not cost much more than a V3 and will allow the option of amplifying balanced gear - in case you run into hum from an AC power ground loop.

Not just balanced gear - you get most of the benefit of a balanced connection even when driving a balanced input from an unbalanced source.

 

[onethreetwowaitthatswrong]

Be advised Fosi is developing a similar amplifier with unbalanced RCA inputs and differential (balanced) inputs. It may not cost much more than a V3 and will allow the option of amplifying balanced gear.

With the sub out as well, that sounds like a very promising future piece of gear ! The lack of remote control might be a deal breaker for some though, even though there will be a trigger

 

[Guddu]

Not just balanced gear - you get most of the benefit of a balanced connection even when driving a balanced input from an unbalanced source.

What do you mean by "most benefit of a balanced connection even when driving a balanced input from an unbalanced source"?
Can you put down some details?

 

[Julf]

What do you mean by "most benefit of a balanced connection even when driving a balanced input from an unbalanced source"?
Can you put down some details?

The main benefits of a balanced connection are cancellation of common mode noise and avoidance of ground loops. If you use a balanced cable and balanced input, you still get the ground loop avoidance and pretty good common mode noise cancellation.

A good piece by Bruno Putzeys (the designer of the Hypex and Purifi amps): The G Word, or How to Get Your Audio off the Ground

 

[Guddu]

The main benefits of a balanced connection are cancellation of common mode noise and avoidance of ground loops. If you use a balanced cable and balanced input, you still get the ground loop avoidance and pretty good common mode noise cancellation.

A good piece by Bruno Putzeys (the designer of the Hypex and Purifi amps): The G Word, or How to Get Your Audio off the Ground

Thanks, but I was referring to what you mentioned about balanced inputs from unbalanced source.

 

[Julf]

Thanks, but I was referring to what you mentioned about balanced inputs from unbalanced source.

Oh, sorry, I forgot a very important part in my reply. Let me try again:

The main benefits of a balanced connection are cancellation of common mode noise and avoidance of ground loops. If you use a balanced cable and balanced input, you still get the ground loop avoidance and pretty good common mode noise cancellation even if your source is unbalanced.

 

[antcollinet]

The main benefits of a balanced connection are cancellation of common mode noise and avoidance of ground loops. If you use a balanced cable and balanced input, you still get the ground loop avoidance and pretty good common mode noise cancellation.

A good piece by Bruno Putzeys (the designer of the Hypex and Purifi amps): The G Word, or How to Get Your Audio off the Ground

Not really - balanced means impedance balanced. If the impedance of the two signals is not identical you lose the common mode rejection. It is quite difficult to balance the Aout/0V - even if you try to put a matching resistor in the 0V connection - and not many do even that.


You can probably get close if you actually measure the output impedance of your unbalanced connection, then put a matching resistor in the 0V(cold) connection.

 

[Julf]

Not really - balanced means impedance balanced. If the impedance of the two signals is not identical you lose the common mode rejection. It is quite difficult to balance the Aout/0V - even if you try to put a matching resistor in the 0V connection - and not many do even that.


You can probably get close if you actually measure the output impedance of your unbalanced connection, then put a matching resistor in the 0V(cold) connection.

Balanced also means differential. Did you read the piece by Bruno Putzeys?

 

[antcollinet]

Balanced also means differential. Did you read the piece by Bruno Putzeys?

Yes, I did, as I have before.

Because yes, differential is necessary for balanced, but is not, on its own, inherently balanced. Bruno specifically states that "balanced" refers to the balancing of impedances. In fact in his sub heading "Balance" all he talks about is impedance.

When a Wheatstone bridge is exactly nulled, the term we use is that the bridge is balanced. That is where the word “balanced connection” comes from. It has nothing at all to do with one voltage going up while the other goes down, but with divider ratios being equal. Don’t think uppy-downy. Think equilib- rium. Zen. Ooohmmmmm...

 

[Sokel]

Yes, I did, as I have before. Did you?

Differential is necessary for balanced, but is not, on its own, inherently balanced. Bruno specifically states that "balanced" refers to the balancing of impedances. In fact in his sub heading "Balance" all he talks about is impedance.

I love the way bruno writes.
It's a common thing amongst people who know exactly what they're talking about to be straight,simple and have humor.

 

[Julf]

Yes, I did, as I have before. Did you?

Would I post a link to a document I haven't read?

Because yes, differential is necessary for balanced, but is not, on its own, inherently balanced. Bruno specifically states that "balanced" refers to the balancing of impedances. In fact in his sub heading "Balance" all he talks about is impedance.

No, impedance is not all he talks about - he talks about voltages too. And just before that sub-heading he states:

To change an output from single-ended to differential all you need to add
is an extra wire to carry the reference potential to the receiver.

 

[antcollinet]

Would I post a link to a document I haven't read?

No, impedance is not all he talks about - he talks about voltages too. And just before that sub-heading he states:

To change an output from single-ended to differential all you need to add
is an extra wire to carry the reference potential to the receiver.

Sorry - that "did you?" was un-necessarily antagonistic without the obligatory smiley. I've deleted it.

He refers to voltages only in terms of how they are impacted by the balance - or otherwise - of the impedances.

And yes - it becomes differential - but as I state above, differential is not balanced.

 

[Julf]

And yes - it becomes differential - but as I state above, differential is not balanced.

Right. Can we at least agree that the differential part is what helps with avoiding ground loops?

 

[antcollinet]

Right. Can we at least agree that the differential part is what helps with avoiding ground loops?

The differential part means that the noise injected by ground loops appears as a common mode signal on the differential pair - that part is correct.

The effectiveness of the rejection of that common mode noise on the differential input depends on the degree of balance in the impdances. Lets look at the maths - and using the CMMR calulations shown by Bruno.

Lets assume the balanced input has a 10K input impedance that is perfectly balanced. The output impedance is (say) 200ohm.

For a balanced ouput using 0.5% resistors, then the mismatch on the output will be at most 1% = 2 ohms

CMRR = -20 log(2/10000) = 74dB

For an unbalnced output, with the ground separated at the output to shield and cold connection:

Then the mismatch on output impedance will be the full 200ohm.

CMRR = -20 log(200/10000) = 33dB

So that 33dB looks pretty good compared to a typical RCA unbalanced connection - but bear in mind that is at the input to the diff amp. Imperfections in the electronics will reduce the CMRR further.

Probably in most systems - even ones where ground noise is a problem for RCA connections - the improvement will be enough. But I'd guess that in systems where ground noise is severe, then the loss of 40dB rejection can make the differerence between the noise being audible or not.

Put it this way - if your system SNR is ground noise dominated, then the difference between just differantial - and balanced differential in the situation above - is 40dB loss of SNR.

On the other hand - the system SNR improvement (again if ground noise dominated) going from RCA to pseudo differential should be 33dB

 

[Julf]

On the other hand - the system SNR improvement (again if ground noise dominated) going from RCA to pseudo differential should be 33dB

Indeed, quite an improvement. I totally agree that fully balanced is by far the best solution, but my point is that you still get quite a benefit from a differential/balanced input, even if the source is unbalanced.

 

[VVR]

Still, in real world we rarely see more than a couple of db improvement with balanced vs single ended signals. And very often there is no improvement at all... even the opposite in poorly designed circuits.
Theory is correct, implementation is another story.

 

[Julf]

Still, in real world we rarely see more than a couple of db improvement with balanced vs single ended signals. And very often there is no improvement at all... even the opposite in poorly designed circuits.
Theory is correct, implementation is another story.

If you don't have any ground loop issues or any electrical noise to start with, there will of course not be much improvement, but I haven't came across any cases where a differential connection would be worse than a single-ended - there is a reason most high-speed digital communications use differential signalling. Of course there can be very poorly designed differential input circuits out there...

 

[VVR]

No doubt it's superior! But I've seen poor results in amplifier measurements in Stereophile
I guess doubling the amount of active devices in poorly designed circuits introduces more noise than the noise cancelation effect from the balancing.
For the PA5 II the difference is about 2db better in favour of balanced. There is more feedback available in balanced and yet not so crazy better! ... And considering that this design is probably as good as it gets from noise perspective. Many older discrete circuits may be quite worse.