Topping PA5 Review (Amplifier)

[aj625]

Sorry you are the one… I see this current delivery argument like some amp is going to spit out 200A.
I suppose if it is a current source and the speaker impedence goose high or low then one could get something.

But if it is applying a voltage gain, then whatever the rail is at, divided by the speaker impedance, will be the current.
It is not force feeding the current to the speaker like a XMAS goose. It is providing voltage which determines the current needed to hold that voltage up.

To say otherwise is something I cannot seem to understand.

Don't forget the impedance of source you can't ignore it. It is the part of circuit and it eats the voltage and thus reducing available current to the load.

 

[pjug]

Curious: how could it deliver 100 watts per channel? Do you see the power supply it is using and the total power available?

I am not sure this question has been addressed. It came up on the other thread too (I also wondered about it) and John said the power supply is 300W for some length of time, and the amp will shut down before the power supply goes into protection. Amir's measurements back him up.

 

[aj625]

@aj625 the current idea comes from the reserve of current stored in the power supply smoothing caps. The current through the speakers is determined by the resistance of the speaker and is not something the speaker draws or as @Holmz wrote is not force fed into the speaker in huge amounts on musical demand. There is a finite amount of current determined by the impedance.

NOTE: On musical transients more current can be delivered to the speaker if the power supply has it available and that is limited by the resistance of the power supply and the amp topology.

That's what I was saying current deliverability for certain power is function of internal resistance( output impedance ) say you have a 1kw amp with output impedance of 1ohm and you attach a 8ohm speaker and use 20w power and now you attach same speaker to a 50w amp with output impedance of 0.1ohm and still draw 20w. Obviously an amp with output impedance of 0.1ohm will be able to supply more current for the transient when speaker impedance dips to say 2ohm than 1kw amp having 1ohm output impedance because out impedance of amp is something which is fixed irrespective of how much power you draw.

 

[Doodski]

That's what I was saying current deliverability for certain power is function of internal resistance( output impedance ) say you have a 1kw amp with output impedance of 1ohm and you attach a 8ohm speaker and use 20w power and now you attach same speaker to a 50w amp with output impedance of 0.1ohm and still draw 20w. Obviously an amp with output impedance of 0.1ohm will be able to supply more current for the transient when speaker impedance dips to say 2ohm than 1kw amp having 1ohm output impedance because out impedance of amp is something which is fixed irrespective of how much power you draw.

I'm not sure where the numbers you are exemplifying come from but the idea is on track. It's called power output linearity when the output power doubles for a halving of the speaker impedance. To do that requires very low output impedance and a power supply transformer that can supply the power to the smoothing caps while all this is going on. I've had such a amp that is linear to 2Ω and the sound is pretty nice and imaging is excellent.

For perspective: These are real scenario numbers.

20WRMS@8Ω is 17.9V peak +/- @ 2.2A
40WRMS@8Ω is 25.3V peak +/- @ 3.2A

50WRMS@8Ω is 28.3V peak +/- @ 3.5A
100WRMS@8Ω is 40V peak +/- @ 5A

 

[PeteL]

That is true, hence the term peak. I was commenting only on the suggested comparison to a tone.

My point was not about judging exactly to compare dynamic music loudness general perceived volume to a single tone. That is difficult to do, but to try to explain that it's a measure of loudness, you measure speaker sensitivity, you calculate The A Weighting of a SPL measurment with a 1k tone as a reference. I don't know, but, I feel you are nitpicking a bit... Saying you can't compare those units to dBFS, you can't apply this to a Tone... At some point we are talking dBs, we have to use these concepts and references. The LUFS Value or a tone recorded at -14 dBFS will be -14 LUFS, that's all I meant a tone is a constant value, you don't need to explain to me that music has peaks, has a crest factor, etc, I know that's the point.

 

[Rottmannash]

I am really wondering what SPL levels most people here are listening to.

32WPC gets you 15 dB on top of the the sensitivity lvl of your speakers. That seems well over the expectations for any nearfield / desktop setup.

I cannot imagine listening to 85+ dB for longer than half an hour a day. I would go deaf really soon otherwise.

This is a great quality amp for the price for anyone who doesn't want to go deaf.

Even purifi amps go about just 4 dB louder than this little beauty.

Would I use it for my main system in the living room? Probably not (I have a purifi based amp). But for everything else it seems to be pushing the boundaries we have known so far. I would not hesitate to buy it for my desktop or bedroom setup. In fact, I will probably get E50 + PA5 combo to replace my Loxjie A30 once the power brick gets certified for use in the EU.

I guess the only question left to answer is: How long can it be run at near full power with demanding speakers (those that dip to low impedance) before shutting down?

 

[aj625]

Sorry you are the one… I see this current delivery argument like some amp is going to spit out 200A.
I suppose if it is a current source and the speaker impedence goose high or low then one could get something.

But if it is applying a voltage gain, then whatever the rail is at, divided by the speaker impedance, will be the current.
It is not force feeding the current to the speaker like a XMAS goose. It is providing voltage which determines the current needed to hold that voltage up.

To say otherwise is something I cannot seem to understand.

That's what I was saying current deliverability for certain power is function of internal resistance( output impedance ) say you have a 1kw amp with output impedance of 1ohm and you attach a 8ohm speaker and use 20w power and now you attach same speaker to a 50w amp with output impedance of 0.1ohm and still draw 20w. Obviously an amp with output impedance of 0.1ohm will be able to supply more current for the transient when speaker impedance dips to say 2ohm than 1kw amp having 1ohm output impedance because out impedance of amp is something which is fixed irrespective of how much power you draw.

 

[aj625]

? Sorry but current delivery and power rating is the same thing... does ohm law tells you something? If what you mean is that the rated power at 8 ohm is not the only important rated power, yes I agree with that, but from rated power at X ohms, we can derive the current.

That's what I was saying current deliverability for certain power is function of internal resistance( output impedance ) say you have a 1kw amp with output impedance of 1ohm and you attach a 8ohm speaker and use 20w power and now you attach same speaker to a 50w amp with output impedance of 0.1ohm and still draw 20w. Obviously an amp with output impedance of 0.1ohm will be able to supply more current for the transient when speaker impedance dips to say 2ohm than 1kw amp having 1ohm output impedance because out impedance of amp is something which is fixed irrespective of how much power you draw.

 

[Doodski]

supply more current for the transient when speaker impedance dips to say 2ohm

Yes, true. Hit it on the spot.

 

[pjug]

I guess the only question left to answer is: How long can it be run at near full power with demanding speakers (those that dip to low impedance) before shutting down?

This doesn't address <4 ohm, but maybe gives some idea (something like a minute with both channels at 100W)

Topping PA5 measurements )

@JohnYang1997 Is it OK for a customer to use a different power supply? Is there a performance gain (dynamic power at 6ohms or 4 ohms) using something like this? https://www.amazon.com/NES-350-36-Switching-Power-Supply-110-240/dp/B01AWQ0Y52 No. And no.

audiosciencereview.com

 

[Koeitje]

I guess the only question left to answer is: How long can it be run at near full power with demanding speakers (those that dip to low impedance) before shutting down?

Full power with what? Sine wave? Couple of minutes probably. Music? 24/7. Full power RMS is the most useless specification ever.

 

[doug2761]

I think Sonos set the ball rolling years ago with the powered speaker and your phone as the source. Now the KEF LS50 wireless puts a better speaker and better DAC on a Class D amp and adds an HDMI input. But the price of both Sonos and KEF can be wallet draining.

I too, was staring at my 60 pound AB amp and adjacent AVR 11.2 box and wishing it would go away and just leave the floor space. I have the LS50 metas, and a Gustard X16 DAC. I also have a Raspi 4. All I needed was a preamp, and an amp the size of the DAC and that could happen. Yesterday I spotted the preamp in a Black Friday sale and pulled the trigger. I searched and searched for a nice Class D and gave up. And then, poof...I see this review of the Topping.

My dream is coming true. I'm still a little nervous about replacing a 200W /ch AB with a 140 W/Ch D so I'll procrastinate a little more.

"staring at my 60 pound AB amp" - I found myself in the same place earlier this year looking at my McIntosh C48 preamp and MC402 power amp that I've had for many years. I built the Hypex DIY NC400 amps and listened side by side with the McIntosh gear for a few months. I wound up selling the McIntosh gear. It took two adults to transport the McIntosh gear. I loved that McIntosh but let's be honest, nobody wants that much weight to carry around. I'm exclusively digital for my source material. I can cary the Hypex amps, DAC, and raspi in a small box.

 

[xrk971]

would also be useful to know what the gain range on this amp is

Wouldn't Air core inductors be ideal? maybe a little bit big... but saturation and inearety would e closes to perfect.

Air core is of course, ideal from standpoint of linear response. It does radiate EMI more though and is quite a bit larger. The crosstalk between air core inductors is not insignificant either as I demonstrate in a video here.

 

[xrk971]

Full power with what? Sine wave? Couple of minutes probably. Music? 24/7. Full power RMS is the most useless specification ever.

Class D amps (assuming TPA32XX core) are about 90% efficient. For a 140W rated amp that is 14W dissipation per channel. I don’t think this little box with a few vents can dissipate 14W x 2 = 28W continuously. Amp power is sometimes spec’d based on music “program power” and that is typically some amount below peak. You want the amp to be able to hit the peak on occasion but max continuous is much less. Assuming we want 20dB headroom then average power is really 14W continuous. That’s quite loud with circa 85dB sensitive speakers, BTW.

 

[617]

"staring at my 60 pound AB amp" - I found myself in the same place earlier this year looking at my McIntosh C48 preamp and MC402 power amp that I've had for many years. I built the Hypex DIY NC400 amps and listened side by side with the McIntosh gear for a few months. I wound up selling the McIntosh gear. It took two adults to transport the McIntosh gear. I loved that McIntosh but let's be honest, nobody wants that much weight to carry around. I'm exclusively digital for my source material. I can cary the Hypex amps, DAC, and raspi in a small box.

Imagine if they were all in one box though..

 

[Koeitje]

Class D amps (assuming TPA32XX core) are about 90% efficient. For a 140W rated amp that is 14W dissipation per channel. I don’t think this little box with a few vents can dissipate 14W x 2 = 28W continuously. Amp power is sometimes spec’d based on music “program power” and that is typically some amount below peak. You want the amp to be able to hit the peak on occasion but max continuous is much less. Assuming we want 20dB headroom then average power is really 14W continuous. That’s quite loud with circa 85dB sensitive speakers, BTW.

Exactly, its pointless to talk about sine wave output. Music is not a test signal. If you can play music with the peaks hitting 140W power draw in a 24/7 situation you are good to go (and probably deaf).

 

[Lambda]

Air core is of course, ideal from standpoint of linear response. It does radiate EMI more though and is quite a bit larger. The crosstalk between air core inductors is not insignificant either as I demonstrate in a video here.

This shape is very non ideal for cross talk and emission.
toroidal inductors help a lot.

 

[xrk971]

Please @JohnYang1997 - transfer this amplifier technology over to something larger, with balanced inputs, a remote, internal power supply and double the power, and you'll have my money in an instant

I am not sure I understand what is meant by a remote internal power supply? If it is internal, how can it be remote? Maybe not in the same PCB? There are commercial amps with a separate SMPS but in the same chassis as the TPA3255, and balanced or single ended inputs in a larger chassis.

 

[xrk971]

This shape is very non ideal for cross talk and emission.
toroidal inductors help a lot.

I agree. But toroidal inductors are no longer air core.

 

[Walter]

I am not sure I understand what is meant by a remote internal power supply? If it is internal, how can it be remote? Maybe not in the same PCB? There are commercial amps with a separate SMPS but in the same chassis as the TPA3255, and balanced or single ended inputs in a larger chassis.

Remote (control) comma internal power supply