Topping LA90 Review (Integrated Amplifier)

[Deleted member 46664]

You think so, right? Really?

Below is a record during listening in home living room conditions with a speaker with some 89dB/2.83V/1m sensitivity.

The apparent clipping is not the amplifier, but it is in the CD data. Tchaikowsky 1812 Telarc, BTW. 35Vp amplifier output.

View attachment 202757

Recording of October 1979: The Telarc 1812 Overture

Tchaikovsky: 1812 Overture (Op.49), Capriccio Italien (Op.45 ), "Cossack Dance" from Mazeppa (LP), plus Marche slav, Op.31, Polonaise and Waltz from Eugene Onegin , Op.24, Festival Coronation March (CD). Cincinnati Pops Orchestra, Kiev Symphony Chorus; Children's Choir of Greater Cincinnati...

www.stereophile.com

The scope panel is reporting 58 volts (297 watts) and the lower level is showing at 20v ( 35w) ... I doubt that most people (myself included) ever listen at those levels. It would be straight up painful. I've done work on theatre systems that aren't driven that hard.

I listen at 80dbz levels... which is just about a watt. Peaks might hit 90 db... or about 10 watts. Which is far more typical.

Some examples of 80db sounds ... Garbage disposal, dishwasher, average factory, freight train (at 15 meters). Car wash at 20 ft (89 dB); propeller plane flyover at 1000 ft (88 dB); diesel truck 40 mph at 50 ft (84 dB); diesel train at 45 mph at 100 ft (83 dB). Food blender (88 dB); milling machine (85 dB); garbage disposal (80 dB).

Comparative Examples of Noise Levels - IAC Acoustics

This blog post compares examples of noise levels. It is broken down by Noise Source, Decibel Level, and Decibel Effect.

www.iacacoustics.com

 

[pjug]

How do you know it is a single polarity PSU? There might be a voltage regulator/capacitance multiplier with artificial center point inside the amp case. Remember that one of the heatsinks gets more hot. If Amir simply opened the box and used the DMM we would have known.

Someone posted an image -- I can't find it now. I think it may have been this one but manipulated to see a little clearer. Can you get anything from this?

TOPPING LA90 Discrete NFCA Power Amplifier / LA90 Power Amplifier

Introducing the LA90 Discrete Amplifier, a high-fidelity audio solution that brings the benefits of discrete NFCA power amplifier circuitry to your listening sessions. This is a brand-new iteration of the LA90 model that features greater power in stereo mode, advanced signal-to-noise ratio for...

apos.audio

edit: are the coils and caps at the top of the image showing same as the standard single ended output example shown in the LM3886 data sheet Figure 3?

 

[Talisman]

If you think €800 is a fair price for the things you have then I would think ~€900 for the LA90 (which outperforms any of your current gear) would also be a fair price.

I promised myself to leave it alone, but I feel like I want to answer that.
"which outperforms any of your current gear"
but it surpasses in what exactly? in something that you can feel and perceive as an improvement?

When some audiophile shows up with a boutique dac claiming they can hear wonders, infinite stages, smooth mid-highs, etc all, ALL insist on doing a blind ABX comparison to determine if they can really hear these wonders without using their eyes.
But when it comes to a product with fantastic measurements no one seems to object that a blind test in ABX with matched levels should be done to determine if those wonders in the graphs represent a REAL improvement in the audio performance perceived to the ear compared to much cheaper products. .
"Listen only with your ears not with your eyes" could be ASR's motto and I totally agree, but not only when the suggestion is given by an elegant wooden case, but also when it is given by beautiful graphics to see.
Without making considerations on reliability, because nobody can know anything yet. In real listening, what improvement would be 800 euro LA9 compared to 350 euro PA5?
Listen with your ears.

 

[Overseas]

Why do you compare LA90 vs PA5?
Did you choose PA5 with your ears?

 

[Talisman]

Why do you compare LA90 vs PA5?
Did you choose PA5 with your ears?

If this is the reasoning, then let's stop going against audiophiles, they don't choose with their ears either, but we blame them and make fun of them for it. We make the same irrational choices but we feel superior because we base them on numbers and not snake oil, but the reality is that just like them we are willing to pay a lot more for something we don't get any acoustic benefit from.
However I compared them because they are two topping products, they are two products with SOTA measurements and similar powers but two very different price categories. Ah ... incidentally, I didn't buy pa3 for almost the same reason, beautiful and performing, but for the same amount I can find something that sounds identical to my ears but with other functions that I think are more useful.

 

[TheBatsEar]

"which outperforms any of your current gear"
but it surpasses in what exactly?

THD and N.
Why do we have to explain this again and again?

in something that you can feel and perceive as an improvement?

If you want feelings you are in the wrong neighborhood.
Just answer this (silently to yourself please): what SINAD do i think i can perceive? Then go buy your amp. Let us enjoy the best, despite what you think is perceiveable or not.

But when it comes to a product with fantastic measurements no one seems to object that a blind test in ABX with matched levels should be done

The heck are you saying? THD and N are key metrics, just like W@8R. You don't say "Well, this amp may have 5 more Watts, but you have to ABX if it sounds better". What is wrong with you? No, really:

It's a measurement of THD+N, the better numbers win.

In real listening, what improvement would be 800 euro LA9 compared to 350 euro PA5?

Then get a PA5 if you don't like the best measurements. What are you even arguing?

 

[Human Bass]

Can it be used with headphones and IEMs the same way some people use the latest SMSL amp or the Benchmark ABH2? Does it need to have an external preamp in order to get fine volume control? Any reason (aside from accidentally blowing your headphones) that you wouldn’t want to do this from a specifications or performance perspective?

Using an speaker wire to 4-pin xlr adaptor? Yes.

 

[mdsimon2]

Not really it shows the current limiter kicks in at 5A. (1 channel driven, 2 ohm)
At 20.5V max. output voltage one can calculate the amp is designed for 4 ohm (thus 8 ohm bridged)
It just doesn't go up in smoke when 2 ohm is applied.

As speakers are never 2 ohm over the entire audible range and music has a wide range of frequencies that may all exist at the same time it is not a big problem if the impedance dips at certain frequencies as the voltage level for those frequencies will never reach 20V only the combined voltage will do so. As not all frequencies will be in 2 ohm the output current limit even in 2 ohm will not be reached (using music).
Only on a measurement bench it will.

I would not recommend to use this amp in bridged mode with 4 ohm speakers though even though it does not blow up it is designed to be used with 8 ohm nominal (meaning higher and lower impedance dips/peaks are allowed).

The power supply can deliver 256W (continuous) and with about 75% efficiency for class A-B = 190W total available.
So 95W per channel (continuous) so power supply is the limiter it seems.

View attachment 202721

Thank you so much for the explanation, that makes sense.

In my ignorance I had always assumed that current draw between say 4 ohm x 2 and 8 ohm x 1 would be identical but after running the numbers they clearly are not. Here is my quick attempt at presenting this in tabular form for power roughly at 1% THD+N for each case. Obviously for bridged cases the voltage is the combined voltage from the two channels.

Z	Ch	P_ch	V	A_ch	P_tot	P_psu	Limiting scenario
8	2	50	20	2.5	100	133	Output voltage
4	2	90	19	4.7	180	240	Current limit / PS
8	1	180	38	2.4	180	240	PS
2	2	50	10	5.0	100	133	Current limit
4	1	190	28	3.4	190	253	PS

Does this look about right? Thanks again for helping me understand this.

Michael

 

[ATAineri]

4A is really not good. In playback acoustic can become 2 or 1 ohm or less. I saw acoustic that needs 50 W, but in playback to play peaks it needs 30A.
I really want this, but I afraid that it will loose all peaks

 

[Talisman]

THD and N.
Why do we have to explain this again and again?


If you want feelings you are in the wrong neighborhood.
Just answer this (silently to yourself please): what SINAD do i think i can perceive? Then go buy your amp. Let us enjoy the best, despite what you think is perceiveable or not.


The heck are you saying? THD and N are key metrics, just like W@8R. You don't say "Well, this amp may have 5 more Watts, but you have to ABX if it sounds better". What is wrong with you? No, really:
View attachment 202762
It's a measurement of THD+N, the better numbers win.


Then get a PA5 if you don't like the best measurements. What are you even arguing?

I do not think there is anything wrong with me, I am making objections based on evaluations that can be shared or not, please do not offend and do not attack personally, I did not.

 

[pjug]

Thank you so much for the explanation, that makes sense.

In my ignorance I had always assumed that current draw between say 4 ohm x 2 and 8 ohm x 1 would be identical but after running the numbers they clearly are not. Here is my quick attempt at presenting this in tabular form for power roughly at 1% THD+N for each case. Obviously for bridged cases the voltage is the combined voltage from the two channels.

Z	Ch	P_ch	V	A_ch	P_tot	P_psu	Limiting scenario
8	2	50	20	2.5	100	133	Output voltage
4	2	90	19	4.7	180	240	Current limit / PS
8	1	180	38	2.4	180	240	PS
2	2	50	10	5.0	100	133	Current limit
4	1	190	28	3.4	190	253	PS

Does this look about right? Thanks again for helping me understand this.

Michael

I haven't looked at your table carefully and I'm not more an expert than you are anyway, but I think there can be also be a limit from the limit on chip amp power dissipation. Possibly built in to the device, thermal or otherwise.

 

[mdsimon2]

I haven't looked at your table carefully and I'm not more an expert than you are anyway, but I think there can be also be a limit from the limit on chip amp power dissipation. Possibly built in to the device, thermal or otherwise.

That makes sense to me, I guess what I am calling "PS limited" could be some sort of other power limit.

What did not initially make sense to me was how low the 2 ohm stereo power was but a limit on output current at about 5 A per channel explains it well.

Michael

 

[solderdude]

Someone posted an image -- I can't find it now. I think it may have been this one but manipulated to see a little clearer. Can you get anything from this?

TOPPING LA90 Discrete NFCA Power Amplifier / LA90 Power Amplifier

Introducing the LA90 Discrete Amplifier, a high-fidelity audio solution that brings the benefits of discrete NFCA power amplifier circuitry to your listening sessions. This is a brand-new iteration of the LA90 model that features greater power in stereo mode, advanced signal-to-noise ratio for...

apos.audio

edit: are the coils and caps at the top of the image showing same as the standard single ended output example shown in the LM3886 data sheet Figure 3?
View attachment 202764

Most likely yes, it is called a Boucherot filter (Zobel network) and helps keeping the amp stable under unwanted conditions.

 

[pjug]

Most likely yes, it is called a Boucherot filter (Zobel network) and helps keeping the amp stable under unwanted conditions.

But also is output through the caps in upper right, feeding the coils, and so the whole thing is a single ended design?

 

[solderdude]

4A is really not good. In playback acoustic can become 2 or 1 ohm or less. I saw acoustic that needs 50 W, but in playback to play peaks it needs 30A.
I really want this, but I afraid that it will loose all peaks

Which speaker do you own that drops impedance to 1 ohm. And when you have then the LA90 should not be on your list.

The specs clearly state ≥ 4 ohm in stereo mode and ≥ 8 ohm in bridged mode.
That excludes >1 ohm and >2 ohm.

When your speaker is that low impedance look for another amp.

 

[pma]

@pjug , 0.7uH//10R behind the output cap is to prevent oscillations from capacitive loading.

 

[anmpr1]

I am happy to wake up and see the new information. After looking at the revised measurements (into lower impedances), my view of the amplifier has changed. For most people in most routine domestic settings, and with most 'normal' loudspeakers, the basic unit should be sufficient, when used within its modest power rating. With highly sensitive loudspeakers, such as horns, the basic unit would be more than enough for almost anyone.

For those needing more power, two units bridged, at $1600.00 would be a good choice. If you need more than two hundred watts/ch then you are either going to have to spend a lot more money to gain similar distortion characteristics (two AHB2 bridged, at six large), or go with something with more measured (but likely not audible) distortion, such as two Vidars, which will give you twice the bridged Topping's output, for less money.

My concerns about warranty/service/support remain, but those issues have nothing to do with the amplifiers basic electrical performance, which cannot be faulted.

 

[solderdude]

But also is output through the caps in upper right, feeding the coils, and so the whole thing is a single ended design?

There is nothing in the specs and measurements that indicates a coupling cap is used at the output.
The power supply either is 64V and some kind of rail splitter is used or ±32V is used.

 

[pjug]

@pjug , 0.7uH//10R behind the output cap is to prevent oscillations from capacitive loading.

Thanks. Are you able to make out whether output goes through the caps like the standard single ended example?

 

[JohnBooty]

@JohnYang1997, your engineering talent is appreciated. Thank you for helping to push the industry forward.