Topping B100 Amplifier Review

[DWPress]

Someone asked if it remembers the power state on AC disconnect. It does NOT. Goes into standby mode.

Thanks Amir. That's unfortunate for my use case.

 

[DWPress]

Ultra-sensitive horns? Those guys won't buy this

If you've read the thread you might note that quite a few of us are considering it.

 

[terryforsythe]

This is a silly discussion. Feedforward is not the patent. Topping could use feedforward, feedback, a combination of, or whatever they want. We had feedforward amplifiers well over 40 years ago.

Perhaps there was a misunderstanding as to what was patented, and when it was patented. They could be referring to a patent that issued long ago and has since expired, or to a patent directed to a component that happened to include feedforward in the design, but the claims were actually directed to another aspect of novelty. It happens all the time.

 

[amirm]

Either I've misunderstood the EU rules on standby power consumption, or Audiophonics shouldn't be selling this.

The low power measurements are not going to be accurate.

 

[kemmler3D]

Not currently.

However, you could use one of their many DACs with both XLR/TRS Balanced and RCA Single-ended Line out.

Connect the balanced out to the B100 and the RCA out to your Sub.

Then use the Sub's crossover knob to match its response to your Mains' natural roll-off.

This is what I do, I run 2 subs off the RCA out of my DX5 Lite with the XLR output going to the mains.

 

[dlaloum]

Any idea how this amp behaves with speakers with <4 ohm?

Don't know, but the rise from 8ohm to 4ohm is less than 50% (goes from 70W to 100W)... which is not promising.

Typically, amps that handle low impedances / high current well, have a rise of around 75% from 8 to 4 ohm and then a bit over 50% from 4ohm to 2 ohm.

Having said that, the Quad current dumper amps are totally stable into any load, although their power delivered drops off dramatically... so ultimately, only testing it will tell.

 

[mhardy6647]

This is a silly discussion. Feedforward is not the patent. Topping could use feedforward, feedback, a combination of, or whatever they want. We had feedforward amplifiers well over 40 years ago.

ayuh (as the Mainiacs Mainers People from Maine would say).

source: https://www.worldradiohistory.com/h.../HiFI-Stereo-Review-1981-01-OCR-Page-0037.pdf

source: https://www.worldradiohistory.com/h...-IDX/IDX/80s/High-Fidelity-1984-06-IDX-54.pdf

 

[totti1965]

There's where you need to understand how much gain you need. This is sensible. Given the power levels at issue, 1.5V would more than drive it into clipping on high gain of around 25dB. It's not an issue since power levels are so low.



Doubt it, unless you're into tubes.


Bingo. I suspect it was a product designed to be measured and top the SINAD chart. THD wars, and all. It was not designed to sell in any sort of rational market since it doesn't fit any actual market need or niche. Passive nearfield monitors? Barely exist. Ultra-sensitive horns? Those guys won't buy this. So who does that leave? Distortion and noise (SINAD) junkies, many of whom will underpower a pair of speakers to get whatever this represents. But that is a market, and I think you're right it hangs out here.

Now, if it didn't have a pair of power bricks, came in a fancy audiophile case, and cost twice as much, from a "branded" name, I could see it. then it would be cheap, even, for the performance, discounting factors like support and long-term reliability.


You assume 1) the patent is even valid, and 2) that the patent covers every possible kind of feed forward correction. A lot of these "patented" technologies are little more than rehashing decades old, public domain idea, with a small twist irrelevant to performance that allows "Company X" to claim it is "patented technology." There are multiple ways to get distortion this low which are long out of patent. There's not exactly any magic sauce here, it's just unique sauce because most audio companies have no clue how to use feedback and rely on decades old designs they don't want to redo.

When companies finally decide to apply the science that has been known for at least 25 years now, they can get it right. See, e.g., the Linn amps reviewed recently in Stereophile (https://www.stereophile.com/content/linn-klimax-solo-800-monoblock-power-amplifier-more-measurements). Shocking performance for a high end manufacturer, and presumably not violating patents, either. When power and gain structure are considered, the Linn is a better amp than a Benchmark (albeit with an insane price). This is there, too, for a fraction of the price but with inadequate power for anything but the smallest of bookshelf speakers or those who always listen quietly. If they could put Linn power behind it and match Linn's measurements and sell it for a few grand? Wow, what an amp that would be. Then put a linear supply in it for reliability (and screw the SINAD) and pop some meters onto the faceplate... Now I'm in! At this point, though, you might just be able to use Hypex switching supplies without reliability concerns. If it pukes, they've presumably sold enough to be able to replace it down the road with available or compatible spares.

Whatever the case, the audio circuit engineering here is very good, even if I don't really understand what flea amp target market this is after. But I've never been in that target market, so perhaps that's the problem. They sell a lot of low powered amps, after all.

There's where you need to understand how much gain you need. This is sensible. Given the power levels at issue, 1.5V would more than drive it into clipping on high gain of around 25dB. It's not an issue since power levels are so low.



Doubt it, unless you're into tubes.


Bingo. I suspect it was a product designed to be measured and top the SINAD chart. THD wars, and all. It was not designed to sell in any sort of rational market since it doesn't fit any actual market need or niche. Passive nearfield monitors? Barely exist. Ultra-sensitive horns? Those guys won't buy this. So who does that leave? Distortion and noise (SINAD) junkies, many of whom will underpower a pair of speakers to get whatever this represents. But that is a market, and I think you're right it hangs out here.

Now, if it didn't have a pair of power bricks, came in a fancy audiophile case, and cost twice as much, from a "branded" name, I could see it. then it would be cheap, even, for the performance, discounting factors like support and long-term reliability.


You assume 1) the patent is even valid, and 2) that the patent covers every possible kind of feed forward correction. A lot of these "patented" technologies are little more than rehashing decades old, public domain idea, with a small twist irrelevant to performance that allows "Company X" to claim it is "patented technology." There are multiple ways to get distortion this low which are long out of patent. There's not exactly any magic sauce here, it's just unique sauce because most audio companies have no clue how to use feedback and rely on decades old designs they don't want to redo.

When companies finally decide to apply the science that has been known for at least 25 years now, they can get it right. See, e.g., the Linn amps reviewed recently in Stereophile (https://www.stereophile.com/content/linn-klimax-solo-800-monoblock-power-amplifier-more-measurements). Shocking performance for a high end manufacturer, and presumably not violating patents, either. When power and gain structure are considered, the Linn is a better amp than a Benchmark (albeit with an insane price). This is there, too, for a fraction of the price but with inadequate power for anything but the smallest of bookshelf speakers or those who always listen quietly. If they could put Linn power behind it and match Linn's measurements and sell it for a few grand? Wow, what an amp that would be. Then put a linear supply in it for reliability (and screw the SINAD) and pop some meters onto the faceplate... Now I'm in! At this point, though, you might just be able to use Hypex switching supplies without reliability concerns. If it pukes, they've presumably sold enough to be able to replace it down the road with available or compatible spares.

Whatever the case, the audio circuit engineering here is very good, even if I don't really understand what flea amp target market this is after. But I've never been in that target market, so perhaps that's the problem. They sell a lot of low powered amps, after all.

Here they are discussing some technical stuff of the Benchmark amp:

At least in case of benchmark, I trust them, if they say that there were two engineering teams, one from thx an one from benchmark which then from 2006 on cooperated and have hold many patents from their AHB-2 project. I don’t remember the time stamp, when the benchmark guy tells this stuff.

The AHB-2 Project was expensive and it supposed to set a new standard in studio work for the professional guys.

 

[ryanmh1]

If you've read the thread you might note that quite a few of us are considering it.

... I should have clarified that with "outside of ASR".. in my mind for some reason, the guys with those niche speakers tend to want some offbeat tube amp, and the goofy yet often cool horns are a way to be able to use a 5W amp. Obviously, this is not so limited.

The challenge, I think, is this: Once you're selling into the objectivist marketplace with power bricks and cheap cases, the dollar per watt race is on, once you're into sonic transparency. This is well, well past that assuming there's nothing undesirable and unmeasured. Why this over something similarly powered and transparent that costs $250? Commoditization quickly becomes a significant marketing problem. Hopefully Topping can integrate this performance level into products that avoid this. Truly some fine engineering work. Would be awesome to see it in a "pride of ownership" product. But maybe $300 gadgets is sustainable? Who knows. So far so good, I suppose.

 

[manisandher]

Buying an amp with 0 dB gain was not remotely possible 10+ years ago, even 5.

My >15-year-old First Watt F4 monos had 0dB voltage gain. They acted as an 'impedance converters' between the source and the speakers. Absolutely fine for high-sensitivity horns. I suspect the B100s in 0dB gain mode would be too.

Mani.

 

[Beershaun]

I have a few questions. What is the use case for the 0 DB low gain mode? I don't understand what the amplifier is actually doing and when you would want to set it in that setting. That seems like an invalid comparison to put it in the SINAD chart. The high gain mode (25db) is much more "apples to apples" comparison on the SINAD chart.

Also, this doesn't seem like enough power at 8ohms to drive anything loud enough below say 150hz. So while the design and SINAD performance is impressive, it seems like you'd have to cross over a subwoofer fairly high to unload the main speakers enough to not clip the amp any time you have anything like a bass guitar or kick drum playing. Seems like it would be fine for chamber music or orchestra sans percussion.

All that said, this is a very impressive piece of engineering from a company that continues to produce high performance at extremely low prices! I love both the Topping DAC and headphone amp I have!

Keep up the great work and keep making more powerful Amps!

 

[staticV3]

What is the use case for the 0 DB low gain mode?

High sensitivity speakers and/or short listening distance, coupled with a DAC that provides high output voltage (e.g. RME, octo, Benchmark, etc).

I don't understand what the amplifier is actually doing

It's providing current and low output impedance.

and when you would want to set it in that setting.

When your DAC has enough output voltage for a satisfying listening experience, despite no additional gain from the Amp.

Also, this doesn't seem like enough power at 8ohms to drive anything loud enough below say 150hz.

Kind of useless blanket statement.

"Enough" depends on a bunch of factors and will mean something different for every person and situation.

For example, I'm using some old, custom-build 3-way floorstanders that play down to ~30Hz with ear-splitting loudness, despite the fact that they're being driven by a 30W Beolink Passive Amp.

 

[bargainguy]

Many thanks for the review, Amir.

I'm somewhat surprised at the antipathy here toward amps with external power supplies. Didn't see that one coming. I get the other caveats about this amp, but having an external PS Is a plus for me, not a minus. Just shows how differently we all think about the same thing.

I'm assuming - perhaps incorrectly - that Topping would not have been able to achieve this level of performance in a single chassis for $299. I'm also wondering if it was designed from the ground up using an external PS to achieve that level of performance.

This amp could be endgame for me as long as I work within its limitations. Stay away from inefficient, hard-to-drive speakers, don't listen too far away and don't induce ear-bleeding levels. Seems pretty simple.

 

[rongon]

Class B amps are only used in RF due to crossover distortion. So the common version is a hybrid Class AB, not class B. Here, feedback is used so much as to nullify crossover distortion in class B. Much like the way Hypex amps use a lot of feedback to deal with all sorts of non-linearities in class D.

In the old texts about audio amplifier design, it was often stated that negative feedback cannot smooth out the crossover notch in the output waveform caused by class B operation, because there is no gain with both push-pull output devices in cutoff, so there can be no negative feedback. I'm sure the Topping engineers have done something to get past this limitation, but I wonder exactly what. Whatever it is, it can't be as simple as just slapping as much negative feedback around the circuit as it can stand before oscillating.

Also, you don't do square wave testing of power amplifiers. That would show any ringing or instability. Since this is a class B design with great gobs of negative feedback employed, stability would be a concern. Would it be possible for you to run a 10kHz square wave through the amp (at a moderately low level, of course) and show us the output? That might be illuminating. Thanks.

 

[Palladium]

... I should have clarified that with "outside of ASR".. in my mind for some reason, the guys with those niche speakers tend to want some offbeat tube amp, and the goofy yet often cool horns are a way to be able to use a 5W amp. Obviously, this is not so limited.

The challenge, I think, is this: Once you're selling into the objectivist marketplace with power bricks and cheap cases, the dollar per watt race is on, once you're into sonic transparency. This is well, well past that assuming there's nothing undesirable and unmeasured. Why this over something similarly powered and transparent that costs $250? Commoditization quickly becomes a significant marketing problem. Hopefully Topping can integrate this performance level into products that avoid this. Truly some fine engineering work. Would be awesome to see it in a "pride of ownership" product. But maybe $300 gadgets is sustainable? Who knows. So far so good, I suppose.

Not sure why anyone pick 2 of these over a much cheaper PA5 II, their power per channels are in the same ballpark.

Besides, most of us are never gonna push 50W to speakers without also giving ourselves hearing damage.

 

[terryforsythe]

This amp could be endgame for me as long as I work within its limitations.

You also can look at it's bigger brother, the B200, if your speakers have low efficiency.

 

[rongon]

This is a silly discussion. Feedforward is not the patent. Topping could use feedforward, feedback, a combination of, or whatever they want. We had feedforward amplifiers well over 40 years ago.

Is 'feedforward' the same as positive feedback (regeneration)? I could see using that as a way to fill in the crossover notch inherent in push-pull class B operation.

Who here understands the concept of class B operation?

Class B operation was employed in the output stages of radio transmitters and giant PA amplifiers back in the day. The idea is to have the opposite polarity output devices at low or zero current draw with no signal, but start to draw current when signal is applied. You get maximum efficiency that way, but there's a gap at the zero crossing of each sine wave where both devices switch off, causing the notch in the output sine wave that you see in the above illustration.

My question is what magic trick did the Topping engineers employ to get rid of that crossover notch? The classical electronics texts state that simply applying great gobs of negative feedback won't do the trick, because both output devices are momentarily shut off, therefore there is no gain, therefore there is no signal to drive the negative feedback loop. So how did Topping do it?

One problem is that an amplifier with great gobs of negative feedback will need to be carefully optimized so that it doesn't ring or oscillate on signal transients. The classic way to test that was to apply a fairly high frequency square wave (like 10kHz) to the input of the amplifier and look at the amp's output on an oscilloscope. If the amp is ringing, the corners of the square wave will look exaggerated, with a spike coming up from the corner of the wave, instead of a perfect 90 degree angle. Something like this:

That's why I asked Amir to show us what the output looks like with a 10kHz square wave input, if possible.
The output waveform should look as close as possible to this:

 

[EJ3]

Either I've misunderstood the EU rules on standby power consumption, or Audiophonics shouldn't be selling this.

Maybe not there?

 

[JSmith]

Is 'feedforward' the same as positive feedback (regeneration)?

No, not the same at all.


JSmith

 

[Old_School_Brad]

Is 'feedforward' the same as positive feedback (regeneration)? I could see using that as a way to fill in the crossover notch inherent in push-pull class B operation.

Who here understands the concept of class B operation?
View attachment 392733
Class B operation was employed in the output stages of radio transmitters and giant PA amplifiers back in the day. The idea is to have the opposite polarity output devices at low or zero current draw with no signal, but start to draw current when signal is applied. You get maximum efficiency that way, but there's a gap at the zero crossing of each sine wave where both devices switch off, causing the notch in the output sine wave that you see in the above illustration.

My question is what magic trick did the Topping engineers employ to get rid of that crossover notch? The classical electronics texts state that simply applying great gobs of negative feedback won't do the trick, because both output devices are momentarily shut off, therefore there is no gain, therefore there is no signal to drive the negative feedback loop. So how did Topping do it?

One problem is that an amplifier with great gobs of negative feedback will need to be carefully optimized so that it doesn't ring or oscillate on signal transients. The classic way to test that was to apply a fairly high frequency square wave (like 10kHz) to the input of the amplifier and look at the amp's output on an oscilloscope. If the amp is ringing, the corners of the square wave will look exaggerated, with a spike coming up from the corner of the wave, instead of a perfect 90 degree angle. Something like this:

View attachment 392738

That's why I asked Amir to show us what the output looks like with a 10kHz square wave input, if possible. It should like as close as possible to this:

View attachment 392739

Instead of assuming they have come up with a method of getting rid of crossover distortion, we should ask if it ever was an audible issue to begin with?
A lot of pro audio PA amplifiers are class B in practice but don't have audible distortion.

Has anyone ever heard crossover distortion? -I haven't. And I've had a lot of pro and consumer gear during the years.