Topping B100 Amplifier Review

[rongon]

At the risk of once again being derided for subjective input, I will tell you my pair of B100s do not sound sterile or closed in at all—I got that sensation when I auditioned an SMSL SA400, which I purchased used and immediately sold right back off because of that sound despite its wonderful measurements. The Fosi Class-D amps and the B100s sound wonderful in comparison, so I politely disagree with your blanket statement in spite of the B100’s triple nested feedback design.

-Ed

That's great to hear about the Fosi and this B100 amp. I know what that 'closed in' sound is like, and it bothers me. I remember hearing it from an NAD power amp from the 1990s, as well as Crown power amps from the 1980s. 'Dull but harsh' is what I'd call it. But I have also heard solid state amps that sound smooth and don't bother me. I have this weird aversion to reproduction that calls attention to its being electronic. I want to be fooled into thinking I'm hearing something like live people making sounds. I'm really intrigued by this B100 amp because it's 1) apparently using a class B output stage, which is unusual, and 2) not expensive, so I could actually afford a pair if I get serious about it.

The two class D amps I've owned so far are a small TPA3116 thingie made by Breeze Audiothat sounds a bit loose in the bass, murky in the midrange, and loads down easily driving difficult bookshelf speakers, and a Behringer A800 which is plenty powerful but has a bit of an electronic 'edge' to its high frequencies (I can't explain it better than that) perhaps a result of more intermodulation at higher frequencies (as measured in the ASR review) than measured in the Fosi V3, for example. Right now I'm suffering with an old Hafler P1000 amp, which drives my speakers well enough for my small space and is relatively smooth sounding. I'd love to be able to improve on that, but I can't afford a Benchmark, Bryston, etc. I'm hoping Topping has made a smooth sounding, just powerful enough, affordable amp that fills the bill, even if it is running its output stage in class B linearized by great gobs of negative feedback!

 

[MAB]

"Clipping behavior will be cleaner, as the clipped peaks of the output waveform are not modulated by the ripple on the supply rails. Having said that, if your amplifier is clipping regularly you might consider turning it down a bit."

Agree with Douglas Self, for sure.

 

[pablolie]

Personally, I shall never ever again waste a minute of my life trying to hear differences between amps that measure "well enough", which I'd define as load-independent with a SINAD of 90ish+ and good characteristics at lower volumes. And anything over 100W is wasted on me, because I am now a firm believer in the superiority of 2.1 or 2.2 systems in the majority of real world environments.

 

[Ismapics]

Blows the Cambridge CXA81 MKII integrated out of the water and you have $ left for better sources.

 

[Ismapics]

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.

This may have been answered before, but: Why can't AMPS power output and distortion be measured using pink noise?

 

[ryanmh1]

This may have been answered before, but: Why can't AMPS power output and distortion be measured using pink noise?

You want to measure distortion using noise? In the terrible analogies department, that's like measuring optical clarity using a lens covered in dirt.

 

[Esprit]

Here:

If the product has a CE certification, it is not up to the seller to verify whether the product meets all the required parameters and characteristics. He is not even equipped to do so (see also for example the regulation against electromagnetic interference).

 

[rongon]

EddNog said:
...the B100’s triple nested feedback design.

oooh. Do you know something about that? How did you find out it's a triple-nested feedback design? (I hadn't seen that info anywhere else.) Two stages, each with a local NFB loop around them, and all of it with a global NFB loop wrapped around it? Or something different from that?

That could be NFB applied along the lines of the old RCA 50-Watt High Fidelity Amplifier from the later RCA Receiving Tube manuals. That's a four-stage design with a push-pull output stage having feedback from plate to grid of output tubes, then output tubes plates to preceding stage cathodes, and a global loop from speaker output to first stage voltage amplifier. It would be pretty cool if Topping was doing stuff like that with the B100. Or is that old news in contemporary solid state amp design?

 

[EddNog]

oooh. Do you know something about that? How did you find out it's a triple-nested feedback design? (I hadn't seen that info anywhere else.) Two stages, each with a local NFB loop around them, and all of it with a global NFB loop wrapped around it? Or something different from that?

That could be NFB applied along the lines of the old RCA 50-Watt High Fidelity Amplifier from the later RCA Receiving Tube manuals. That's a four-stage design with a push-pull output stage having feedback from plate to grid of output tubes, then output tubes plates to preceding stage cathodes, and a global loop from speaker output to first stage voltage amplifier. It would be pretty cool if Topping was doing stuff like that with the B100. Or is that old news in contemporary solid state amp design?

It’s how they describe this, “T’ang-ku-la,” amplification module. I know nothing beyond that, it’s basically just marketing lingo as far as I am aware.

-Ed

 

[oleg87]

oooh. Do you know something about that? How did you find out it's a triple-nested feedback design? Two stages, each with a local NFB loop around them, and all of it with a global NFB loop wrapped around it? That would be NFB applied along the lines of the old RCA 50-Watt High Fidelity Amplifier from the later RCA Receiving Tube manuals. That's a four-stage design with an output stage having feedback from plate to grid of output tubes, then plate to preceding stage cathodes, and a global loop from speaker output to first stage voltage amplifier. It would be pretty cool if Topping was doing stuff like that with the B100. Or is that old news in contemporary solid state amp design?

I don't follow this stuff closely because frankly I'm quite happy with 0.01% distortion levels, but I know high-order loops are standard in the Hypex/Purifi class D amps because the designer has talked about it. Applying such techniques to linear amplification should be easier.

 

[rongon]

It’s how they describe this, “T’ang-ku-la,” amplification module. I know nothing beyond that, it’s basically just marketing lingo as far as I am aware.

-Ed

Yes, I saw that, replied to a post about it. It's marketing stuff. No real details in it.

 

[rongon]

Something else just popped into my head.
What's the problem with a unity gain speaker level amplifier module if you have a line stage that can drive it?
You're going to need what used to be called a 'control amplifier' anyway. You need input selection, volume control, you might want tone controls, and so on. You might even want something with room correction and gain.
If that 'control amplifier' had +24dB of gain, and the power amplifier has unity gain, the gain apparent to you the listener would be as expected (maybe about 1V rms signal to full power, or something like that).
That would work, wouldn't it?
So this B100 with just a little gain isn't so horrible. Of course, putting another little amplifier stage in front of it is going to increase distortion and noise by at least a tiny bit. I think it's nice to have the flexibility to choose where that 24dB of gain will come in your system. Maybe...

 

[terryforsythe]

Of course, putting another little amplifier stage in front of it is going to increase distortion and noise by at least a tiny bit.

No need for another amplifier stage. The B100 has three gain settings:

Balanced: 0, 10.4dB and 19.9dB

Single Ended: 6dB, 16.4dB and 25.9dB

 

[rongon]

Yes, I know, but increasing the gain on the B100 also increases noise and distortion, but by such an infinitesimal amount that it's not going to be audible, or even worth mentioning.

If you want all those other features then you could pick the low gain option on the B100 and add a 'control amplifier'. I like the flexibility for system building.
I think that's a nice feature, or it could be for me.

 

[tomchris]

It is great to see a new addition to Topping's product line of amplifiers.

Measurements look impressive.

Personally, I prefer the ability to handle very low ohm loads over hunting for lower THD at this order of magnitude. Perhaps, one day in the future, Topping will add a monoblock with low Ohm handling and high current capability. One can only hope.

I can't wait to see what the future holds.

 

[sarieri]

Some of us are getting all cynical and trying to find faults...

Topping is known for its unreliable QC on amps, so there might be a reason for all the critics here. Topping may very well deserve it.

Just some personal experience with Topping here. I have about 5 Topping amps, 2DACs, and few other preamps. Two LA90d dead after one month of use, one ClassD RA3 dead one day after delivery, one Pre90 preamp with popping noise. And guess what does the cooker from Topping say? JohnYoung said “you should avoid Topping since you have no good luck.” lol, hell I will, there are gears out there measures better.

 

[Doodski]

The problem is that NFB destroys the sound, hence why the THX AAA Benchmark AHB2 and the likes sound dead and closed in, a cold sterile tool with no enjoyment.

You apparently do not know how feedback works. It is an algebraic summing process and has nothing to do with deadness, closed in and sterile.

 

[ryanmh1]

It’s how they describe this, “T’ang-ku-la,” amplification module. I know nothing beyond that, it’s basically just marketing lingo as far as I am aware.

It's gobbledygook. There is no such defined term as triple nested feedback. Could mean 1 global loop with three local loops inside that, or something else, like a loop in a loop in a loop. Doesn't matter either, since feedback is feedback, and they've applied it spades, implying potentially more than 2 poles and a complex feedback design. Hard to stabilize. Hopefully the protection engineering is good, too!

 

[rongon]

Right.
Stability is the question.
Put a 10kHz square wave through it and see what comes out.
That would tell us how well the circuit's been stabilized.
Right?

 

[mhardy6647]

These amps' behavior into a particularly reactive load (e.g., a Quad ESL-57 loudspeaker --or connection to most any loudspeaker via a Polk Audio "Cobra Cable", both well documented killers of marginally stable amplifiers) might be interesting to study.