Reinventing the Audio Power Amplifier: Mark Levinson No 53

By Amir Majidimehr

Mark Levinson No 53 Power Amplifier

The power amplifier has been around for decades. Countless companies manufacture them at all price points and performance levels. In this crowded field it is hard to stand out. But Mark Levinson No 53 does, literally from its vertical configuration and circuit design point of view.

In a traditional amplifier design, the output transistors are operated in the so called “linear” region where the output of the device is more or less proportional to the input. Unfortunately transistors are not very efficient when operated in this mode and as a result the total power output relative to the input is roughly 50% in practical implementations.

This is more of a problem in the US with our home electrical voltage being 120 as compared to other countries which use 240 volt power. This means the maximum amount of power we have is half as much. For low power electronics this is OK. But our high power stereo amplifiers can starve for power when the low frequency transients arrive in our music. There is only so much power that can be stored inside the power supply of the amplifier after which, the output power must reduce and with it, additional distortion created.

Efficiency also hurts us with respect to modern speaker designs. Today, the trend is toward more attractive speakers in diminutive enclosures. This reduces the footprint of the speaker letting it blend better into our home décor but has the down side that it takes more power to produce the same level of loudness than it used with the boxy designs of yesteryear.

The solution in the professional sound reinforcement industry (e.g. audio for live concert concerts) has been to use a newer class of amplifier with far more efficient design called switching or class D amplifier. The efficiency of switching amplifier comes from its name. The output transistors are not working proportional to the input signal as they are in traditional amplifiers. But rather, they are either on or off. This sharply reduces transistor losses as the device is not being asked to produce and intermediate output level.

But how do you amplify a signal by just jumping from extreme low to extreme high voltage? The answer has two parts. First, in the input stage the analog waveform is converted into a set of pulses whose width depends on level of the input signal. The louder the input, the more packed the square wave pulses and vice versa. The plus train is then used to drive the output transistors in switching mode as they toggle between zero and maximum voltage to get our high efficiency.

Alas, we can’t drive our speaker with that signal or we would destroy it in an instant. The square wave has infinite bandwidth and hence would instantly damage our tweeters. Fortunately we can get back to our original signal if we filter out the switching frequency before sending the output to the speaker. This is easier said than done because the output of the amplifier is high power and hence, we cannot use active filtering techniques (such is used in tone control or EQ). Instead, we have to resort to passive components such as used in speaker crossovers. Without going through the circuit design details, let's say it is challenging to create a passive filter that filters out the switching frequency but leaves the audio band perfectly flat. You see the ramification of this in response anomalies of some class D amplifiers that “ringing” at higher frequencies (frequency response has oscillations to it).

Output filter design becomes simpler if we can push out the switching frequency way higher than the audio band. That way, even a gentle filter that starts to roll off after the audio band will effectively suppress the switching frequency. Laws of physics interfere yet again because increased switching frequency can stress the output transistors and reduce their reliability or cause outright destruction. Designers attempt to balance the two factors and usually wind up with a switching frequency somewhere in the 300 KHz to 400 KHz range.

Another issue with the output filter is that it can interact with your speaker cable and speaker itself resulting in the amplifier response varying based on the load. This is responsible for the negative reputation of switching amplifiers sounding different with different speakers. We can see a great example of this in the simulate loads tested against a major brand class D amplifier by the Stereophile magazine:

Each color represents a different output load. As you see the variations are quite aparent and likely audible.

Harman, the company behind major audio brands such as Crown, Lexicon and Mark Levinson set out to solve the above problem. The charge came from the professional Crown group which a decade back, completely changed the dynamics this design. The solution called for dividing the waveform into “phases” where each transistor carries the load for part of the time. It is like a relay team rather than having a single runner.

The so called “interleaving” method has a great advantage over traditional class D designs as the interleaving factor acts as a multiplier on the switching frequency. As an example an 8-way interleaved output stage run at 500 KHz, acts the same as a class D amplifier running at 4 MHz! This makes filter design much simpler as we have plenty of time to go from conducting the full audio response, say up to 50 KHz, to full reject of the 4 MHz effective signal. Indeed that is how the Mark Levinson No 53 works. Here is a comparison of the distortion prior to filtering for the traditional 2-way interleaving to 8-way method used in Mark Levinson No 53:

Mark Levinson Performance

You can readily see the huge reduction in spurious signals on the 8-way interleaving used by Mark Levinson No 53 on the right.
Inside Mark Levinson 53

Being a “statement” product, the Mark Levinson No 53 sports a traditional linear power supply from the Mark Levinson Reference 532 amplifier. The temptation in class D designs is to use a switching mechanism in the power supply itself feeding the amplifier. While this provides improved efficiency it aggravates a weakness of switching amplifiers which is their very high sensitivity (compared to linear amplifiers) to power supply voltage variations and noise which unfortunately get worse with switching supplies. Copious amount of “negative feedback” can be used to compensate for this but that leads potentially to amplifier instability (and some would claim compromised audio fidelity).

So when we are not trying to build the smallest and lightest amplifiers (a highly desirable theme in professional world but not so much in consumer amps), use of switching power supplies is best avoided. You can see the Mark Levinson No 53 power supply including its massive capacitors to store power in the front of the amplifier in the picture on the right.

Going back to efficiency, the Mark Levinson No 53 definitely provides it despite its use of linear power supply. Here is a nice comparison:

Mark Levinson No 53 Power Efficiency

But How Does it Sound

OK, lots of technical talk but does any of this impact the sound? You may know that there are two schools of thought here. One that says all amplifiers more or less sound the same. The other says the exact opposite with each sounding different like the smell of two different flowers. I won’t take a position in that food fight :). But instead, speak of a much less controversial issue of pure power delivery.

As I noted earlier, with the trend of less of efficient speakers and somewhat limited power available from our wall sockets, the amplifier can run out of steam before your desire for dynamics does. This usually translates into the amplifier sound becoming leaner at higher volumes, together with increased high frequency distortion, and less than impactful bass.

In comparison testing I have done, switching amplifiers using the classic class D configuration always sport incredible low frequency control and power. They beat out linear class AB amplifiers almost regardless of price. What they give up though is high frequency fidelity which I find somewhat harsh. The distortion is highly non-linear and challenging to spot but it is there. The Mark Levinson No 53 is the first switching amplifier I have heard which does not have this compromise. Its bass is amazingly authoritative: tight and powerful. Yet the rest of the response is absolutely neutral and pleasant.

If you have not heard these unique amplifiers, I highly encourage you to come into our showroom for a listen. We have a pair on hand driving our Revel speakers. I am confident that they will improve the sound of your current speakers given the ease with which they can drive any load regardless of how difficult they might be (and many high-end speakers are difficult to drive). We are happy to let you evaluate them with your own system to see the benefits of this technology. Hearing this amplifier was an eye-opener for me. I think it will be for you too.

Further reading: Mark Levinson Design Overview, specifications

Back to Articles