Classé Delta Mono monoblock power amplifier

Canadian audio manufacturer Classé Audio was founded in 1980 by engineer Dave Reich and entrepreneur/audiophile Mike Viglas. The name "Classé" was a pun on the fact that Reich was a firm believer in an amplifier's output stage operating in class-A, where the output devices never turn off (see sidebar). Though the brand was established with the 25Wpc DR-2, the first review of a Classé amplifier to appear in Stereophile, by Larry Greenhill, was of the later DR-3, in December 1985. No fewer than 22 reviews of Classé products are available in our free online archive. These include the CAM-350 monoblock amplifier, which was reviewed by Brian Damkroger in January 2001 and has been used by Jon Iverson, AVTech's self-described "web monkey," as his long-term reference amplifier.

My own experience of a Classé amplifier dates from March 2011, when I reviewed the CT-M600 and CA-M600 monoblock amplifiers. So positive was that experience that, when Editor Jim Austin asked me if I would be interested in reviewing Classé's new third-generation Delta Mono amplifier ($21,998/pair), I didn't hesitate to say "yes."

First, some history
Classé Audio became part of the Bowers & Wilkins Group in 2001, with Viglas staying on as chairman and the day-to-day running of the company being handled by brand-development VP Dave Nauber. (Reich had joined Theta Digital by that time; Viglas retired in 2010 and passed away in 2018.) In May 2016, B&W was acquired by EVA Automation; a year later, the new owner shuttered Classé's operations in Montreal and all employees, excluding then-president Dave Nauber, were let go.

Fortunately, Sound United, whose brands include Denon, Marantz, Polk, Definitive Technology, and Boston Acoustics, then acquired Classé. Sound United resumed Classé's operations under the leadership of Dave Nauber, now brand director.

The third-generation Delta products—stereo and mono amplifiers and a preamplifier—are the first fruits of the new ownership. Their development had started when Classé was still owned by B&W; in a video interview, I asked Nauber how come it had taken so long for these products to come to market.

620classe.soundunited

"After Sound United purchased Classé, we continued to develop the Delta 3 designs but had to move the manufacturing from the B&W factory in China to the Shirakawa Audio Works about 130 miles north of Tokyo, where the higher-end Denon and Marantz products are made. . . . Shirakawa has the most comprehensive manufacturing and testing capability that I've ever seen. Everything from how a board gets made, to how a product gets assembled, to how the packaging is tested—it's beyond anything that we've ever done. However, the parts procurement, all sorts of things had to be moved to Japan in order to be able to build Classé in Shirakawa. As you might imagine, most of the parts that we use are not parts that are used by Denon and Marantz, so the individual certifications were lacking, new parts had to be created in their systems, and so on. So it really took longer than any of us thought it was going to take."

Designing the Delta
I asked Nauber what the design goals had been for the new Delta products (footnote 1). "We wanted to shift from what had been a high-power, high-voltage amplifier to a somewhat lower-power, higher-current amplifier design [in order] to optimize it to drive lower-impedance loudspeakers effortlessly. At the same time, there were a few other things at work. We were building a less-expensive range of products, called the Sigma series, which freed us up to go with the Delta series at a higher price. When you have a discussion with the engineers, and you tell them that you want this series to be much, much better, they ask, "How are we going to do that?" Because if you don't spend a lot more money, while you can get incremental gains, it's hard to do something really significant. So, because we had the Sigma series, it was possible to take the shackles off a bit and spend more money, as long as we could justify the cost."

Romanian engineer Sergiu Ignat, who relocated to Montreal in 2000 and started working for Classé in 2012, joined our conversation. Ignat is the project leader for the new Delta products. I asked him what the implications of shifting to a higher-current philosophy had been.

"We maintained the same VA in the transformer—we had a 2.4kVA transformer in the CA-M600 amplifier—but we lowered the voltage rails. The rating into 8 ohms dropped from 600W to 300W, but we have a lot more current available for low impedances. . . . However, when you reduce the voltage rails, in order to maintain the same energy storage, you need to use more capacitance. We actually increased the capacitance by a factor of four to keep the same amount of energy stored in the power supply."

620classe.Pcb-Mundorf-Cap

Ignat confirmed that the Delta Mono uses expensive Mundorf 4-pole electrolytic capacitors in the power supply—22 of them: "Yes, as they are well-rated for voltage and ESR and other parameters." Nauber pointed out that there is more than half a kilometer of copper wire in the transformer, "which is why it weighs 30lb!"

I asked Ignat about the devices used in the Delta Mono's output stage.

"They are lateral MOSFETs. There are two MOSFETs in a single TO247 package, and we have 16 of them, which would be the equivalent of 32 classic bipolar transistors in terms of Safe Operating Area . . . they come with a higher bias, but also when you parallel so many of them, it brings a good class-A output region. The amplifier delivers 35 watts into 8 ohms in pure class-A (though less, of course, into lower impedances)—that's a benefit you wouldn't have with other devices."

I asked why he chose lateral MOSFETs rather than the bipolar devices that had been used in earlier Classé amplifiers.

"The MOSFETs that we are using are factory-matched, not only to keep a good match between same-sex devices but also for both sexes," Ignat explained. "Lateral MOSFETs are the only devices that come with matched transconductance for P-channel and N-channel devices. With all other MOSFETs and bipolars, the N part usually has a higher gm—higher gain, if you want—while lateral MOSFETS actually match both sexes."

Negative feedback is always a contentious issue in power amplifier design, I ventured.

"We can drastically reduce the THD in the treble towards 20kHz and even past it," replied Ignat. "We use something called TMC (Transitional Miller Compensation). It's not a classical single-pole compensation, it's a bit more complex and it's one of the ingredients of obtaining this lower distortion in the treble. . . . This type of compensation ensures that the circuit maintains the loop gain even past 20kHz. Of course, the tradeoff is that you have to be very careful about stability; . . . you can have some problems in the megahertz region, especially with MOSFETs, where they are prone sometimes to behave erratically.

"Rather than trying to get marginal gains from the prior models through additional polishing, we stepped back and made several major changes that opened the way to a significant step forward. I'd had all sorts of things running through my head. Most prominent was the reduction in stages and the shortening of the signal path compared with the prior models. There is now one common input 6-layer printed circuit board with an improved instrumentation amplifier that feeds the antiphase signals into two identical power-amp pcbs (layout-wise). Adding to this, 0.1%-tolerance gain-setting resistors were used all the way, in some cases combinations of two equal values which give effective 0.07% tolerance. The end result is very good CMRR and cancelation of asymmetries."

620classe.pwerstage

Features
At almost 98lb, the Delta Mono is a heavy piece of kit. It continues its predecessor's styling feature of having a front panel that curves around to form the side panels. The front panel also hosts a large air intake used by the amplifier's ICTunnel active cooling system (see sidebar). The intake's louvers have an airfoil profile for low-noise airflow and are magnetically attached to allow removal for the internal filter to be cleaned. Next to the air intake is a first for a Classé product: a VU meter. Rather than connect the meter circuit to the output stage, it is first galvanically isolated with an optical coupler, then the signal is passed to a 10-bit A/D converter. After the digital data are converted to a logarithmic scale and processed for the appropriate ballistics, they are converted back to analog to drive the meter mechanism.

"It was the 20-minute auto-standby timer required in the European Union that drove us to this kind of implementation, since we needed a way to detect the signal and start a counter," explained Ignat. "The challenge when implementing our real-time VU meter was fitting the ADC's dynamic range to a logarithmic scale. With about 60dB of dynamic range on the meter's scale, we really had to put all 10 bits to good use. We had to process the data fast enough so as to not rob the micro-controller of precious time to execute its main function (protection) and to massage [the data] in such a way that the ballistics don't betray their digital origin (showing 'steps' at the low end or hitting a hard-stop before the other end of scale—a real challenge especially due to the quasi-logarithmic law."

620classe.meter

Why VU ballistics rather than an instantaneous peak power reading, I asked.

"A VU response is preferable in the context of a 'power meter,'" Ignat replied, adding that "a VU will certainly miss some of the peaks in a musical program, but we considered that showing potential clipping more often (as a PPM type would do) may discourage users from listening comfortably to a satisfactory level. . . . While less spectacular than a PPM, the VU-type response was the judicious choice."

The Delta Mono's rear panel features unbalanced and balanced inputs at the top right—a strap is used to short the XLR jack's pins 1 and 3 when the single-ended input is used—next to which is the exhaust for the ICTunnel fan. Two pairs of Furutech Torque Guard terminals are used for connection to the loudspeakers. These have rhodium-plated copper posts, and a clutch mechanism prevents the relatively soft copper from being damaged by overtightening.

620classe.bac

On the top left of the rear panel is the IEC 15A AC connector, a fuse, and a bunch of RJ-45 ports for network and IP control, and 3.5mm jacks for infrared control and DC trigger functions.

The Delta Mono is a powerful amplifier, rated at 300W into 8 ohms, 600W into 4 ohms, and 1kW into 2 ohms. Despite the class-A operation for the first 35 8-ohm watts, the chassis remains cool. Warm air can be felt emerging from the vent on the rear panel. Late at night, when the music stopped, I could just hear the fans operating as a very faint "ssssh." This noise was inaudible with music playing.

620classe.bac2

Listening
Classé provides a high-quality AC cord with the Delta Monos, made by DR Acoustics and fitted with Furutech connectors. However, to avoid making too many changes in the system I used to audition the Delta Monos, I stuck with the AudioQuest Dragon High-Current AC cables I had been using for other recent reviews, these plugged into an AudioQuest Niagara 5000 conditioner.


Footnote 1: A video on the design of the series 3 Delta products can be found here.
COMPANY INFO
Classé, a division of Sound United, LLC
380, rue McArthur Saint-Laurent
Québec H4T 1X8
Canada
ARTICLE CONTENTS

COMMENTS
Bogolu Haranath's picture

Are you ready for a brand new beat? ...... Summer's here and the time is right ...... For dancing in the 'seat' (street) :-) ......

supamark's picture

"Fortunately, Sound United, whose brands include Denon, Marantz, Polk, Definitive Technology, and Boston Acoustics, then acquired Classé."

Sound United has shuttered Boston Acoustics, the brand effectively no longer exists. Per Wiki, "Currently they maintain the brand name, but no engineering is budgeted and there is no future production planned for Boston Acoustics." Y'all should probably stop mentioning Boston Acoustics when talking about Sound United, except to note that they liquidate some of the brands they aquire. Maybe they'll sell the brand name to someone who wants to make (sealed box) speakers... but doubtful.

Ortofan's picture

... active cooling solution, then try the Rotel Michi S5 or M8.

https://www.hifinews.com/content/rotel-michi-p5s5-prepower-amplifier-lab-report

https://www.rotel-inzahlungnahme.de/fileadmin/daten_rotel/Rotel_MICHI_Stereo_01_20_-_Der_Weg_der_Vollendung.pdf

John Atkinson's picture
Ortofan wrote:
If want a power amp with an active cooling solution, then try the Rotel Michi S5 or M8.

Stereophile will be reviewing the Rotel Michi S5 in a fall issue.

John Atkinson
Technical Editor, Stereophile

Bogolu Haranath's picture

It would be interesting to see MF review the Rotel Michi M8 (1080W/8 Ohms and 1800W/4 Ohms) with his Wilson Alexx speakers :-) ........

rom661's picture

Right. I'm auditioning the Classe stereo amp right now. You're right of course. The fact that the Rotel and the Classe have cooling fans makes them equivalent. Also, please note, a Hyundai and a Ferrari both run on petrol so, of course, they are the same. It's nice that you apparently like Rotel. I sold them for years. Your comment is nonsensical.

Ortofan's picture

... chi-chi Mundorf capacitors.

With Rotel, you have to settle for British patented slit-foil, high-efficiency, low-ESR bulk storage capacitors.

Mikk's picture

I always enjoy gear reviews, but can't help feeling that using a few other speakers with this amp would provide greater insight into how it performs, overall.
Given the amps price, and the apparent engineering that's gone into it, i imagine Classe would have hoped it got a more thorough workout too?
Sorry to nitpick an otherwise detailed review!
(Edit- apologies if recent events within Stereophile made further tests not do-able).

John Atkinson's picture
Mikk wrote:
I always enjoy gear reviews, but can't help feeling that using a few other speakers with this amp would provide greater insight into how it performs, overall.

That's a valid point, but it wasn't possible during the time I had the Classe amplifiers in my system.

John Atkinson
Technical Editor, Stereophile

Bogolu Haranath's picture

MF or JA2 could do a follow-up review of the Classe amps with the full-range speakers they have :-) ......

georgehifi's picture

Classic con yet again, manufacturers understating the 8ohm wattage by 27% to make the 4ohm look as though it doubles. and the 2ohm becomes even more of a joke. When will this industry con stop?
Would have had a chance if top and bottom were both N channel fets.

Cheers George

Jim Austin's picture

I would call this "conservatively rated." You call it a con?

There is a common misunderstanding that doubling into half the impedance is a crucial metric. It only matters if you'll be operating the amp at or near its maximum power. What matters is having enough power to drive the load. For power-hungry 4 ohm speakers, it's much better to have a 380W amplifier that outputs 610W into 4 ohms than to have a 200W amplifier that doubles its output into 4 ohms and so provides 400W. For a low-sensitivity 8 ohm speaker, 380 ohms is better than 300.

If I were a manufacturer, and I knew about this widespread misconception--that doubling is important regardless of the output power--then I'd be tempted to under-specify the 8 ohm impedance, too. In any case, underspecifying the power is not a con.

Jim Austin, Editor
Stereophile

Ortofan's picture

... power measurements, is there any plan to include in the reviews some form of transient peak power (aka dynamic headroom) testing, perhaps using a 200 ms burst signal, as was discussed in the following article?

https://www.stereophile.com/asweseeit/489/index.html

The article also suggests that equipment would be measured prior to sending it out for listening tests, in part to cull under-performing (let alone defective) samples. Was this procedure ever implemented?

michelesurdi's picture

could you tell us exactly how to go about level matching?

Bogolu Haranath's picture

They place both components side by side and place a 'spirit level' on top of them and then level match ..... Just kidding :-) .......

John Atkinson's picture
michelesurdi wrote:
could you tell us exactly how to go about level matching?

I use the 1kHz, -20dBFS warble tone on Stereophile's Editor's Choice CD and equalize the RMS output voltages at the loudspeaker terminals to within 0.1dB.

John Atkinson
Technical Editor, Stereophile

michelesurdi's picture

hooboy,what a scream

georgehifi's picture

"I would call this "conservatively rated." You call it a con?"

Ok Jim, then why wasn't the 4ohm figure conservatively rated as well by the manufacturer?? It's almost spot on, out by just 2% on what JA measured?? Not 27%!!!! like their 8ohm figure

A blind man can see what the manufacturers are doing, and it just keeps getting done in their specs, "it's a con", and it's about time you guys called them out on it being reviewers, at least JA puts up the measurements, but his silence is a little muted about it.

You guys used to get yourselves and us sooo excited with those big Krell like amps, because when you tested them they did almost double down to 1ohm!!! let alone 2ohms, with words like you can weld off these speaker outputs!!!! And now with speakers like the Alexia ect ect, this spec is even more important.

I'm seeing even more exaggerated examples of "this con" with Class-D, when/if the 2ohm wattage specs are mentioned

Cheers George

Jim Austin's picture

It is possible that some companies understate their 8 ohm power output in order to make it appear that the output doubles into half the load. But if they do so, it is a marketing decision aimed at people who do not understand what matters and what doesn't. As I wrote above, what matters (among much else) is not doubling but the maximum power an amplifier can deliver to a particular load. If a company understates its power output into 8 ohms in order to appear to double into 4 ohms, then, to people who understand, they are making the product look worse, not better. If it is an attempt at a con, then it is a failed attempt. To those with a correct understanding, they would be making themselves look worse, not better.

Jim Austin, Editor
Stereophile

Bogolu Haranath's picture

Some companies like PS Audio specify their power amp output as minimum power outputs ....... For example, minimum power output into 8 Ohms .... Minimum power output into 4 Ohms etc . .....

Hi-Fi News measures dynamic power outputs in addition to continuous power outputs for the amplifiers ...... In many cases, lot of amplifiers put out more power during dynamic output than during continuous output :-) ......

georgehifi's picture

Whatever you say to protect them Jim, doesn't get around the fact that their own specified 8ohm wattage is grossly underrated and in the same breath their own specified 4ohm wattage is correct. And to me it's done just to make it "look like" it can double down.
And the only way it's exposed is in JA's bench tests which 90% of your readers have no idea what their looking at, they just like to read the reviews.

THEY CAN'T HAVE IT BOTH WAYS

Cheers George

Jim Austin's picture

I'm protecting logic and good sense.

Is power in an amplifier bad or good? Because if power is good, then perpetrating a "con" by "grossly understating" it makes no sense. I'm not expecting you to change your mind, but I do want to make sure that other people reading these comments are not misled.

Jim Austin, Editor
Stereophile

Bogolu Haranath's picture

George is making much ado about nothing :-) ..........

Lars Bo's picture

It seems to me, that EPDRs of 1-2 ohms are not at all uncommon for speakers typically matched with big, drive-it-all power amps (or vice versa).

Bogolu Haranath's picture

Speakers with EPDRs of 1-2 Ohms sound better with amps like the Classe Delta monos :-) .......

georgehifi's picture

"grossly understating" it makes no sense.

Yes understate by 27%!!! on the 8ohm figure. Then by only 2% on the 4ohm figure. (You never see them or others overstate the 8ohm as then will make the 4ohm will look real sad indeed)

I consider that a gross understatement, for a means to baffle/confuse/con/deceive call it what you wish.
It's not right that so many manufactures are doing this now, this is not an isolated indecent, it's only serves one purpose.
I've been calling on this now for a couple of years, it has crept into the manufacturers specs and it's not right.

Cheers George

RoryB's picture

Whether chicken or egg, it is a marketing decision. That's why they are called "nominal" specifications. You're only making yourself look silly.

You got 27% more than you paid for? This is an OUTRAGE! (27% more is right around 2 dB, based on 20 log 1.27. Not enough extra to smoke a tweeter.)

Edit: So my math was wrong. Because power comes from the square of voltage, the actual formula is 20 log 1.127, which comes out to 1.03 dB. Cha-ching. Here, let me turn that down 1 dB for you.

a.wayne's picture

Great measurements there Classe , fantastic except for the 2 ohm current limiting , these measurements puts this Classe into a similiar league to the big Boulder ...

Regards

Lee S.'s picture

JA, A few years ago, you reviewed the Classe CA-M600 mono amps and gave them high marks. You said they were the best sounding amplifiers you had reviewed at the time. Your review immensely influenced my decision to purchase a pair. Best sounding amps I have ever owned. I know it’s been several years, but could you comment on the sound attributes between the CA-M600 and the new Delta monos to the best of your recollection. Thanks, Lee

georgehifi's picture

Once again, and your silly if you can't see it.

"Yes understate by 27%!!! on the 8ohm figure. Then by only 2% on the 4ohm figure to make it look better."

Cheers George

Mike-48's picture

JA, I auditioned a previous Classe amp (CA-2300), and while I thought the sound was wonderful, I could occasionally hear the fan running during quiet passages. Are the fans in these monoblocks quieter?

X