[berni]

I am a little unclear on what exactly is going on with this Emotiva design.

According to Emotiva, the actual amplifier gain stage and output devices are biased as conventional Class A/B, but the power supply is described as Class H.

The power supply has a 2.5KVA transformer, a 240,000 uF capacitor bank and requires a 20 amp circuit.  It has all the ingredients of and sounds very much like a linear power supply, but there must be some power supply switching operation to be called Class H.

Doesn't seem to be any white paper or technical explanation of this, though, and I don't recall previously seeing a Class A/B amplifier combined with a switching power supply.

Steve

I am not 100% certain, but the Jeff Rowland 625 has this combination.

[rollo]

For many years our Maggie 3As [now retired ] were driven by the Audio research DR250 Servo Mk2 amp. 240w of Triode bliss. A new kid on the block has taken over. With 1000W into 4/Ohms the Arion class "D" hybrid monoblocks have won out in our system. Now drive GT Audio speakers.
    There are many choices for amps with Maggies.  High powered amps seem to give one that ease with headroom. Dynamics are more evident. The jump factor as well. The amp choice is subjective to the synergy of the system as well as the speaker. For us an amp that is not lean, bright or over detailed will work wonders on Maggies.


charles

[josh358]

I am a little unclear on what exactly is going on with this Emotiva design.

According to Emotiva, the actual amplifier gain stage and output devices are biased as conventional Class A/B, but the power supply is described as Class H.

The power supply has a 2.5KVA transformer, a 240,000 uF capacitor bank and requires a 20 amp circuit.  It has all the ingredients of and sounds very much like a linear power supply, but there must be some power supply switching operation to be called Class H.

Doesn't seem to be any white paper or technical explanation of this, though, and I don't recall previously seeing a Class A/B amplifier combined with a switching power supply.

Steve

You're more likely to see Class H in sound reinforcement amps. Also, it's commonly used to drive headphones in portable devices. I guess the best known audiophile examples were the Carvers. I don't know what Emotiva has done -- there are various ways of achieving it. There's a lot of information in that thread on some of the approaches that have been used, as well as their pitfalls.

[Davey]

Maybe a clue to the Emotiva configuration is the very large heatsinks they are using.  I wouldn't expect that in a Class-H amp since the dissipation of the devices should be reduced greatly.

If you look inside a Sunfire amp you won't see any massive heatsinking at all.

Cheers,

Dave.

[josh358]

I just did a search to see if I could find some technical details on the XPR's and it came up empty, but I did find this:

"The only comment from a senior person at Emotiva that compared sound quality of the upcoming XPR to the current XPA amps was in response to someone asking if the XPR-1 will have better sound quality than the XPA amps. Lonnie's response was that the XPA-1 will continue to be the best sounding amp in the Emotiva line. Lonnie continued to say that the XPR amps were designed with HT in mind and will offer tremendous dynamic headroom."

http://emotivalounge.proboards.com/index.cgi?board=amps&action=display&thread=20623

[josh358]

Found a bit more, although it refers to the XPR-5:

We have a two rail design in the XPR-5. The low rail, which is what the amp is powered from on idle up to about 50% of the amps voltage swing, and then the high voltage rail with takes it all the way to maximum output. By sitting on the lower rail most of the time, quiescent losses are greatly reduced. The amp runs really cool under normal operation.

But when you need it, the rails instantly switch to the high voltage mode and presto, a beast is born! There is a high speed comparator that looks at the input signal and predicts where the output is going, and when needed, it turns on the high voltage rail BEFORE the output gets there. You can't fool it, even with high frequency transients, as its switching time is faster than the rise time of a 20kHz signal. Sweet!

So, you get the SQ of a really well designed Class AB amp, with a huge efficiency improvement, and no sonic downside.

We've been building Class H amps for a long, long time and we have the control loop down pat. It's more expensive than Class AB, but at a certain power level, it's just the only way to go. You can't justify the power losses on an amp this big.

There is really no down side to a well designed Class H amp. But remember, it's all in the details. You have to start with a great sounding amplifier stage. And the Class H controller has to be done very carefully in order to minimize sonic artifacts. When done correctly, (like in the XPR-5) the result is amazing."

http://emotivalounge.proboards.com/index.cgi?board=amps&action=display&thread=25083

. . . Which leaves me puzzled as to why they're calling it Class H rather than Class G.

[Davey]

Yeah, much more similar to Class G me thinks.  A comparator as a sensing device would just trigger on a threshold and the power supply rails would simply have two levels or "steps."

The Sunfire amps used an analog all-pass filter to create the required delay so the tracking power supply could follow the output waveform very closely.  It was actually a "stepped" power supply also, but with many more than two steps.

Anyways, it appears Emotiva doesn't really understand the difference/distinction between Class G and H. 

Cheers,

Dave.

[Rclark]

Well, for the price there I think more than a few people are going to bite for that kind of headroom. It'll be interesting to see what they're like. I'm going to guess they're probably pretty nice, considering how well received some of their new products are (Stealth 8 monitor received almost ironically absurd pedestal placement last RMAF). These are also the same people who just bought Bob Carver, according to SteveFord.

I know that the XPA-1's had a decent enough rep for SQ, and apparently have a lot of performance left available via modding.

[mojave]

Tommy O'Brien, from the Digital Amp Company circle here at AudioCircle, wrote an article for ProSoundWeb.com called Riding The (Moving) Rails: Detailing Class-G And Class-H Amplifier Topologies. Based on his description, if the high voltage rail in the Emotiva is variable then it is a Class H amplifier. The quote "the high voltage rail with takes it all the way to maximum output" seems to indicate the high voltage rail is variable and supplies power from 50% up to maximum.

[Rclark]

For many years our Maggie 3As [now retired ] were driven by the Audio research DR250 Servo Mk2 amp. 240w of Triode bliss. A new kid on the block has taken over. With 1000W into 4/Ohms the Arion class "D" hybrid monoblocks have won out in our system. Now drive GT Audio speakers.
    There are many choices for amps with Maggies.  High powered amps seem to give one that ease with headroom. Dynamics are more evident. The jump factor as well. The amp choice is subjective to the synergy of the system as well as the speaker. For us an amp that is not lean, bright or over detailed will work wonders on Maggies.


charles

Trung said a lot of nice things about those Arions. He kept telling me I should buy a pair of these and try them instead of buying Ncores. But the 'cores were a known quantity and getting world beater reviews every day at that point last year. 1000 watts though huh? Nice. They do look pretty sweet.

[SteveFord]

It'll be interesting to see what comes of that Emotiva/Bob Carver connection.
I'm hoping that he'll lower the price of his tube amps and he'll bring some of his ideas to the Emotiva brand which will offer great bang for the buck, just like the Carver and Sunfire brands of old.
I'm sure he's learned a thing or two since then.
From the sound of that podcast, it seems like Emotiva is really interested in becoming a major player in the subwoofer arena.

[Rclark]

The only complaints people seemed to have about their latest X-REF subwoofers (built in EQ, pretty snazzy) is their lack of low end extension. I'm sticking with Epik though, that dual opposed design really is magic. I can't have a front firer any more.

[medium jim]

The only complaints people seemed to have about their latest X-REF subwoofers (built in EQ, pretty snazzy) is their lack of low end extension. I'm sticking with Epik though, that dual opposed design really is magic. I can't have a front firer any more.

Did you have the Epik repaired?

Jim

[satie]

Found a bit more, although it refers to the XPR-5:

We have a two rail design in the XPR-5. The low rail, which is what the amp is powered from on idle up to about 50% of the amps voltage swing, and then the high voltage rail with takes it all the way to maximum output. By sitting on the lower rail most of the time, quiescent losses are greatly reduced. The amp runs really cool under normal operation.

But when you need it, the rails instantly switch to the high voltage mode and presto, a beast is born! There is a high speed comparator that looks at the input signal and predicts where the output is going, and when needed, it turns on the high voltage rail BEFORE the output gets there. You can't fool it, even with high frequency transients, as its switching time is faster than the rise time of a 20kHz signal. Sweet!

So, you get the SQ of a really well designed Class AB amp, with a huge efficiency improvement, and no sonic downside.

We've been building Class H amps for a long, long time and we have the control loop down pat. It's more expensive than Class AB, but at a certain power level, it's just the only way to go. You can't justify the power losses on an amp this big.

There is really no down side to a well designed Class H amp. But remember, it's all in the details. You have to start with a great sounding amplifier stage. And the Class H controller has to be done very carefully in order to minimize sonic artifacts. When done correctly, (like in the XPR-5) the result is amazing."

http://emotivalounge.proboards.com/index.cgi?board=amps&action=display&thread=25083

. . . Which leaves me puzzled as to why they're calling it Class H rather than Class G.

This sounds like the NAD circuit. It is essentially a secondary winding on the PS trannie that provides a high voltage rail and the headroom, in the case of NAD there are also additional output devices that are activated when the high rail gets current (IIRC the voltage is always on, but there is no current without activation. The high voltage rail can't take current for prolonged periods and will cause thermal shutdown when it is activated too often or over too long a time. The effect on the rest of the amp's SQ when not operating is minor but noticeable according to folks who worked this circuit in DIY mods to NADs. The extra output devices idling in parallel are the culprit in compromising SQ - however small the impact may be.

Please don't ask me to look up the links, I don't remember the sites anymore.

[mg3720]

Have you had a chance to compare the XPA 1 with the Krell?

I have not done any serious comparison between the XPA-1 and the Krell 400e.  I got the Krells at the same time I got the 20.7s so the speakers had not had any break in time.   Since then, the 20.7s got Mye Stands with 100 pounds of counter weight and are breaking in very well.  I should try a comparison now.

When I was auditioning the Krells, the dealer had McIntosh MC601’s for a comparison, with well broken in 20.1’s.  Within seconds I heard the difference.  The low end response and transients were remarkable with the Krells.

[Rclark]

Did you have the Epik repaired?

Jim

Just waiting on the amp. It's fine, I love running 2.0 in the meanwhile, they put out plenty of bass.

[klao]

I was very much satisfied with the sound I got from the Bryston 7B-SST² monoblocks & Maggie 3.6's.  The 1.7's with the same amps are also nice.

Decided to get a pair of pre-owned Bryston 28B-SST² to drive the 20.7's when my new, larger room (19.5 x 27.5 x 11.5) will be finished in a few months.  One 7B would remain to service, in the future, when CCR & DMW center channel arrrives.  I'm getting the Wyred4Sound small mAMP pair for the 1.7's in rear channels. 

Actually the 7B's might just be enough for the room and speakers, but whenever I attended a concert and heard the symphony orchestra making cresendos to full fortissimo, I always thought extra power in the system can't hurt.

Hope I'm good for the next 10-15 years in terms of amplification & my Maggies. 

[josh358]

The Sunfire amps used an analog all-pass filter to create the required delay so the tracking power supply could follow the output waveform very closely.
Cheers,

Dave.

Clever. I was wondering how he kept he rails from lagging.

[josh358]

This sounds like the NAD circuit. It is essentially a secondary winding on the PS trannie that provides a high voltage rail and the headroom, in the case of NAD there are also additional output devices that are activated when the high rail gets current (IIRC the voltage is always on, but there is no current without activation. The high voltage rail can't take current for prolonged periods and will cause thermal shutdown when it is activated too often or over too long a time. The effect on the rest of the amp's SQ when not operating is minor but noticeable according to folks who worked this circuit in DIY mods to NADs. The extra output devices idling in parallel are the culprit in compromising SQ - however small the impact may be.

Please don't ask me to look up the links, I don't remember the sites anymore.

It's a vivid description, no links necessary (I saw a schematic with a similar topology the other day).

[medium jim]

For many years our Maggie 3As [now retired ] were driven by the Audio research DR250 Servo Mk2 amp. 240w of Triode bliss. A new kid on the block has taken over. With 1000W into 4/Ohms the Arion class "D" hybrid monoblocks have won out in our system. Now drive GT Audio speakers.
    There are many choices for amps with Maggies.  High powered amps seem to give one that ease with headroom. Dynamics are more evident. The jump factor as well. The amp choice is subjective to the synergy of the system as well as the speaker. For us an amp that is not lean, bright or over detailed will work wonders on Maggies.


charles

Charles, are you sure that the Audio Research DR250 is 240 watts per in Triode?  I don't think and octet of 6550's per channel can produce that much power in triode, but can in Ultra Linear, still more than enough power to drive a pair of 3A's, or any Maggie for that matter.

Jim