[Brian Cheney]

Testing this amplifier has been a highly informative and sobering experience for me.  John Curl, on whose equipment the measurements were made, has tested many switching amps (he had several on hand, including a Tripath output stage) and was not as surprised by the results as I was.

If I had seen the ZR1600 on my lab test bench in 1965 (when I got started) I would have been alarmed.  I would have checked for proper biasing of the output stage and the feedback loop for enough compensation.  The distortion measurements on the ZR1600 were reminiscent of the early 60's.  Here they are:

IM distortion 60/7000Hz, 1W level: 0.5%.
THD 1W/5 kHz: 0.8%
THD 10W 5 kHz: .28%
Power output at clipping, 8 Ohm load, one channel driven: 200W
Power output at clipping, 4 Ohm load, one channel driven, 220W
Dynamic headroom, 8 Ohms: minus 1.76 dB
Dynamic headroom, 4 Ohms: minus 4.5 dB
Power bandwith at any level below clipping: 10Hz to above 40kHz
1 kHz rise time: 1 ns (!!)

1W 1 kHz distortion spectrum: 14 harmonics visible, 2 and 3rd, 4th and 5th about equally high

Noise: considerable, rising steeply above 20 kHz
Waveform fidelity: sine waves at all frequencies thickened with switching noise riding along
1 kHz square wave: 1 cycle of overshoot and large amount of switching noise on top of wave which was greatest at low (1W or less) listening levels

All of these measurements with the exception of the square wave risetime are exceptional poor.  The unit failed to reach its rated power before clipping, and oscillated when doing so.

It is the nature of switching amps that one output device is one while the other is off, leading to a measurement phenomenon that looks rather like crossover distortion.  The amp designer has decided not to filter the switching artifacts from the output and they are a major component thereof.  It is likely he counts on the self-inductance of moving coil tweeters (like the treble horns in pro speakers) to roll off the switching frequency, suggesting the amp might sound better on tweeters with higher self-inductance (most domes, for example) than with ribbon tweeters which lack such a component.

Summary: as John Curl says, it is difficult to know how high the distortion of the amplifier is, since noise is its biggest component.  On some measurements distortion was within the manufacturer's 0.5% THD spec, on others the measurement was in excess of 3% (with 80kHz filter engaged on the distortion meter). By any standard there is a LOT of distortion.

I will contact the manufacturer for a suggested output filter so that the power measurements may be made more accurately.  It is quite possible the waveform clipping we observed was a noise component and not the signal at all.  Special equipment and test procedures may be needed in order to make accurate power and distortion measurements.

On to critical comparative listening.

[JCC]

I was amazed by your measurements. Yesterday you noted that at 200 hours the amp was sounding quite good, leading me to wonder if there are problems relating to measuring this new type of amplifier, which sounds great. One of mine has been listened to by Stan Warren who was also impressed.

Anyway, I am looking forward to your comparative listening tests. It will be interesting to compare the results with moving coil tweeters and ribbons.

I wonder what the measurements would be with a BelCanto or one of the new Audio Research units, since both are Tripath based.

[jackman]

A couple observations:
1.  Seems a bit strange that Brian Cheney suddenly gets "specification happy" when evaluating an audio pruduct.  Isn't this the same guy who doesn't furnish any specifications on the speakers he sells?  Presumably because "specs don't matter"...
2.  He tests the amp with a guy who designs amps for a competing company.
3.  Didn't an amp manufacturer (Ampzilla) provide the design idea for his latest and greatest speaker?  Where are the specifications for that speaker?  I have never seen any spec's, however people say they sound very good.  Wait a minute, sound's like some of the people who have heard the Carver...

Anyway, don't want to cause WWIII, however it seems a bit inconsistent for a guy to post all of these spec's on a product when he doesn't post any frequency response measurements for any of his own speakers.

Fire away,
Jman

PS- maybe he could post some measurements that show how the "hand peeled" caps that are matched with a tollerence of .0000000000001% (because VMPS laughs at the 1% many other companies use) have less distortion or measure better than his standard caps.  Or even better, show scientific proof as to how caps measured to such a close tollerance matters when the drivers that are used are not close to this tollerance.  That would be a more interesting topic.

[Brian Cheney]

Amplifier testing is quite standardized, speaker testing is not.  I posted the measurements in order to check for two things:
1. Is the amplifier working properly?
2. Does it make its specifications?

If you doubt the measurements' veracity make some yourself.  I didn't say whether the measurements correlate with listening quality or not.

The ZR 1600 did not make its power spec, and it oscillated at clipping.
It could be we need to change the test protocol (adding an output filter or using specialized equipment).  The amplifier tested poorly with our hgh quality conventional equipment.

As for our speakers, jackman, you've heard them, so why be concerned about measurements?  They were good enough to win the "Best of CES" award for High End Audio the past two years running.  Now that's a measure of quality you can hold on to!!

[_scotty_]

Testing this amplifier has been a highly informative and sobering experience for me.  John Curl, on whose equipment the measurements were made, has tested many switching amps (he had several on hand, including a Tripath output stage) and was not as surprised by the results as I was.

If I had seen the ZR1600 on my lab test bench in 1965 (when I got started) I would have been alarmed.  I would have checked for proper biasing of the output stage and the feedback loop for enough compensation.  The distortion ...

Hi Brian,thanks for posting the test results. Standard practice for measuring THD in digital amps is to use AES 17 filter specs. This filter offers a sharp cutoff at around 20kHz and minimises the intrusion of noise
into the in band distortion measurement. Tripath recomends a three pole filter to reduce some of the HF noise. This is not the same spec filter as the AES17.Interestingly enough the Tripath
EVAL board based on TA0103A module has considerably better measurements than the ZR1600. Is there any information available
on the deadband time Carver used.The longer the deadband time in ns
the worse the IM measurements will be.

[Brad]

Thanks for the measurements, Big B.

Interesting how the results can correlate or not correlate to the listening experience  

[Brian Cheney]

You should probably ask Carver Pro about these matters.  Based on the square wave risetime (about the best I have ever seen) the "dead time" between output devices should be better than 1 ns.

John Curl has the Tripath demo module and it measures about the same as the ZR1600, according to him.

A sharp 20kHz filter should help, if done in the digital domain.  It would cause havoc sonically if done analog.  The designer has elected not to include such filtering.

I placed the ZR1600 to work in our factory in the "speaker burnin" line, where it ran 24/7 for 10 days playing music at a 2W average level (loud) prior to bench testing.  It got better and better with subs and cone dynamic speakers.  I have not listened to it yet on ribbon speakers.

I bought this amp from Carver Pro based on good web buzz and because I wanted to be able to recommend it to my customers.  We shall see if I can do that or not.

[Zero]

I have always felt that ascertaining satisfaction with any given product is the ultimate desired goal.  It appears as if such a thing has occured for many individuals in regards to this particular product.  

While the test results were indeed intriguing, and may point towards an over-amplification of capabilities, they do not state whether the product is good, or bad.  Apparently, there has been an over-whelming response in favor of some of these new Digital Drive amplifier's.  Perhaps, there is something to this that stats simply cannot tell us.

[JCC]

I just spoke to Stan Warren, and he gave me permission to quote him. Stan as you might know made a Tripath based amp himself. When making this amp he noted that it provided amazing clarity and transparency. He also indicated the same was true with Carver ZR that I sent to him for evaluation.

When Stan measured with traditional equipment and approaches the results were poor. The sound was great. Stan suspects that the superior sound results from the extremely low output impedance, low hysteresis distortion, and the extremely high switching speed of  Tripath.  

Stan also noted that for years the audiophile community has raved about tube transparency, every though distortion was often 2% to 3%. The point that he made to me was that there are often things that we can hear but cannot measure. When something sounds good, we should try to figure out what is happening, rather than trying to explain it with inappropriate, or outdated measurements. In other words, the problem is in the measurements.

Take a look at _Scotty_  post on the AES 17 filter specs and deadband time. I suspect that you will find more about how to measure these amplifiers when you speak to Carver or Tripath. Regardless, the measurements might look poor to you, even though the sound should be impressive.

[warnerwh]

A big Thank You for your time and efforts Brian. With all the talk here and at Audio Asylum I almost bought one.  Specs do matter to some degree to ME. I couldn't live with this amp and be happy even if it sounded like a 20,000 dollar tube amp.  Also amplifier specifications can and do tell us something about the amplifier's sound.  Specs also tell us something about the engineering quality too, at least in this case.

[RussKon]

A big Thank You for your time and efforts Brian. With all the talk here and at Audio Asylum I almost bought one. Specs do matter to some degree to ME. I couldn't live with this amp and be happy even if it sounded like a 20,000 dollar tube amp. Also amplifier specifications can and do tell us something about the amplifier's sound. Specs also tell us something about the engineering quality too, at least in this case.


wow...what a big surprise....the poster who was dissing the whole concept of digital amps in about three other threads says that specs really ARE important!!!...lol...what a joke!!!!!

[warnerwh]

Russkon: Maybe I should clarify it for you. If specs are THAT BAD then they WILL have an audible affect on the sound which could be bad. I assumed they'd be somewhere in the real world, my error. Not too many  audiophile SS amps made in the last 30 years spec as horrible as the Carver Pro amps.  This is clearly unheard of in the audiophile world.  Go check the specs Brian listed on  random SS amps and see if you can find anything close to these, even in the cheapest amp you can find.  
1. Dynamic headroom was measured as MINUS 4.5db into 4 ohms. (spare me)
2. Power output at clipping, 8 Ohm load, one channel driven: 200W
This amp is rated at 300wpc. If you can't hear this difference then you are deaf.
3. 1 kHz rise time: 1 ns (!!)  Anybody HOME in there?

Then again maybe you don't know what the specs mean.
This is of course if these measurements are accurate. If the approaches to these measurements created errors then maybe the results could be much different.  It seems odd that John Curl would not know how to measure these amps properly but this may well be the case as I don't know how much he knows about digital amplifiers. Let's wait and see.  If these specs are correct though the engineering is poor imo. There may be more here than we know but those specs are enough to make me gag, sorry.

[Rob Babcock]

The specs don't look good but I've heard plenty of stuff with stellar measurements that sound like ass.  The specs say something, but they don't tell you jack about the s/q.

That said, it is possible that the methods used to test analog amps may yeild squirrely results measuring a digital one.  I suppose we'll have to wait for B's further reports from JC.

I'm curious to see the measurements, but much more interested in the results Brian gets with his two best test instruments:  his ears.

[RussKon]

warner...

aren't you one of the posters looking forward to "golden ears brian"'s report???

right up until the measurements, brian was reporting that this amp sounded very good.....

mmmmmm...... do you care more about the spec sheet or about the sound that comes out of your speakers?

from your posts....its the spec sheet....do you have the spec sheet of your current amplifier framed and posted in your listening area to show all your guests????

its obvious that all of brian's earlier posts about the sound of this amp have had no effects on you....

[_scotty_]

A big Thank You for your time and efforts Brian. With all the talk here and at Audio Asylum I almost bought one. Specs do matter to some degree to ME. I couldn't live with this amp and be happy even if it sounded like a 20,000 dollar tube amp. Also amplifier specifications can and do tell us something about the amplifier's sound. Specs also tell us something about the engineering quality too, at least in this case.


wow...what a big surprise....the poster who was dissing the whole concept of digi ...

 I have to agree,the specs are important but unless the measurements
done have the correct methodology the specs generated will not reflect what the amp is really doing and will be numbers with no meaning. The fact that the amp oscillated with 8ohm test resistor as a load concerns
me more than any specs generated by the test. Tripath recomends in their application notes that a 2nd order 80kHz LC filter be part of the design implementation. Tripath has obtained good results with Lf=18uH
and Cf=0.22uF for a nominal impedence of 8 ohms. Tripath also recommends that a RC damper  be used after the low-pass filter. No load operation can create significant peaking in the LC filter. which produces
strong resonant currents that can overheat the output MOSFETs and /or other components. I am wondering if the Carver implementation of the Tripath digital technology has these two items critical to sucessful operation as part of the circuit design.

[Rob Babcock]

One question for Brian:  did JC test the power output with the unit bridged to mono?  I am anxious to see if it makes close to rated mono power.  At the price a pair used bridged would still not be too steep, and those who've done this have raved about the improvement (maybe due to the extra headroom?).

Thanks.

[RussKon]

there is still the main question.....why do many people like the way these amps sound?????

[warnerwh]

Russkon:  Could you stay on topic. When you grow up you'll understand why this is best.

[RussKon]

You are talking out your a__.

thank you for your intelligent response.....lol

have you even bothered to read the other threads out there concerning these amps????

you must have because you posted to this thread...

http://www.audiocircle.com/index.php?topic=4782&postdays=0&postorder=asc&start=0

you must have a short memory....

by the way.... you haven't told us what type of frame that you have your current amp's specs in.... is is gold...or is it  a simpler wood frame????

[RussKon]

warner...

i see that you edited your post....

lol

and the question is "when i grow up?"

yeah right......