Carver Raven 350 Review (Tube Amp)

[staticV3]

The sound is more majestic, more realistic, and the soundstage is larger and more compelling than it would be without this unique, spooky and strange ability.

 

[Deleted member 50971]

MacIntosh as brand is not what it once was ,as so many hifi brands that's just husk of what they really where .

back in the day when they mattered , they where legit .

Hi. Are you saying that their MC275 amp from 50 + years ago was better than the MC275 being produced today?

 

[EJ3]

This is a review and detailed measurements of the Carver Raven 350 tube monoblock amplifier. It is on kind loan from a member and (I think) costs US $4,750.
View attachment 225095

The 350 is not a bad looking amp. It has the same "orange peel" paint job of the previous Carver 275 amplifier I reviewed. At 42 pounds, the unit is pretty heavy. With all the weight in the back, the front handle is kind of useless when it comes to lifting it.
View attachment 225100
There is an XLR input but I believe it is for convenience and performs the same as RCA. I measured the DC resistance of the 4 and 8 ohm terminals (after the measurements) and realized they are both the same! Here I was switching back and forth between them. Anyway, measured DC resistance is 0.47 ohm. The 1-2 ohm tap has a lower DC resistance of about 0.3 ohm.

Unlike the 275 amplifier, the fuses did not blow and the amp survived my sweep tests. Bias is to be set at "80" on the front dial and that was the case when I powered on the unit.

There is a gain control but you are advised to set it to max which is what I did for testing.

There is a flip switch in the front which changes the amount of feedback. I tested the amp in both settings.

Here are the specs:
View attachment 225121

Carver 350 Measurements
Let's start with high-feedback performance using XLR input:
View attachment 225102
There is copious amount of distortion causing SINAD to be dominated by it. At 44.1, it ranks as the second worst amplifier ever tested:
View attachment 225104
It even performed worse than the Carver 275 which had a SINAD of 46. Switching to lower feedback doesn't make things that much worse:
View attachment 225106

Note that gain is reduced and therefore, volume will be different. This makes AB testing tricky. Here is a more detailed FFT showing the small difference between low and high feedback:
View attachment 225115
Notice how distortion is made up of both 2nd and 3rd harmonics so you can't make the argument that it is "2nd harmonic goodness."

As noted, RCA performance is the same as XLR:
View attachment 225107

Power supply 60 Hz noise in both cases causes severe intermodulation higher up in frequency. No amount of grounding impacted that so it is endemic tot he design.

Noise performance is decent for type of amp it is:

View attachment 225108

Intermodulation distortion rears its ugly head again in multitone test:
View attachment 225109

I don't know how anyone could hear more "detail" with such an amp where so much of the music signal will get lost in the distortion "grass."

Frequency response should be flat but it is not:
View attachment 225110
It naturally will have load dependency due to output impedance.

Let's see the power situation as we had serious shortfall in the 275 amp:
View attachment 225111

We don't meet the spec there but come closer with 8 ohm:
View attachment 225112

Back to 4 ohm, here is our 1% THD (double the company spec) max and burst power:
View attachment 225113
We seem to get the same shortfall. There is momentary reservoir though allowing the peak output to shoot way up. This was not always consistent though.

I also tested 2 ohm capability:
View attachment 225114

You are still current limited so no more power.

Changing the test frequency gives us the following power curves:
View attachment 225116

There is some instability at 20 Hz and fairly significant power drop. This points to power supply not having enough capacity as the lower frequency taxes it for longer period. Some drop occurs in every amplifier by the way so it is a matter of how much.

Finally, the amplifier warms up quickly and is pretty stable:
View attachment 225117
Testing occurred after this warm up.

Conclusions
What an upside down the world of audio is. Folks want to pay so much more to get dirtier sound. You want dirty? The Carver 350 gives it to you. Even at 5 watts there is copious amount of harmonic distortion. Power supply mixes with that at such high level that it creates its own spread of distortion. A video must come with these amps that hypnotizes you into thinking you are getting great sound....

I can't recommend the Carver 350 monoblock amplifier.
-----------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

Heck, the internal amplifier to my 1977 ADVENT 300 beat that! (I only use the ADVENT 300 as a preamp anyway, as the pre-outs run into a pair of NAD 2100's [done by Peter at Quirk Audio in their early days]).

 

[Mnyb]

Hi. Are you saying that their MC275 amp from 50 + years ago was better than the MC275 being produced today?

it migth be the same amp , I honestly dont know , but they don,t inovate much and create very expensive stuff nowadays

 

[Deleted member 50971]

it migth be the same amp , I honestly dont know , but they don,t inovate much and create very expensive stuff nowadays

I understand, and I don’t know a whole lot about them, even though their headquarters is only an hour away from me. But I see a lot of expensive gear going out their doors, I don’t think the passion is there anymore.

 

[B&WTube]

Hi, I agree with some of that, but just because someone puts bigger better transformers in, and add some tubes doesn’t mean it’s going to perform any better. As witnessed by the measurements that Amir did, the original ST 70 measured better.

What about my preamp that Bob supposedly turned into a super preamp? Was it necessary? Is it more transparent than mine? Where’s the data to back it up, or is it just part of a sea of online folklore? Here it is in black-and-white, if you’re going to claim bigger better stronger faster show me the specs.

Maybe Bobs preamp measures better than my stock SP-14 but I seriously doubt it, and I seriously doubt it sounds any better. That’s the extent of my knowledge, and does my SP 14 that Roy Mottram built sound any better because I opted for the best capacitors he had to offer, the NOS tubes, The XLR inputs and outputs, and a couple other things that were upgraded? I’ll bet in a double blind test no one was picking mine out over the less expensive version.

Everybody hears differences… When they’re sighted. If you go on Audiogon there is more opinions than there is gear out there, and for the most part it’s all money based… Spend more money get better quality. Untrue, and unless something is engineered properly it could cost $20,000 and sound no better, possibly worse, and possibly measure worse. That’s my two cents, although I get what you’re saying and theoretically it should all work out. Don’t forget that they had tube amps pretty perfected for the most part especially by the 1960s, so decades of design goes out the window with these many of these new pieces in my opinion.

Not to disagree with you entirely, I do get what you’re saying.

Seems like you are making pretty large assumptions leaps by taking a preamp that Roy Mottram designed and are comparing it to a totally different designed preamp, and then extrapolating that to VTA-70 vs original ST-70, which you have not really studied, let alone heard.
This is not some debated topic in the Dynaco world. Literally, everyone who has had many of the Dynaco derivatives recognizes the substantial superiority of the VTA over the original. Even the few who prefer the original, recognize the better clarity, range, and speed of the VTA- they just prefer the sound they are used to.
This puts us with:
All Dynaco Experts: VTA70 is clearly better
ASR: ST70 is better

There is an explanation for this. I don’t know exactly what it is, but if I had to guess (which I prefer not to do) it might have to do with the carrying of higher voltage which raises SINAD, but not enough to negate how much cleaner and more dynamic the actual playing of music is.

 

[Ken Tajalli]

An OTL can have quite high output impedance ? so with most speakers the frequency response will be slightly modified , it might possibly be in such way that it's sounds "richer"

yeah, but mine can handle 6 ohm. choose your speakers wisely.
They are mostly current amps.
Don't be scared of feedback, and they can sound crystal !

 

[solderdude]

Looks like the neutral pin is not connected to a chassis ground, answering my earlier question. I would not take my chances with this amp.

Most likely it is not. This is on purpose. Acc. to Bob it is not needed and double insulated.

 

[DonR]

ASR has the measurements to show us. Who are all of these "Dynaco experts", and what do they have to show us? Or you just talking about someone who has an opinion?

Remember:

View attachment 225657
Jim

In before Pace, Rhythm and Timing are mentioned... or am I too late?

 

[audioholic63]

I agree. I’ve also spent a long time in the dark until sites like this. I’m not an engineer and depended on Hifi shops and magazines for information. Stereophile was all I was using as a source for measurements and they were extremely hard to understand as there was never any context or comparison of the results. Just JA at the ended summing it up with something like “it measures respectably”. With the wealth of information out there now though, I rarely feel too bad for someone not willing to do a little research before making a $9k amp purchase. Hell, I spent 3 days researching the last $200 vacuum I bought. But that’s just me

In the pre-internet world I once spent a year shopping for a tea kettle. Them's were the days...

 

[Tks]

I don't get it either as I know of Bob's reputation for solid state design, not tubes. When I asked him in person why he is designing tubes, he said because he likes it. And later read him saying that there is a lot less competition with tubes than transistors. The people buying it are seemingly doing so based on his reputation as an audio genius designer. How much he sells, I don't know. He is definitely not in the circle of "high-end" audio buyers.

Could then also follow with not being in the circle of "high-price" audio sellers if that's the case.

 

[Ken Tajalli]

Most likely it is not. This is on purpose. Acc. to Bob it is not needed and double insulated.

.... and will prevent earth loop noise.

 

[Holmz]

My dentist is into tube Schiit gear. Admittedly, he considers himself a bit of an audiophile but when questioned on room modes and adjustments he seemed to draw a blank.

So the room is a sort of cavity then?

 

[solderdude]

I would love an explanation of how this works...

It doesn't and it can't.
Of course you want an explanation. The 'trick' is the feedback resistor. It is placed in series with the speaker.
Amir measured 0.3ohm series resistance for the low imp range and 0.47 ohm for the high range.
As an OPT transformer consists of windings of thick copper wire and the difference between the 2 taps is 0.17 ohm the series resistance used is going to be 0.3 - 0.17 = so around 0.13 ohm (my assumption = 0.1ohm).

All current through the speaker (lets assume 8ohm) thus runs through the speaker, amp and the resistor.
Given the output resistance of the amp being 1.7ohm and the 'magic' resistor being 0.1ohm about 6%.
Now lets assume the amp produces 5W = 6.3V over 8ohm. There will be 0.78A. The resistor thus will have a voltage across it of 78mV.
This voltage is going into the feedback loop (which is that great invention).

Now for the dubious claim of 'listening to the room'.
The theory is that a speaker also acts as a microphone. And it does. The speaker works both ways around. This is the exact reason as to why the impedance of a driver seemingly 'rises' b.t.w. A higher impedance with a constant voltage (6.3V) means the resistance is higher which comes from counter EMF from the speaker as it loooooves to vibrate at that frequency. So it generated a 'counter voltage' which is 'generated' by an 8ohm source (woofer coil). 6.3V - 3V (yes assumption but realistic) = 3.3V/8ohm = 0.41A.
But as the speaker receives 6.3 and 'generates' the -3V the current drops from 0.78A to 0.41A.
6.3V/0.41A = impedance rise to 15ohm (real speakers can easily reach higher)

This is where the damping factor plays a role. Consider a tuning fork. You tap it. It loves to ring. The only damping it has is the air. So keeps ringing for long.
Now if you put some cloth over it, it will ring shorter. Put a wet cloth over it much shorter. That is damping.

In speakers there is mechanical damping and also electrical damping. Think electric cars that when releasing the 'gas' deliver current back which 'brakes' the car.

Same with speakers. Here is is not voltage that 'brakes' but current.
Now we already established the speaker acts as a mic (generator) that has a source resistance of 8ohm. It generates 3V and with 8ohm this means 0.375A.
But.. for current the speaker and amp output R are in series. For voltage they are in parallel but damping is current and not voltage.

So Rout being 1.7ohm + 8ohm (generator) the current will not be 0.375A but 3V across 9.7ohm = 0.31A.
This is 1.6dB lower in damping current compared to when driven by a 0.1ohm amplifier (say a SS amp with a poor damping factor).
So there is a little less 'damping' current (0.83 x less) something else is also happening.
Voltage division. As the amp is 1.7ohm out and the speaker imp. varies between 8 and 15 ohm something happens.
To create a voltage of 6.3V in 8ohm there is 0.78A flowing. But also across the internal resistance (1.7ohm) = 1.3V.
This means when the speaker is disconnected we would see a rise from 6.3V to 7.6V.
When the speaker is 8ohm at 1kHz but 15ohm at 50Hz then a different voltage division happens.
The voltage across the speaker will not the 6.3V (which it is at 1kHz) but at 50Hz the amp is loaded by 1.7 ohm + 15 ohm = 16.7ohm.
This means the voltage across the speaker, by voltage division at 50Hz will be 0.9x 7.6V instead of 0.83x 7.6V.
At 50Hz we see 6.8V instead of 6.3 at 1kHz which means 50Hz will be reproduced 0.66dB louder and will be damped less.

So far damping factor (which, with 8 ohm is about DF5) where your average amp is around DF100 or so.

What does this have to do with 'reading the room' and 'acting as a microphone'. The sad part is that the same division also takes place when acting as a microphone (which is the claim).

Assume we send in a pulse at 1kHz. speaker reproduces it and send out a sound wave. A speaker has a certain efficiency. We put in 5W. The speaker is a good 100dB/W so 107dB. This spreads in a pattern and several walls/objects bounce part of it back. This means only a tiny fraction of the that sound bounces back, scattered in time.
This means only a fraction of that sound bounces back and hits the small surface of the driver.
Because of this a very small voltage will be generated (and shifted in phase opposite the stimulus) we are talking about -20dB or so which is 0.1x 6.3V = 0.63V.
In reality this will be much lower and rather a few mV but lets go with the huge 0.63V.
The source resistance of the mic is 8ohm. The output R = 1.7ohm so the 0.63V generates a 65mA current.
That current also flow to the 'magic' resistor of 0.1ohm = 6.5mV.
Remember the generated voltage IRL will be much lower.

So the resistor has a voltage across it from the source 0.31V and is slightly (a few mV) altered by the reflections. Both of these signals are 'added' to the overall feedback as well so in the end nearly nothing of the voltage 'generated' by room reflections actually makes it into the amp which by itself (being a tube amp) has very little open loop gain.

This means, in practice none of the sounds reflecting back on to the speaker (time and phase delayed/spread) will actually be found back into the signal going to the speaker.
So.. BS story.. complete nonsense.
What Bob Carver believers should do is ask him how he envisions this works and what it does to the signal provided to the speaker.
He can prove it by showing the output waveform from the amp when measured in a garden and once in a bathroom.
Then he can show the world how 'well' the room is sensed and what the amp does with it.
No such a thing will ever be attempted by anyone so we'll never know but science says it is complete and utter nonsense... no matter what owners of such amps claim to hear.

 

[Deleted member 50971]

Seems like you are making pretty large assumptions leaps by taking a preamp that Roy Mottram designed and are comparing it to a totally different designed preamp, and then extrapolating that to VTA-70 vs original ST-70, which you have not really studied, let alone heard.
This is not some debated topic in the Dynaco world. Literally, everyone who has had many of the Dynaco derivatives recognizes the substantial superiority of the VTA over the original. Even the few who prefer the original, recognize the better clarity, range, and speed of the VTA- they just prefer the sound they are used to.
This puts us with:
All Dynaco Experts: VTA70 is clearly better
ASR: ST70 is better

There is an explanation for this. I don’t know exactly what it is, but if I had to guess (which I prefer not to do) it might have to do with the carrying of higher voltage which raises SINAD, but not enough to negate how much cleaner and more dynamic the actual playing of music is.

My point is Bob Latino redid both. So is it possible he made each worse? That’s not a dig on him, but the proof is in the pudding. I will guarantee that there are also many other ST70 fans out there crying… why would you mess with affordable perfection. By all accounts it was already a good amplifier, and it just did what it was supposed to do, amplify a signal.

Get one of each right next to each other, put them both on a transparent RCA switcher, level match them and bring in those experts, then blindfold them… then let testing begin. That’s a level playing field, and not subjectively pulling sound out of the air. Basically prove it, because until then on a science based forum nobody’s going to take it seriously.

I’ve debated it enough, have a good rest of your day.

 

[B&WTube]

ASR has the measurements to show us. Who are all of these "Dynaco experts", and what do they have to show us? Or you just talking about someone who has an opinion?

Remember:

View attachment 225657
Jim

Talking about people who have owned them for decades and have owned many iterations. The ST70 is like the Chevy 350- it has a huge following and some people with no knowledge, as well as great circuit designers. It seems like we are here: “I measured Rosie o'donnell and Scarlett Johansson and Rosie is actually more symmetrical, so she is more attractive.” I am gonna say, “I believe your measurements and methodology, however, I think there might be some factors to consider or weigh differently- because I have personally seen them, and I gotta give it to Scarlett and everyone who has seen them kind of agrees.”

Obviously, it is everyone’s right to disregard, what almost all first hand experiences are. I know subjective reviews and first hand experiences are rife with issues, and this forum is great at removing that…at the same time, seems like objective and subjective should have a little better correlation. Especially, when all of the designers I know of agree that objectively great measurements doesn’t mean it sounds good. They battle to achieve both.

 

[Deleted member 50971]

Talking about people who have owned them for decades and have owned many iterations. The ST70 is like the Chevy 350- it has a huge following and some people with no knowledge, as well as great circuit designers. It seems like we are here: “I measured Rosie o'donnell and Scarlett Johansson and Rosie is actually more symmetrical, so she is more attractive.” I am gonna say, “I believe your measurements and methodology, however, I think there might be some factors to consider or weigh differently- because I have personally seen them, and I gotta give it to Scarlett and everyone who has seen them kind of agrees.”

Obviously, it is everyone’s right to disregard, what almost all first hand experiences are. I know subjective reviews and first hand experiences are rife with issues, and this forum is great at removing that…at the same time, seems like objective and subjective should have a little better correlation. Especially, when all of the designers I know of agree that objectively great measurements doesn’t mean it sounds good. They battle to achieve both.

OK I’ll bite lol. You mean all the new designers, or many of them, because all of the old school engineers measured everything, they were competing against each other with real measurements. Why would they go through all those measurements and list their specs if it didn’t mean anything? Contrary to popular belief, everything we hear can be measured, and not only that it was set in stone many moons ago.

That’s all I got to say about that.

 

[PeteL]

MacIntosh as brand is not what it once was ,as so many hifi brands that's just husk of what they really where .

back in the day when they mattered , they where legit .

I never owned one but according to this list compiled by a member, the MC462 output 720 Watts in 4 ohms and achieve a SINAD of 108 dB I don't know but to me it's quite SOTA numbers, making it at Benchmark-Purifi-Hypex levels but with considerably more power. What's not to like? What makes them less legit than the old ones?

Amplifier SINAD list ASR + 3rd parties

List Name,THD+N (5W, 4 ohms),SINAD (dB) (5W, 4 ohms),Power (W) (1% THD+N, 4 ohms),Price (USD),Source Topping LA90 (stereo, bypass, low gain),0.00009%,121,90,800,<a href="https://www.l7audiolab.com/f/topping-la90/">L7 Audio Lab</a>,Suggestions, corrections and more info here: Topping LA90 Discre...

docs.google.com

 

[Deleted member 50971]

I never owned one but according to this list compiled by a member, the MC462 output 720 Watts in 4 ohms and achieve a SINAD of 108 dB I don't know but to me it's quite SOTA numbers, making it at Benchmark-Purifi-Hypex levels but with considerably more power. What's not to like? What makes them less legit than the old ones?

Amplifier SINAD list ASR + 3rd parties

List Name,THD+N (5W, 4 ohms),SINAD (dB) (5W, 4 ohms),Power (W) (1% THD+N, 4 ohms),Price (USD),Source Topping LA90 (stereo, bypass, low gain),0.00009%,121,90,800,<a href="https://www.l7audiolab.com/f/topping-la90/">L7 Audio Lab</a>,Suggestions, corrections and more info here: Topping LA90 Discre...

docs.google.com

I admittedly know nothing about Google docs, although I share one document with a business associate. Is there a link to those documents from ASR, or way to search?

 

[PeteL]

I admittedly know nothing about Google docs, although I share one document with a business associate. Is there a link to those documents from ASR, or way to search?

To be honest I don't know much neither but for me it opens on both safari and chrome so seams you don't need the purely Google environment.
This was linked from this post.

Amplifier SINAD list ASR + 3rd parties

I wanted to compare amplifiers measured elsewhere to the ASR standard measurements: SINAD at 5W in 4 ohms, and power at 1% THD+N in 4 ohms. So I chose some amplifiers that interested me more (let me know if you wish I can add more), read their graphs, converted to SINAD, color coded and created...

www.audiosciencereview.com