Record breaker: Raphaelite CS30-MKII tube amplifier review and measurements

[amirm]

Nicely done review! Promoted to home page. Nice to see focused set of measurements in the review which makes for easier reading.

 

[Blumlein 88]

Very nice review, nicely done including the history.

I'm sort of not surprised. Results in the past have shown with music 3% (-30 db) or less is usually good. At lower power you've shown these meet that. You can also find this or that music that 3% is audible in some cases. The FR changes with level are a bit less easy to think about though a little like frequency shaped compression as mentioned up thread.

I've mentioned this many times before, but if you liked the sound while owning speakers needing more than 5 watts you could send the signal thru these. To get the frequency effects with level you'd need something similar to your loudspeakers as a load in place of just a resistor. With modeling these days figuring out a few caps and coils with a resistor to get a near mimic wouldn't be very hard. I also wonder how many times these amps are heard driven far too hard even if you have relatively efficient speakers.

 

[MaxBuck]

Another demonstration that "high fidelity" may not be what everyone really is looking for. Great post!

 

[GXAlan]

@John Atkinson @Kal Rubinson

The Stereophile simulated speaker load doesn’t have the low impendence as you get into the treble. It starts rising at 4 kHz and at 20 kHz it’s off the charts.

(My Error: my phone didn’t open the next page. My mistake was only reading page one before page two).

 

[John Atkinson]

@John Atkinson @Kal Rubinson

The Stereophile simulated speaker load doesn’t have the low impendence as you get into the treble. It starts rising at 4 kHz and at 20 kHz it’s off the charts.
View attachment 280814

That was the original load designed by Ken Kantor. I modified it by adding a Zobel network - a resistor and capacitor in series - across the input terminals so that the load is now moderately demanding but not untypical. See the revised circuit diagram and impedance plot at https://www.stereophile.com/content/real-life-measurements-page-2 . The impedance above 5kHz is significantly lower.

John Atkinson
Technical Editor, Stereophile

 

[GXAlan]

That was the original load designed by Ken Kantor. I modified it by adding a Zobel network - a resistor and capacitor in series - across the input terminals so that the load is now moderately demanding but not untypical. See the revised circuit diagram and impedance plot at https://www.stereophile.com/content/real-life-measurements-page-2 . The impedance above 5kHz is significantly lower.

John Atkinson
Technical Editor, Stereophile

Thanks! That’s so weird. On my phone the “view more” / next page button didn’t load! But now I see it! I sheepishly apologize.

 

[tmtomh]

This is one of the best tube amp reviews I’ve ever read. So much specific info and data really showing how the topology and behavior of the amp likely influence the perceived sound. The “inverse loudness curve” behavior seems like such a good match for a lot of what tube fans describe about the sound. And the history stuff is an added bonus. Thanks @GXAlan - great stuff and really informative!

 

[jhaider]

Interesting review, thanks for the history and observations!

I’m curious if the bad old cheap class D amps mimic this amp’s level dependent FR. The Sonic Impact T-amp, a plastic Tripath chip amp that was like $30 in its day, in particular made me think of triodes.

Thanks! That’s so weird. On my phone the “view more” / next page button didn’t load! But now I see it! I sheepishly apologize.

Stereophile’s phone mode is terrible. Unreliable, and even when it works does not let you go to what you actually want to see. The first thing I do when reading something from Stereophile on my phone is scroll down to “Switch to desktop theme.”

 

[eddantes]

One reads this and then compares against something like PA5... and the realization is:

• amps and dacs are hardly material

What does matter?

• speakers matter
• subwoofers matter
• rooms matter

And as a corollary: one should always have DSP/PEQ at ones disposal becuase

• Preferences matter most

 

[Saponetto]

Premise, I am biased... when I see tubes I see the light.
So, all I can say is "what a nice, hefty piece of eye-candy amp!"
Thanks a lot for your review, I appreciated it a lot!

 

[GXAlan]

This is one of the best tube amp reviews I’ve ever read. So much specific info and data really showing how the topology and behavior of the amp likely influence the perceived sound. The “inverse loudness curve” behavior seems like such a good match for a lot of what tube fans describe about the sound. And the history stuff is an added bonus. Thanks @GXAlan - great stuff and really informative!

Thanks! This is my first time with a SET. Since the unit was physically damaged from shipping, I used my “burner” pair of speakers to test it out first (just to make sure I don’t get DC blowing out the speakers). And then the sound was incredible.

Then I did the measurements and saw the FR drop and then sort of tried to figure out why my hearing was so different from what I was measuring and then I stumbled upon the answer.

• amps and dacs are hardly material

What does matter?

• speakers matter
• subwoofers matter
• rooms matter

And as a corollary: one should always have DSP/PEQ at ones disposal becuase

• Prefferences matter most

Agree. You should do the double blind test for fun if you haven’t already.

The thing is, DACS and to a certain extent, amps, are pretty easy to sell/buy/ship. Rooms matter the most, but we typically have the least opportunity to modify this, based upon actual home construction as well as aesthetic considerations.

Speakers and subwoofers make the biggest difference in sound quality whether you are an objective or subjective audiophile, but it’s hard to ship/move speakers on your own.

 

[Kal Rubinson]

I’m curious if the bad old cheap class D amps mimic this amp’s level dependent FR. The Sonic Impact T-amp, a plastic Tripath chip amp that was like $30 in its day, in particular made me think of triodes.

I have one of those in a metal case stowed at the bottom of my closet. If anyone wants to try it, let me know.

 

[ObjectiveSubjectivist]

HiFi was lost long time ago in audiophile world.
Just sad.

 

[GXAlan]

HiFi was lost long time ago in audiophile world.
Just sad.

The other way to look at it is that music reproduction from the 1930’s is surprisingly good. We just focus on ergonomics and speakers.

 

[misureaudio]

I enjoyed reading very much. Thank you!

 

[John Atkinson]

Thanks! That’s so weird. On my phone the “view more” / next page button didn’t load! But now I see it! I sheepishly apologize.

No problem. And as someone else posted today, switching to "desktop theme" on the iPhone works better for anything other than superficial browsing.

John Atkinson
Technical Editor, Stereophile

 

[AndreaT]

Nice review! So much progress with new amps! No nostalghia for the very imperfect amplifying technology of the 1930’s…Very happy listening to Hypex modules.

 

[Blumlein 88]

Interesting review, thanks for the history and observations!

I’m curious if the bad old cheap class D amps mimic this amp’s level dependent FR. The Sonic Impact T-amp, a plastic Tripath chip amp that was like $30 in its day, in particular made me think of triodes.



Stereophile’s phone mode is terrible. Unreliable, and even when it works does not let you go to what you actually want to see. The first thing I do when reading something from Stereophile on my phone is scroll down to “Switch to desktop theme.”

View attachment 280861

Yes, I had a Tripath based Tact amp. It also reminded me somewhat of triode amps when I used it.

 

[audiotron]

Very cool review!

 

[Axo1989]

What am I missing?

22dB is the LOWEST 5W SINAD of any amplifier measured here?

View attachment 280722

I think that's what "record breaker" in the headline refers to.

Looking at the profile more closely, it's 2nd harmonic that dominates and some people argue that's innocuous or even euphonic. The vertical scale on that graph made me take another look, by the time we get to 3rd and higher we are down around -80 dB. And less distortion at lower output levels.

Spoiler: don't peek if you are going to try the comparison

While it turns out I apparently did hear it in the samples in the other thread, it was subtle and didn't sound at all bad.