Al, Ralph and OP.
As Al and Ralph know, I have been acutely, maybe obsessively, interested in the OP's Q. I drive a pair of Paradigm S8s (v3) with an ARC VS-115 tube amp. I can't think of a worse match because the S8s have one of the most wacko impedance curve and phase angle plots I have ever seen. Indeed, when I called Paradigm, I was told the S8s were designed and voiced to be driven by a high current, high power SS amp. OTOH, ARC and Paradigm said I could still get away with my VS-115 -- whatever that means.
I was feeling a bit down until I read the review and bench test reports on two ARC tube amps. Stereophile's report on the Ref 150 and Soundstage's report on the VS-115 (my amp). Both use about 12 to 14 dbs of negative feedback, which I will touch on below.
Now here's the interesting point that I always overlooked. Take a look at the graphs that report the FR output results of both amps when driving a standard dummy load. I copied the URL sites of both reports. They dummy loads are pretty similar and may be standardized. Does power output as a function of frequency vary -- yes!. But not as much as I would have thought: +/- 1 db, or so with both amps driving the dummy load off the 8 ohm taps. Could that variance "flavor" the presentation? I think so.
John Atkinson's comments (re the Ref 150) about this phenomenom (sp?) are interesting: "All three taps offer quite a low source impedance for a transformer-coupled design; as a result, the modulation of the amplifier's frequency response, due to the Ohm's Law action between that impedance and that of our standard simulated loudspeaker, was relatively mild. From the 8 ohm tap (fig.1, gray trace), it was ±0.8dB; the 4 ohm tap offered ±0.4dB, the 16 ohm tap ±1dB."
In other words, driving a speaker off the 4 ohm tap resulted in a lower variance of FR as a function of frequency.
Up until now, I had been driving my speakers off the 8 ohm tap. I always thought the presentation was a bit "forward." I think I now know why. The S8's 28 ohm impedance peak at the 2.2K Hz crossover point should have resulted in an exaggerated FR output and I think it did. Even at 1 or 2 db, I think it changed the acoustic presentation.
I recently switched off to the 4 ohm tap. At times I think the presentation is less defined, maybe flatter. But I turned up the gain a bit and I think the detail is all there. Even the bass is flatter, more extended, maybe more honest. I've been using the 8 ohm tap for so long, I need to re-educate my ears. Time will tell.
Perhaps Al and Ralph will chime in here. But I think the reason for my experience may be that the use of negative feedback lowered my amp's output impedance. This may have the effect of making my tube amp perform somewhat "SS-like" when presented with varying impedance curves. Dunno.
Well that's all I have to say, which has been a lot. I look forward to reading Al's and Ralph's responses. The URLs I referred to are posted below.
http://www.soundstagenetwork.com/measurements/amplifiers/arc_vs115/
http://www.soundstagemagazine.com/measurements/test_amplifiers.htm
http://www.stereophile.com/content/audio-research-reference-150-power-amplifier-measurements
http://www.stereophile.com/reference/60/index.html |
Thanks Al. One thing that the reports I cited answered for me was the magnitude of the FR variance. I didn't know how to convert the impedance and phase angle data into useable information.
As you can see from both reports, the ARC Ref 150 and VS-115 use roughly the same amounts of NF and have similar output impedance stats. Further, the amps were tested by different people at difference times which gives me some confidence that the data is reliable.
As I mentioned above, it appears that using the 4 ohm tap should yield a better FR result. I think it just might be doing that. Short of having an acoustician do precise measurements, I'll go by my ears. Just have to get use to a different acoustic presentation off the 4 ohm tap.
This post and others like it provides important information for folks who want to use tube amps. I think the take-a-ways are the White Paper Paradigms present the extreme cases. As we can now see, my tube amp doesn't neatly fit into the Power Paradigm category. The other take-a-way is that folks looking to match tube amps with speakers should be somewhat mindful about the speaker's impedance and phase angle swings and their tube amp's output impedance. Of course, there's no substituting for a serious listen.
My thanks to Al and Ralph for hanging in there with me. |
I copied below John Atkinson's bench tests of the Revel Ultima Salon 2, a speaker that most would agree is a very high end performer. While the impedance and phase angle plots are more "user friendly" than my Paradigm S8s, it is important to note that even the mighty Revel's impedance plots have peaks and valleys. The low being a little less than 4 ohms in the bass region, with a high of 14 ohms in the mid/treble FR range.
So . . . based on the comments above, I would expect that if the Revels were driven by a "pure" Power Paradigm amp with a relatively high output impedance, the acoustic presentation could very well be colored to some extent in the mid/treble FR range. But as Al said, such may not necessarily be the case because many tube amps are configured to have lower output impedances than one might expect, probably because of NF.
In light of the bench test results of the ARC Ref 150 and VS-115, see above, if those amps were used to drive the Revels, I surmise that the acoustic coloration might not be too far off, especially if the 4 ohm taps are used. I don't know if the Ref 150 or VS-115 have enough "oomph" to get the best out of the Revels, but I don't think you would get a headache listening to them.
Btw, I seem to recall that the Revel web site recommends a high current/high output SS amp to get the best results out of the Revels. I understand that the Revels are somewhat inefficient.
I hope our membership reads these OPs because I think there is much to learn. I think this OP and other OPs are asking good questions.
Btw, btw, as I am retraining my ears to get used to listening to my S8s driven off the 4 ohm tap, I don't think the S8s sound all that bad -- maybe even better than before. It's probably all in my head. ;>')
Cheers,
http://www.stereophile.com/content/revel-ultima-salon2-loudspeaker-measurements |
Al and Bombaywalla, it's a shame that only now am I starting to appreciate the import of impedance and phase angles plots. Had I known this before, I would have approached my speaker and amp purchases differently.
As I posted above, even the "mighty" Revel Ultima Salons have roller coaster impedance curves, albeit more tube friendly than my S8s. At least Revel had the curtesy of including in their sales literature advice that their speakers sound best when driven by a high current/high power SS amp.
At this point, I'm standing pat for now. As I mentioned, I switched over to the 4 ohm taps on my ARC VS-115 tube amp. I think the S8s sound less forward because the midrange is toned down. Although the bench test reports I posted on the ARC Ref 150 and VS-115 suggest that using the 4 ohm taps smooth out the FR variations at the amps' outputs, I'm not sure how that translates into speaker output.
My understanding is that the S8s are ruler flat when driven by a solid state amp. Perhaps, using the 4 ohm taps lowers the amp's output impedance (maybe .5+ ohms) enough to smooth out the speaker's actual FR to approach a near zero output impedance of a SS amp. Dunno.
If my wife would let me switch out gear, I'm not sure whether I would flip the speakers or the amp. The S8s are the darlings of the reviewer community, especially because of the Be tweeters and top grade engineering, R&D and QC that goes into their manufacture. The VS-115 also has garnered high grades.
If I was to flip speakers, I probably would opt for the Revel Studio 2s, but would need to also flip for a high quality tube amp, maybe the new Ayre VX-5. Dunno??? |
Charles1dad -- Bombaywalla makes a fair point that many fine speakers have roller coaster impedance and phase angle plots. As I posted above, even the "mighty" Revel Ultima Salon 2's impedance plot is a bit of rock and roll.
OTOH, I think Bombaywalla's comment about me being in "sonic hell" may be a bit extreme. As I also posted above, the ARC Ref 150 and VS-115 amps were bench tested by different labs. Yet the results were consistent. As Ralph and Al explained, the use of NF reduces the output impedance of the amp, thereby making them more amenable to driving speakers with roller coaster impedance and phase angle plots. That is, somewhat SS-like.
Al also made a very important comment that even though the ARC amps FR plots were pretty flat at the speaker input interface, the most important stat is what comes out of the business end of the speaker. In the case of the S8s, they are rated, and have been bench tested to be very flat -- when driven by SS amps.
I surmise that by using the 4 ohm tap on my VS-115, which as Al thinks might reduce the output impedance of the VS-115, the S8's FR output might approach that achieved with a SS amp. Based on my subjective experience, since switching over to the 4 ohm taps, I think the S8s have a tighter bass response and are less forward as compared to using the 8 ohm taps.
BTW, I seem to recall that ARC uses Wilsons to voice their amps.
Bottom line: if I were to switch out my speakers to the Revels, I would probably also switch to a high current/high power SS amp, as recommended by Revel. At least Revel was honest enough to tell the consumer the type of amp that should be used.
P.S. Bombaywalla -- I think it's unfair to imply that the S8s are junk. I have no intention of starting a speaker war, but suffice to say, I think if properly tweaked and set up, they do a very credible job. |
No offence taken Bombaywalla. Actually, in some of my private communications with Al and Ralph, I too mused that it would be a great boon to the hobby if speaker designers could, with the flip of a switch, change the electronic crossover for SS or tube application. It might require an external crossover -- dunno. But I too take great exception when the industry doesn't disclose the simple fact like Revel, that "so and so speaker is designed to be driven by a [___] amp." Further, where are the tube amp manufacturers? No disclosure there either. The modus operandi is plug and play. IMO, bad form. |
Unsound -- fair point. But if nothing else changes, A'gon members who read this post and other like it will realize that it takes more than lip service to think it's an easy task to match speakers and amps.
Not sure this is true, but I surmise that most (not all) of the hi-end speakers on the market have roller coaster impedance and phase angle plots. Al and Ralph have made this point many times -- it's important to understand what the designer had in mind when designing a particular speaker.
As I mentioned above, my S8s are advertised to have a pretty flat FR plot, as confirmed in a number of independent bench tests. And maybe it's true. But as I also mentioned above, it was only after I called Paradigm that I learned for a fact that my speakers love high current/high power SS amps. Almost eats them up for desert.
So the lesson for the rest of us is that until there is more disclosure, caveat emptor is the rule. If one is interested in a particular speaker, unless the sales literature recommends a particular type of amp, call the factory and ask questions until you get a credible answer. Let's face it, speakers and amps can cost a lot of bucks. The smart consumer will ask questions to ensure the pieces fit together.
Cheers and I hope our fellow members will read this post. |
Charles1dad, I'm not an engineer, but try to be a pragmatist. I also stayed at a Holiday Inn last night. LOL :>')
I think others have made the point that amp and speaker designers have to make trade-offs to achieve a certain result at a certain price point. I appreciate that there are a number of "tube friendly" speakers on the market (I am aware of at least 3 or 4 off the top of my head), but I suspect that most are not. Consequently, I surmise that most speaker designers assume that their gear will be driven by high output SS gear and if compromises have to be made, it could be impedance and phase angle attributes.
So maybe the real issue is that while tube amps possess certain sonic attributes that SS amps can only hope to mimic, perhaps if something has to "give," tube amp manufacturers will have to figure out a way to make their products able to do double duty and handle speakers that are designed and voiced to be driven by SS amps. Ralph and Al have made a point that such amps are heavy and costly. Perhaps that will be the design challenge of tube amp manufacturers if they want to compete in the present and future marketplace.
In the meantime, I have tweaked and configured my gear whereby I am better able to enjoy the music without experiencing listener fatigue. I don't know how flat the FRs of my S8s are, but the output is musical. I'm standing down for now with respect to switching out gear. Next time, I'll be a little smarter when putting gear together. We all live and learn. |
IME, there is wisdom in what Ralph says about musicality and distortion. As I mentioned above, after switching output tranny taps, I think the midrange "forwardness" of my speakers has been tamed somewhat and bass seems more honest and extended.
And as far as turning up the volume, I do not detect any break-up or audible distortion. However, to be fair, when the SPL get's really high, my ears hurt. So what's the point of that exercise?? ;>')
As a point of scientific interest, I am curious how my speakers would perform if FR was measured in an anechoic chamber. But as Mapman said, and I agree with him, musicality has a significant subjective element to it.
Heck, even if my rig was ruler flat as proven in an anechoic chamber and my room didn't add any coloration, some recording are so bad, I think I'd prefer listening to pure distortion. LOL. Lately, I've been trying to get my hands on old London LPs. Many sound pretty good. OTOH, I've picked some LP dogs that are so bad, the neighborhood dogs howl. And I don't think that's because of NF or impedance peaks or negative phase angles. Some recorded music just stinks.
Great thread. I thank everyone for their terrific comments. |
Mapman, I'm not sure where you are holding with SS vs tube amps. But I think it's important to restate for emphasis sake that what Al and Ralph have been trying to explain all along is IME accurate.
If you read my posts about the SS/tube amp Paradigms and varying speaker impedances, I think Al was spot on when he said that an important factor, maybe the most important factor, is a tube amp's output impedance. The lower it is, the more "SS-like" it will perform.
I posted bench test reports from Stereophile and Soundstage for the ARC Ref 150 and ARC VS-115 (my amp). I found John Atkinson's comments quite interesting. He said that the FR output performance of the Ref 150 significantly "flattened" when driving a simulated speaker load off the 4 ohm taps as compared to the 8 ohm tap. Atkinson measured the Ref 150's output impedance to be lower off the 4 ohm taps as compared to the 8 ohm taps.
That result is consistent with what AL said about tube amps having low output impedance performing somewhat "SS-like" when driving speakers that were designed and voiced to be driven by a SS amp -- like mine.
Of course the reason the Ref 150 and VS-115 have low output impedance attributes is likely because of NF, which as Ralph says introduces odd ordered harmonics. So in the end, I guess it comes down to design trade-offs, with respect to both speakers and amps.
Bottom line: My Paradigm S8s have an impedance peak of 28 ohms at 2.2K Hz, which corresponds to the midrange/tweeter x-over point. In contrast, there is an impedance "saddle" of 4 ohms at 100 Hz. Paradigm advised me that the S8s were designed and voiced to be driven by a high current/high output SS amp. Ooopps.
I always thought the S8s sounded a bit "forward" when driving them off the VS-115's 8 ohm taps. At Al's and Ralph's suggestion, I tried the 4 ohm taps. I had to get used to the change in presentation, but now I like it better. Less listener fatigue, less forward, slightly crisper bass, and so forth. My anecdotal experience is consistent with what Atkinson predicted when he bench tested the Ref 150.
Not sure what else to say. Seems to me that if one can pick up a great speaker with relatively flat impedance and phase angle plots, the issue of tube versus SS amp compatibility will be largely mooted. But based on my read of reviews on some of the "big boy" speakers out there, like Magicos and Revel Salons, you are looking at speakers that need to be driven by some monster SS amps to sound their best. But don't take my word for it. Pull some of the bench test reviews and see for yourself.
Cheers and thanks again Ralph and Al for your patience and help. |
