Concert Experience for $5k?

I would strongly urge you to reflect upon the following paragraph in Stereophile’s measurements of your amplifier:

...allowing for some variation in the calculated output impedance with the level and load impedance used, we’re looking at source impedances of 5.5 ohms (8 ohm tap) and 2.8 ohms (4 ohm tap) in triode mode. While these are both high, they are exceeded by the ultralinear figures: 12 and 6 ohms, respectively! Fortunately, the impedances don’t change much across the audioband, but, as figs.1-3 reveal, there will be a large and audible change in frequency response depending on which loudspeaker is used with the Cary and which transformer tap and mode of operation are used.

The review is at www.stereophile.com/content/cary-cad-280sa-v12-power-amplifier-measurements.

While many of the suggestions made above can be expected to be good ones in most circumstances, unless a speaker being considered is known to have an impedance curve that does not vary significantly as a function of frequency, I would not count on it sounding similar with your amplifier to what others may report based on experiences with other amplifiers.

An audition prior to purchase would seem to be even more essential in this case than in most others.

A speaker such as some of the Zu models, however, which in some cases have very high nominal and minimum impedances (e.g. 16 ohms nominal, 10 ohms min), will minimize that concern. Although that kind of impedance characteristic would introduce a different concern, which is that your amp does not provide a 16 ohm output tap. Using an 8 ohm tap with a 16 ohm speaker may reduce the amp’s power capability and increase its distortion, perhaps not greatly but the degree is hard to predict.

So I would add to Doug’s good suggestion of an 8 ohm speaker having 92 or 93 db sensitivity, or thereabouts, that it would be highly desirable for the speaker to have as little variation in its impedance vs. frequency curve as possible, especially if an audition is not possible. Impedance vs. frequency curves are generally included in Stereophile’s speaker reviews. Given the output impedance characteristics of your amp, as a general rule of thumb frequencies at which speaker impedance is high will receive significantly greater emphasis with your amplifier than frequencies at which speaker impedance is low. And, conversely, frequencies at which speaker impedance is low will receive significantly less emphasis with your amplifier than frequencies at which speaker impedance is high.

Good luck in your search. Regards,
-- Al

Regarding the Legacy Focus 20/20, see its impedance curve here:

www.stereophile.com/content/legacy-audio-focus-2020-loudspeaker-measurements

While the nominal impedance is 4 ohms, the actual impedance is 2 ohms in significant parts of the deep bass region, and varies up to about 9 ohms in the mid-range. For the reasons I stated in my post above, I would definitely not expect this speaker to be a good match for the OP’s amp. And in particular, the interaction of the amp’s very high output impedance with the very low impedance of the speaker at deep bass frequencies will result in very weak deep bass. And based on that impedance curve and on the output impedance characteristics of the amp, I would expect that weakness to probably extend up to as high as around 200 Hz.

Also, John Atkinson provides the following comment in the review:

The Legacy’s sensitivity was slightly lower than specified, but still a very high 94.5dB(B)/2.83V/m—this speaker will play very loudly even with a low-powered amplifier. However, its plot of impedance (fig.1) reveals that it needs to be used with an amplifier that can deliver high current into low impedances. Not only are there two minima in the bass of less than 2 ohms, and another of less than 3 ohms in the mid-treble, but there is an amplifier-crushing combination of 3.3 ohms magnitude and 60 degrees capacitive phase angle at 20Hz. Fortunately, it is rare in music to have high levels of energy this low in frequency. However, a good, beefy solid-state amplifier will probably work better than a tube design.

Regards,
-- Al

P.S: The GoldenEar Triton One which has been suggested by a couple of posters will also not be a good match for your amp, especially if you use the amp’s higher powered ultralinear mode (since as indicated in the measurements I linked to the amp’s ultralinear mode has much higher output impedance, which is referred to by Stereophile as "source impedance," than triode mode). The impedance curve for the Triton One is shown here:

www.stereophile.com/content/goldenear-technology-triton-one-loudspeaker-measurements

In this case, the result of the interaction between the amp’s very high output impedance (especially in ultralinear mode) and the speaker’s variation of impedance over the frequency range would be an under-emphasis of the mid-bass region, and excessive brightness in the upper mid-range.

I have no particular knowledge of the speakers that have been suggested other than the ones I have specifically addressed (Legacy Focus 20/20, some Zu models, Triton One). But again, unless a speaker is known to have a relatively flat impedance curve I would caution against assuming it would sound similar with your amplifier to how it sounds with most other amplifiers, including most other tube amplifiers.

Regards,
-- Al

John (Jmcgrogan2), as you probably realize but others may not, it’s not a matter of horsepower (i.e., watts). It’s a matter of the effects of amplifier output impedance on tonality, resulting from the interaction of that output impedance with the variations of the speaker’s impedance over the frequency range.

BTW, there are of course quite a few speakers having relatively flat impedance curves, as well as relatively high sensitivity, that I would expect to be great matches for this amp. Including the Daedalus speakers I use, and I believe also models from Coincident, Tannoy, and others. But I have no specific suggestions to offer that I feel confident would provide the sonic character the OP is looking for (I am mostly a classical music listener, not a rock concert-goer), and that would also meet his price point.

Best regards,
-- Al

Hi Gdhal,

I’ll begin my answer to your question by referring you to this Wikipedia writeup on voltage divider networks:

https://en.wikipedia.org/wiki/Voltage_divider

In the first figure, consider Z1 to be the amplifier’s output impedance, Z2 to be the speaker’s impedance, Vout to be the voltage seen by the speaker, and Vin to be the voltage the amplifier would be putting out if no load were connected (putting aside the fact that tube amps should not be operated without a load). Vin, in other words, corresponds to the voltage provided to the amplifier’s input at any instant of time multiplied by some gain factor.

If Z1 is significantly greater than zero (i.e., if it is 1 or 2 or several ohms, as is the case for most tube amps), and if Z2 (the speaker’s impedance) varies significantly over the frequency range, then since (as shown in the writeup):

Vout = Vin x (Z2/(Z1 + Z2))

... it can be seen that for a given input voltage to the amplifier the speaker will see a voltage that varies significantly depending on the frequencies which comprise that input voltage. In this situation (Z1 high, Z2 varying significantly as a function of frequency), the result (compared to a situation in which Z1 is negligibly small, as is the case with most solid state amplifiers) will be increased voltage and hence increased emphasis of frequencies for which speaker impedance is high, and decreased voltage and hence decreased emphasis of frequencies at which speaker impedance is low.

In other words, the frequency response of the signal provided to the speaker will be significantly non-flat.

The resulting tonal balance will still be essentially correct, however, if the amplifier’s output impedance is in the area of what the designer of the speaker was anticipating the speaker would be driven with. But since few amplifiers have output impedances as high as the OP’s, especially if its higher powered ultra-linear mode is being used (the main exceptions probably being some very low powered SETs), it is safe to assume that a considerable majority of speakers are not designed with the expectation that they will be driven from such high output impedances.

As I indicated previously, though, if a speaker has a relatively flat impedance curve (in contrast to the Triton One) its tonal balance will vary much less dramatically as a function of amplifier output impedance, and the choice of suitable amplifiers will be much broader, everything else being equal.

I’ll add that none of this necessarily has anything to do with the musical resolution of either the amplifier or the speaker. As I’ve said in a number of past threads, the ability to resolve musical detail and the ability to resolve hardware differences do not necessarily correlate with each other.

I hope that clarifies more than it confuses :-)

Regards,
-- Al