I used DRA Labs' MLSSA system with a calibrated DPA 4006 microphone to measure the Harbeth Super HL5plus XD's behavior in the farfield and an Earthworks QTC-40 mike for the nearfield responses. I measured the speaker without the aluminum-frame grille and performed the primary response measurements on the super-tweeter axis. I measured the speaker's impedance magnitude and phase with Dayton Audio's DATS V2 system, but for consistency with Stereophile's presentation of impedance graphs, I plotted the impedance measurement with the Audio Precision System One software.
My B-weighted estimate of the Super HL5plus's sensitivity was 85.7dB/2.83V/m, which is within experimental error of the specified 86dB. The Super HL5plus's nominal impedance is specified as "6 ohms, easy to drive." The speaker's impedance magnitude (solid trace) is higher than 6 ohms for almost all the audioband, dropping slightly below that value between 270Hz and 430Hz. The minimum magnitude was 5.8 ohms at 362Hz, which does suggest that this speaker is a relatively easy load. However, the equivalent peak dissipation resistance, or EPDR (footnote 1), lies below 4 ohms in three regions: between 73Hz and 147Hz; between 320Hz and 660Hz; and above 6kHz. The minimum values are 3.02 ohms at 93Hz and 2.54 ohms at 456Hz. If used with tube amplifiers, the Super HL5plus will work best with their 4 ohm output transformer taps.
The traces in fig.1 are free from the small discontinuities that would suggest the existence of panel resonances. When I investigated the vibrational behavior of the enclosure with a plastic-tape accelerometer, I found resonant modes at 121Hz and 180Hz on the side and back panels (fig.2) and another, at 230Hz, on the top panel. These modes were all relatively low in level, but all had a relatively low Q (Quality Factor), so they might still be excited by music.
The Super HL5plus's impedance-magnitude plot has a saddle in the bass centered on 29Hz, which will be the tuning frequency of the port on the front baffle and which implies extended low frequencies. As expected from fig.1, the woofer's nearfield response (fig.3, blue trace) has the reflex notch at 29Hz. The output of the port peaks between 20Hz and 60Hz (fig.3, red trace) and the port's upper-frequency rolloff is clean. The complex sum of the woofer and port responses is shown as the black trace below 300Hz in fig.3. The boost in the upper bass is at least partly an artifact of the nearfield measurement technique. However, the Harbeth's reflex alignment appears to be slightly overdamped.
The Harbeth speaker's farfield output averaged across a 30° horizontal window centered on the super-tweeter axis (fig.3, black trace above 300Hz) is impressively even, with small peaks balanced by small reductions in energy. There is a very slight downward slope in the treble, with then a sharp rolloff above 11kHz followed by a return to full level at 16.2kHz and then by a 15dB-high narrowband peak at 26.3kHz, the frequency the primary dome resonance of one of the tweeters. The cursor position in the speaker's horizontal dispersion graph (fig.4) suggests a lack of presence-region energy at extreme off-axis angles. The radiation pattern also narrows above 7kHz, presumably due to the relatively wide baffle.
The Super HL5plus's dispersion in the vertical plane is shown in fig.5, with the off-axis responses normalized to that on the super-tweeter axis. With the 17"-high TonTräger stands, which KM used for his review, this axis is 39" from the floor. A survey performed by Thomas J. Norton in the 1990s found that the typical ear height of a seated listener was 36" from the floor. Fortunately, the Harbeth's response 5° below the super-tweeter axis is virtually identical to that on-axis. A suckout at 3kHz develops 10° above and below the reference, presumably related to the 3.5kHz crossover between the bass/mid driver and the tweeter.
In the time domain, fig.6 shows the Super HL5plus's step response on the super-tweeter axis. The super tweeter is connected in inverted acoustic polarity, and its output arrives first at the microphone. The two lower-frequency drive units, which are connected in positive acoustic polarity, arrive next, that of the tweeter arriving before that of the woofer. The positive-going decay of the super tweeter's step blends with the beginning of the tweeter's step, but the decay of the tweeter's step arrives a little too early to blend smoothly with the start of the woofer's step. This suggests that the optimal listening axis will be between the two tweeters, although the Harbeth manual says to use a stand that optimally brings "ears level with tweeters."
The Super HL5plus's cumulative spectral-decay, or waterfall, plot (fig.7) is dominated by ridges of delayed energy at the frequencies of the top-octave and ultrasonic peaks seen in fig.3. Other than a slight discontinuity at the cursor position in this graph, just below 3.2kHz, the Harbeth's waterfall plot is commendably clean.
I was interested in examining how the Harbeth Super HL5plus XD measured because it is the same size as the 1970s-era Spendor BC1, which I used for a while back then, and features a superficially similar array of drive units. I remember that distant ancestor offering excellent measured behavior for its time, but the Super HL5plus's measurements are even better overall. In addition, its low frequencies will be better-controlled than the BC1's, which were always a touch loosey-goosey.—John Atkinson
 | |||||||||||
|
|
| |||||||||
COMMENTS
Would be interested to read your take on the Graham LS8/1, their updated version of the Spendor BC1, as designed by Derek Hughes. I have found contentment with those (famous last words). I have owned other speakers that do this or that thing better, but when I listen to the LS8/1, I am grateful for the music and don't want anything else.
It is interesting that Ken did not find the cabinet resonances to disturb/alter the sound profile, though it is somewhat unclear from the review ("decay tails were ... off the charts"). I read that to mean the box sound did not deteriorate performance from this subjective perspective. The measurements section implies the resonances may alter this speaker's sound, as does the quote of the the manufacturer himself - "you've got this extremely clean midband and this sort of warm, involving low end, which is ideal for some music.""
I wonder why the reviewer measured the frequency output from the supertweeter, and whether it interfered with the output of the 'regular' tweeter, to create the dip from 10-15kHz? If it was measured on axis, would the response be flat? It seems like it has to be an anomaly considering Harbeth's focus on even frequency response, and not a deliberate design choice, to utilize a tweeter system that would create a significant dip in an area where many users can still hear (10-15kHz), but then a high amount of energy above that, where most buyers of this speaker will not be able to hear... I wonder if Ken noticed this.*
Take it for what it is worth, this review lacks comparison to other speakers, outside of "magical, life force, beauty." Such comparisons are usually in other reviews and are very helpful to the reader who might not have the opportunity to audition this speaker in person against competitors. It also gives no hints at what the sound of the speaker actually is. How does a speaker have "life force"? Maybe that means it does not exhibit dynamic compression. But compared to what? Respectfully this could be more helpful to the reader.
*P.s. is this academic for most readers of this magazine, who will likely experience some hearing loss in this region?
These speakers with an EPDR of 3.02ohms in the bass were a far better speakers to use to review the Audio Research I/50 integrated tube amplifier with last week, than 2 x pairs that had an EPDR's less than 2ohms.
Cheers George
Thanks for great review. The retail price of $7995 is 50% premium over U.K retail price of $5335 in current exchange rate. Please note I deducted the VAT in the calculation because the consumer in U.S. is not subject to it. Just because U.S. consumers are not aware of the facts a U.K. price already includes 20% of VAT (sales tax) importers always charge more to U.S. consumers. Ex) a typical U.S consumer will think this speaker’s U.K. price is $6700 in USD just because the tax system are different between two countries. With shipping and import process, a U.S. consumer is fooled as if they are paying a reasonable price. But, the importer knows about this very well and takes advantages of it. Call it what you think, but I personally am disgusted by it. It’s your money so please spend it however you want. Also, bringing a pair of loudspeakers overseas isn’t easy. So, if you like this brand, you really don’t have a choice. Now you read this, I hope you can get over the fact you paid 50% more just because you live in U.S. and the same importer may charge only 20% premium in other part of Europe. Depending on the exchange rate, sometimes you can pay as much as 80% premium over U.K for this brand for no reason really. So, at least now you can feel happy you are only paying 50% more as of now.
This is very true in regards to BBC or BBC inspired speakers, and also European gear generally -- unfortunately. American retail mark-ups are often absurd, even after accounting for shipping cost, currency fluctuation cushion, customs, etc. It's not uncommon to see a mark-up approaching 100% (over suggested retail abroad) . It appears that in the U.S., distributors/dealers charge what they think they can get away with, whereas in Europe, they charge what they think is fair (there are exceptions of course -- some U.S. dealers are not trying to make a living off hifi, and charge more reasonable prices).
This is true as well when roles are reversed (check out what American gear costs in the UK/Europe...yes it needs to be marked-up, but the spread isn't as absurd as it is in the other direction, when UK/European gear is imported into the U.S.)
People have to decide for themselves if it is worth it to pay the premium. For me, there are plenty of US speaker companies that make great speakers for which I don't have to pay import fees etc..
I agree with both of you. I think it's important to discuss things like this in public when a distributor profits 2x or 3x compared to other distributors over Europe hurting both consumers and manufactures from the lost sale due to the higher price. Like we all said, consumers will decide for themselves. At the end of the day, I wish more people know about it to help making a right decision, and I wish this kind of practice will diminish as time goes.
An 8” mid-woofer crossovered to tweeter at 3500 Hz invites some problems.
Mid-woofer diaphragm diameter is almost twice the wavelength of sound at 3500 Hz.
This results in beaming, starting just below 2000 Hz up to the crossover frequency.
This is clearly seen on polar response graphics.
No loudspeaker design engineer would probably start with a design like this.
But does it sound good? Probably yes. (But can’t do Led Zeppelin, or Yes :)
SHL5 Plus version (tested in Stereophile May, 2015 and that I owned and enjoyed for several years) has more to do with an incredible reduction in cabinet resonance than new crossover components.
The non XD Plus has an abundance of cabinet resonance while the new HD has cabinet resonance that is much lower in amplitude, has a huge decrease bandwidth and decays much more quickly.
Not a bad thing but definitely not text book Harbeth lossy design either.
Herb has been noted to describe speaker sound by noting that drivers sound like the material they are made of. (Pardon me if I paraphrased poorly.)
In my experience with Harbeth speakers I have found that to be stone cold true, but it doesn't seem to get mentioned much.
Thanks for a great read, by the way!
___
I like this part: "Good stuff, it makes me happy, but satisfaction is fleeting and soon I—we—need another new fix. You find yourself exploring again..."
In matters of romance and Hi Fi... ;-D
I am never surprised when I meet someone listening to vintage gear and hear them say, "I had those. Man, I wish I hadn't sold them." There is a season for each speaker, under Heaven...but we know seasons are cyclic. Hi Fi works that way, too.
I have found a great way to stay fresh: I keep all my old stuff and when I get that "New is better" urge, I rotate something I already own and get 99% as much joy, on the cheap.
Harbeth says its drivers are made of a "bespoke plastic compound"(https://harbeth.co.uk/harbeth-about-us-harbeth-technology/). So these speakers sound like bespoke plastic compound? Even if yes, I have no concept of what bespoke plastic compound sounds like. Again with respect to the reviewers, this is not helpful to those reading to understand the sound of these or any speakers. My constructive criticism with this review is it doesn't really answer the question of what the speaker sounds like.
Imagine whacking the material in question, or bending and quickly releasing it, or the sound of mylar snapping in air. I think Herb makes a great point to think about.
Thump a plastic cone versus paper or metal....different materials seem to have an intrinsic sound.
Herb could audiophile-splain it better. Check his columns and reviews, this pops up from time to time.
I think the concept of "speaker drivers, when playing music, sound like what they are made of" is based more on a sort of intuition, a folk notion of "how things would work" more than actual engineering principles.
First, it wouldn't make sense that speaker drivers could produce the vast array of sounds they could, if you were likely to hear the sound of the driver materials overlayed. And I'm not sure there are good engineering/psycho-acoustic reasons to expect it either. Yes in principle you can get audible break up or cone resonances, but they can be controlled to below audibility.
I used to have that intuition, e.g. "metal dome tweeters sound like metal..they do cymbals great, but add a metallic bite to other sounds." But even anecdotally I've heard speakers with metal drivers sound organic, and speakers with paper drivers sound more metallic (which I think has more to do with dispersion/frequency response characteristics).
I have owned many speakers (and still own a variety), that use paper drivers, plastic, metal, carbon fibre, etc. I currently own Thiel 2.7 speakers, all metal drivers, driven by CJ amps, and I detect no metallic character overlayed at all. Even compared to my Spendor S3/5 speakers, or older speakers I have using paper drivers.
I read those articles to describe the sound of enclosures more than drivers, if I remember right, but maybe I don't. It is one thing to say that it "sounds like a paper driver," or "it sounds like a plastic driver," or even, "it sounds like a thin-walled wooden box enclosure," and another to describe what the sound of a paper driver or thin walled box enclosure sounds like. Rambling off topic.
Anyway I was just wondering about the high frequency response and wanting to understand how/if the cabinet resonances impacted the sound to the listener, if noticed at all. On re-read, I suppose they were not noticed
You wrote that the Harbeths had not played satisfactorily with ZZ Top and Led Zepelin. At this point, I would like to point out that you can never exclusively test a loudspeaker or any component and then evaluate it afterwards. When you test, and this applies equally to all situations, you are always testing one component, within an existing system of other components that have an enormous influence. A loudspeaker can only give out what has been put into it. If one tests, one never tests only one component alone, it is always also the sum of the further devices involved, which lead to the result. Only if one would get to the bottom of this loudspeaker in many other systems, a more complete impression would be achieved. This is not to say that your system would not be able to do this, but rather that the combination of the various elements can lead to this impression. I wouldn't be surprised that these speakers could definitely rock in another system.
... high-power solid-state amplifier with the Harbeths.
At UK hi-fi shows, where Alan Shaw can choose the amp, his speakers are driven by a Hegel amp that can supply over 350W continuous and 400W peak power into a 6 ohm load.
For rock music, consider adding a pair of REL S/812 subwoofers.
A full review should utilize different amplifiers to report how the speakers respond. I appreciate people like tube amplifiers, but there is no disputing that tubes alter the response of the speaker. I think he used PrimaLuna, and the measurements of those amplifiers change the response significantly and more than many of the other tube amplifiers this magazine reviews and measures. Best to have a 'fair fight' with a clean solid state amp to describe how the speaker actually sounds.
It is super annoying that the speakers again were measured on the super tweeter axis, whilst they are are tuned to be listened to and measured on the tweeter axis. This went wrong the first time the speakers were measured and now it happens again. Why?
| Loudspeakers Amplification Digital Sources | |
Analog Sources Accessories Featured | |
Music Columns Retired Columns | |
Show Reports | |
Features Latest News Community | |
Resources Subscriptions |
© 2024 Stereophile
