|
|
发表于 2003-6-24 23:26:55
|
显示全部楼层
STEREOPHILE REVIEW:
有助你学英文!!
HeadRoom Supreme headphone amplifier
By John Atkinson, January 1994
"Uhh! What is it?" I was being prodded on the arm. Admittedly it was gentle, almost polite prodding, but prodding it still was, a rude disturbance of the cocoon I had woven around myself in seat 31J of the American Airlines MD-11 winging its way across the North Atlantic. I pushed Pause on the Discman, insensitively not waiting for an opportune cadence in the Brahms Piano Quintet that had been my erstwhile virtual reality.
It was a flight attendant. "You're annoying the other passengers," she said, "Your stereo is way too loud."
I was embarrassed, of course. I was also pissed. I was also intrigued. Could the Brahms leaking from the Sennheiser HD560 Ovation headphones be audible above the sound of air rushing past the fuselage and engines turning kerosene into carbon dioxide and water? I tried listening to the headphones with them in my lap. I tried holding them in front of me about as far away as I was from my all-elbows neighbor. Just a faint rustling noise could be perceived above the 100dB low-frequency ambient background.
My listening time had already been partly eaten into by the new blanket rule forbidding the use of "approved electronic devices" until the airplane was well embarked upon its voyage. I realized what was happening. Rival headphone amplifier manufacturers were trying to abort my forthcoming review of the HeadRoom. And well they'll need to, I thought, as I took advantage of the opportunity by asking the stewardess to bring me another beer. (American's flight attendants could usefully learn from British Airways in the art of keeping passengers well supplied with alcoholic beverages?Boozehounds Fly British!" I must've read somewhere.)
Wait a minute. What other headphone amplifier manufacturers? Especially manufacturers of cute little portable battery-powered headphone amplifiers? Yes, the Melos SHA-1 has been selling very well since Corey Greenberg reviewed it for Stereophile in 1992 (footnote 1), but the bulky tubed Melos is strictly a housebound creature. Grado sells a battery-powered amplifier to go with its HP series of headphones, but the amp costs $750梩oo rich for my blood! And mail-order musician catalogs have always offered pro-audio headphone amplifiers that look like they've escaped from the early days of solid-state. But mobile high-end headphone amplification is a rather deserted play area.
Which brings me to Tyll Hertsens' unique HeadRoom amplifiers.
Inner space
The quest to make the headphone listening experience more equivalent to normal speaker listening is not new. Back in the ླྀs, the late Ben Bauer of CBS Laboratories devoted a lot of study to various equalization, interchannel crossfeed, and time-delay schemes intended to work the trick (footnote 2). Indeed, an article in the November 1962 Audio, authored in part by engineer John Eargle (better known today for his excellent-sounding Delos recordings), described a DIY device based on Bauer's crossfeed circuit. The amplitude responses of the Bauer circuit's two channels and the time behavior when fed a signal to just one channel, calculated using a circuit-analysis program, are shown in fig.1 (footnote 3). The driven channel has a degree of HF lift applied to it, while being crossfed to the undriven channel with an approximately complementary amount of HF cut so that the total power in the treble remains about the same. In addition, the crossfeed signal to the undriven channel is delayed by about 0.4ms below 1kHz.
Fig.1 Calculated stereo frequency response, compensated for insertion loss, of Bauer stereophonic-to-binaural network from JAES, April 1961, with delay to undriven channel also plotted. Input is swept tone input to left channel only (2dB and 0.2ms/vertical div.).
Why should it be necessary to so drastically mutate a stereo signal? When a binaural recording (one made with a dummy-head two-channel microphone) is auditioned on headphones, all the sounds appear to come from outside the listener's head. When a conventional amplitude-stereo recording is listened to via headphones, however, musicians and singers seem to be sitting on a clothesline stretched from ear to ear inside the head. "The instruments form a 'musical hat' on the observer's head," was how Bauer described this phenomenon; his EQ, time-delay, and crossfeed are intended to eliminate it by synthesizing the natural ear-to-ear crosstalk signals that a listener experiences when hearing a sound source from the front. (Sound to the side of one ear is heard by the other ear delayed by up to 0.7ms; it is also rolled-off in the treble due to the acoustic obstacle presented to the sound wave by the head.) This processing is intended to render the headphone listening experience more natural, more comfortable.
Fast-forward a quarter century. Any number of researchers are using Digital Signal Processing (DSP) to work the inverse trick: make listening to conventional stereophonic recordings via a single pair of speakers resemble the binaural experience, with sounds coming at the listener from all directions. (This is also something that was first suggested by Ben Bauer; QSound is the commercial realization of one of these algorithms.) Yet I thought Bauer's stereo-binaural quest had been long forgotten梪ntil I saw an advertisement from a new Montana company, HeadRoom, in the March 1993 Stereophile.
Head room
The HeadRoom dual-channel headphone amplifier can operate conventionally, amplifying its inputs and driving a wide range of headphones of differing impedances and sensitivities. But when its unique Process switch is thrown, it feeds each channel through to the other with an amplitude and time delay that are carefully controlled with frequency. In this manner, it parallels what the obsolete Bauer circuit did: synthesize the around-the-head crossfeed information that a listener hears from frontally placed sound sources.
The HeadRoom's block diagram is shown in fig.2. The input impedance is defined by 100k shunt resistors; these also tie the inputs to ground in case the DC-coupled amplifier is turned on with a source component not connected to its inputs. With the Process disengaged, the signal passes through to the volume control, then the output op-amp, discrete complementary buffer stage, series 8 ohms resistors, and the ? headphone jack. (There is no 3.5mm stereo jack: this amp is meant to drive Arnold-sized cans.) A gain-setting resistor is looped around the output stage. With the Process switched in-circuit, each audio signal passes through Crossfeed EQ and Crossfeed delay networks and is summed into the opposite channel prior to the volume control.
Fig.2 HeadRoom headphone amplifier, block diagram. The numbers refer to the pins on the HeadRoom module. (Click on picture for larger image.)
Construction is to an excellent standard. The entire amplifier is housed in an extruded aluminum case. Two dual-AA battery holders from the English Bulgin company occupy the center-rear of the housing; the circuitry and front-panel hardware are carried on a U-shaped, double-sided printed circuit board that surrounds and is supported by the battery holders. On the right rear is a pcb-mounted AC power jack that takes in 5V DC from the supplied AC adaptor; on the left rear are two pcb-mounted RCA jacks, these recessed inside rear-panel holes so that the RCA connectors are securely supported by the HeadRoom's chassis.
--------------------------------------------------------------------------------
Footnote 1: See CG's review in Vol.15 No.10, October ྘, p.195, and my 1992 " roducts of the Year" feature in Vol.15 No.12, December ྘, p.7.
Footnote 2: See "Stereophonic Earphones and Binaural Loudspeakers," B.B. Bauer, JAES, Vol.9 No.2, April 1961, reprinted on p.373 of Stereophonic Techniques, edited by John Eargle and published in 1986 by the Audio Engineering Society. This excellent anthology also includes a 1977 paper (p.382), "Improving the Stereo Headphone Sound Image" by M.V. Thomas, Ph.D., which implements an actively realized version of the Bauer circuit using LM301A op-amps. (Is this the same Dr. Martin V. Thomas who did amplifier design work for the British Musical Fidelity company in the early ྌs, I wonder?)
Footnote 3: A similar response diagram appeared in the January 1968 issue of what was then plain Hi-Fi News. I only include this information for pedantry's sake, however, as HFN/RR now only keeps a limited supply of very recent back issues. Back before I edited that magazine, its publisher decided to save warehouse space by trashing almost all its stock of back issues. It might just have been the fact that I embarked upon my current career because I was a regular and enthusiastic reader of HFN, then HFN/RR, from 1967 through 1976. Nevertheless, I regarded this act as a tragedy, given the superb quality of its technical articles. There was nothing any of the magazine's editorial staff could do about this historically wrongheaded decision.
The DC supply voltage is smoothed with a 100礔 electrolytic, then fed to an encapsulated module that converts it to the ?V-and-ground supply required by the active circuitry. (This 100mA-capable DC/DC converter is said to be the most expensive single part used in the amplifier.) The front or base of the "U" carries the ? headphone jack, the front-panel switches, and the volume control, the last a conductive-plastic component from Clarostat. The other leg of the "U" carries a handful of resistors, 100礔 supply capacitors, and the heart of the amplifier: a largish, 24-pin module that contains a smaller epoxy-encapsulated board carrying all surface-mount components (fig.2). (This HeadRoom module is available separately at relatively modest cost.)
The amplifier is available in three versions: Standard, Premium, and Supreme, costing $199, $299, and $399, respectively (footnote 4). Apart from the Supreme's front-panel Filter switch, all three look identical, the differences being internal. The Standard's circuit module uses surface-mount LM833 op-amps, whereas those in the Premium and Supreme use the Burr-Brown 2604 op-amp?a truly killer audiophile quality device," according to Hertsens. The more expensive Supreme also uses Caddock 132 metal-film resistors on the main board, while the Standard uses ordinary 1% 0.25W metal-film resistors. The Supreme's filter applies a moderate amount of high-frequency boost to compensate for recordings that sound too dull with the crossfeed processing active.
Up a little, left a little...
I started this review last Spring with one HeadRoom Basic amplifier sample. Before I even began my listening, Tyll Hertsens sent me an upgraded unit, followed by what was the first of six different Supreme samples. What was happening was that, in addition to Tyll's discovering better-sounding parts, his experience of different headphone models was being broadened. It had become apparent to him that his "one size fits all" philosophy regarding the amplifier's Process and Filter parameters was not going to work. The Sennheiser HD580s, with their weighty low frequencies and extended highs, required different equalization from the very smooth but rather dark-sounding Grado HP-1s and ƈs. And the totally in-the-ear Etymotics had different requirements again. The design-optimization process is best described in the words of the indefatigable Tyll:
"This revision activity was catalyzed by the Sennheiser HD580 headphones...Our first reaction was to significantly change the filter. We basically brought the whole high end of the curve up by about 5dB. This brought the highs up close to where they should be, but slightly overemphasized the information just before the notch. The box imaged [well], but the tonality was still hurt by the notch. If we brightened up the box enough in the notch region, the direct-channel component (which is tonally unaffected by the processor) was too bright, and made the HeadRoom sound hard. If we eased off, we started to get the old problem back. What to do?
"We decided to turn down the crossfeed channel a little. The danger is [that, according to theory], the crossfeed level isn't much lower in amplitude than the direct channel. But...we knew that time differences are significantly more important than amplitude differences in correct localization. Therefore, cutting the crossfeed level in half should only slightly affect imaging. But梚nsert trumpet fanfare here, 'Ta-da'梒utting down the crossfeed level will have a dramatic reducing effect on the depth of the comb filter notch...It definitely sounds better tonally."
I congratulate Tyll for daring to depart from theory when his ears tell him the theory is not quite right. The amount of bass-shelf boost applied to the sum information and bass-shelf cut information will still depend on the particular headphones used, however. What Hertsens now does, therefore, is to offer a limited range of customization options. When the customer phones HeadRoom, he or she is either supplied an amplifier whose EQ is optimized for the customer's existing 'phones, or is sold a matched amplifier/headphones package.
Powering up
The HeadRoom's case gets hot after a couple of hours, implying that its circuit has a thirst for juice. I found alkaline bunny batteries worked best in the HeadRoom, giving about two and a half CDs' worth of listening pleasure, after which the sound degenerated into a clipped, crackling roar. NiCads were a pain, four AA cells not even lasting the 73 minutes of Beethoven's Symphony 9梖or long journeys, a little more juice is required. Enter HeadRoom's optional external battery holder, which holds four D cells and plugs into the power jack. With alkalines, this gives a supply good for up to 20 hours. To my surprise, rechargeable Radio Shack NiCad D cells didn't last any longer than their AA cells. (I've since been told that these are a scam, merely mounting an AA NiCad cell inside a D cell's case.)
Listening
I used the various HeadRoom amplifiers Tyll Hertsens sent me with a variety of headphones: Sennheiser HD560 Ovations and the new $349 Sennheiser HD580 Precisions; Etymotic ER4S Ear Canal Phones; Sony MDR-484 ear buds; and Beyerdynamic DT901s and 鯧s. In the comfort of my listening room, I used a Mark Levinson No.31 and Panasonic 3700 DAT recorder (footnote 5) driving either a Krell Reference 64 or a Counterpoint DA-10 (UltraAnalog DAC version) as source. For music on the move, I used a Sony WD-D3 Walkman Pro cassette recorder, an Aiwa HD-S1 portable DAT recorder, or my own Panasonic SL-NP1A portable CD player梖ollowed by a Denon DCP-150 portable CD player on loan from HeadRoom. (CDs were carried in the excellent Laserline case, which holds 12 discs by the center holes. Unlike the cases which hold the discs in clear plastic wallets, the Laserline keeps both top and bottom disc surfaces away from anything that might scratch them.)
I mainly used the Straight Wire LSI Encore minijack-RCA cable provided by HeadRoom to link source components to the HeadRoom amp. This was sufficiently oversized that it wouldn't plug all the way into the Denon, however, so for that player I used a premium OFC cable from Sony.
I first compared the amplifier with the headphone output of the ancient Advent 300 receiver I use in my office system.
--------------------------------------------------------------------------------
Footnote 4: According to a cryptic note Tyll Hertsens posted last October on Compuserve's Consumer Electronics bulletin board, there will also be a Limited Edition HeadRoom Home Amplifier, packaged in an Anvil Case with a Micromega Microdrive CD transport and a Theta Cobalt D/A processor, to give what Hertsens modestly refers to as "the world's best transportable stereo."
Footnote 6: Readers should be warned that the enhanced loudness capability of the HeadRoom makes it even more important to take care about how loudly they listen to headphones. I'm concerned that an entire generation is reaching adulthood with crippled ears, due to their habitual playing of portable tape and CD players at hearing-damage levels throughout their teen years.
Well, used as a simple amplifier, the HeadRoom stomped all over the Advent. The AC-powered receiver's bass sounded lumpy and spongy compared with the little battery-powered amplifier. This was probably due to the Advent not having a true headphone output, instead padding down its speaker-level outputs with series resistors. But it sounded so grainy in the highs compared with the HeadRoom that I ultimately ruled it out of contention.
The Denon CD player, on the other hand, has a surprisingly good headphone output, capable of driving Sennheiser 骜s to moderately high levels without strain. However, it didn't go as satisfyingly loud as the HeadRoom. With levels matched at 1kHz, the CD player sounded rather thin compared with the separate amplifier, lacking both the HeadRoom's bloom and its clarity. Its midbass also lacked a little definition, the HeadRoom sounding considerably more authoritative throughout the bass due to its greater low-frequency extension and power reserves.
Against the mighty Melos SHA-1 I habitually use as a preamplifier, however, the match was more even. The tubed preamp sounded warmer in the lower midrange, the solid-state having more apparent clarity in the treble. Surprisingly, given the 6DJ8/6922 tube's reputation for having a rather bright character, the upper midrange and lower treble of the two amplifiers were very close in tonal quality. Where the Melos walked away with the honors, however, was in the feeling of unrestrained power it gave to low frequencies. Where the HeadRoom scored over the direct output from the CD player was where the Melos went even further. Recorded bass drum gave more of an impression that it was going to smooth out the folds in your cerebellum (footnote 6). But all things considered, the little HeadRoom still gave a very good account of itself in this exalted company.
How about HeadRoom's Process process?
With early versions of the HeadRoom, I couldn't get past the bass boost applied to dual-mono signals to appreciate any changes in spatial perception. With recordings with a strong low-frequency mono content梫irtually all rock recordings, which have the kick drum and bass guitar panned to a central position梩he sound muddied up to the point where I preferred using the unit as a straight amplifier. With the current versions, while a warming-up of central images was still noticeable, this didn't get in the way of other aspects of the sound.
Well, to my surprise, I didn't get the out-of-the-head, binaural-type imaging from stereophonic recordings that I expected from reading both HeadRoom's literature and the original Bauer and Thomas papers. (Neither did ླྀs HFN correspondents who had tried the Bauer circuit, as I later discovered.) What I did get was a more coherent presentation of the soundstage within my head. Off-center images didn't seem to clump at the far-left and -right positions as they did with straight headphone listening. Simple recordings made with a spaced pair of omni mikes lost some of their characteristically unstable center imaging, and far-away soundsources captured by this mike arrangement did appear to come from behind my head.
But this making-more-coherent factor was, at best, rather subtle compared with the tonal changes. Simple A/B-switching on the fly paradoxically made it harder to detect. Switching the Process in or out, then settling in to appreciate a recording's gestalt, was much more effective in revealing its effects. Chronicles, the excellent collection of Stevie Winwood classics (Island 9 25660-2), provided a suitable test circuit for the HeadRoom to negotiate, with its combination of driving bass guitar and bass synth coupled with rather hashy high frequencies.
With the Process bypassed, I was all too quickly annoyed by this disc's treble quality (though the Sennheiser HD-580s (footnote 7) alleviated this considerably compared with the earlier 骈s or Beyerdynamic DT901s and 鯧s). With the Process engaged, there was a satisfying solidity to the music's presentation which made the album much more listenable. Ambient information, too, seemed to be more coherently associated with the sound sources exciting it, giving rise to a greater sense of realism to the image. In this respect, switching in the HeadRoom's Process made its musical presentation more believable, in the sense that the unprocessed Melos's presentation was more tangible than that of the unprocessed HeadRoom (or the same sense that LP is, in broad terms, more tangible than CD). Given the choice between the Melos driving the Sennheisers and the processed HeadRoom, I'd opt for the HeadRoom.
True binaural recordings梥uch as one I made at the final hairpin of the 1992 Montreal Grand Prix, and which I shall include on Stereophile's forthcoming Test CD 3梤eproduced in the Bypass position with a true sense of the sounds taking place outside the head, of the listener being immersed in the acoustic environment. Front-placed images, however, as always appears to be the case when a generic dummy head is used (footnote 8), were perceived as being positioned in my temples rather than in front of me. Switching in the HeadRoom Process appeared to have no effect at all on this recording, even though the recording time and amplitude cues were then effectively doubled.
Overall, my time spent with the HeadRoom amplifiers was very satisfying. I spent a lot more time listening to music on headphones than I had done before梐lways a sign that something good is going on.
Summing up
With some products, to see them is to want to possess them; the HeadRoom headphone amplifier falls into that category. It also sounds superb. Even though I couldn't produce any significant out-of-head imaging with the Process switch engaged, it did sound more natural on nonclassical recordings. Don't bother with the Standard unless you're really strapped for cash. Buy the $299 Premium if you don't need the HF-boosting capability of the Filter; otherwise do what I'm going to do and spring for the $399 Supreme. A remarkably well-thought-out, well-engineered product at an excellent price. Long-distance travel will not be the same again. Highly recommended梘ive that 800 number a call!
--------------------------------------------------------------------------------
Footnote 6: Readers should be warned that the enhanced loudness capability of the HeadRoom makes it even more important to take care about how loudly they listen to headphones. I'm concerned that an entire generation is reaching adulthood with crippled ears, due to their habitual playing of portable tape and CD players at hearing-damage levels throughout their teen years.
Footnote 7: I'm much more impressed with these headphones than Alvin Gold appears to be in this issue's "Industry Update." Yes, they sound very smooth compared with earlier Sennheiser cans. But the lack of aggression is due to their more natural presentation of high frequencies, not to a lack of pace. I accordingly used the 骜 headphones for monitoring during Stereophile's most recent recording activities (see this issue's "As We See It").
Footnote 8: Recent research suggests that this is because the brain is very sensitive to the response-shaping effect of the ear's pinnae. Binaural recordings made with probe microphones actually placed in the listener's ear canals don't suffer from this frontal image collapse梖or that listener only!
HeadRoom Supreme headphone amplifier: Specifications
Sidebar 1: Specifications
Description: Battery or AC-powered portable headphone amplifier available in three versions: Standard, Premium, Supreme. Number of inputs: one pair, on RCA jacks. Number of outputs: one ? stereo headphone jack. Controls: On/Off, Volume, Process On/Bypass, Filter On/Bypass (Supreme only). Input impedance: >50k ohms. Rated input: 0dBU (775mV) Standard; 4dBU (1.24V) Premium & Supreme. Rated output: 22mW into 100 ohms, Standard; 60mW into 100 ohms, Premium & Supreme. Maximum output (<0.02% THD at 1kHz): 400mW into 100 ohms (equivalent to 6.32V RMS). THD: <0.01%, 20Hz-20kHz. IMD: <0.01%, SMPTE/DIN 50Hz/8kHz. S/N ratio: >100dB. Headphone impedance range: 30-600 ohms. Headphone sensitivity range: 80-110dB/mW. Power source: 4 AA cells or 5V DC 700mA AC adaptor (supplied). Battery life: >3 hours alkalines, >1.5 hours NiCads, Standard; >2 hours alkalines, >1 hour NiCads, Premium & Supreme.
Dimensions: 5.25" (134mm) W by 6.2" (158mm) D by 1.1" (28mm) H. Weight: 1 lb (1 lb 4oz with batteries).
Accessories available: Case Logic DM-1 ($19.95) and DM-2 ($24.95) carrying cases; external 4D cell battery holder ($24.95), 12-disc CD wallet ($12.95), Standard cable ($9.95), Premium Straight Wire cable ($39.95), desktop enclosure ($35.95).
Finishes available: any the customer desires, as long as it's black-anodized aluminum.
Serial number of samples reviewed: 930710, none found on most recent samples.
Prices: $199, Standard; $299, Premium; $399, Supreme; upgrades are priced at the difference between what you paid for your current unit and what the new unit costs. The HeadRoom module is also available for $59 (Standard) and $89 Premium. Approximate number of dealers: 1 (direct sales via mail-order). HeadRoom also sells by mail-order portable CD players from Denon and headphones from AKG, Grado, Sennheiser, Beyerdynamic, and Etymotic Research.
Manufacturer: HeadRoom Corporation, 611 North Wallace Avenue, Bozeman, MT 59771. Tel: (800) 828-8184. Fax: (406) 587-9484. Web: www.headphone.com
HeadRoom Supreme headphone amplifier: Measurements
Sidebar 2: Measurements
The HeadRoom amplifier was non-inverting for all control settings, and its input impedance was a little lower than specification at 21k ohms at 1kHz (footnote 1). Channel separation in Bypass mode was better than 70dB at low frequencies, worsening to 37.5dB at 20kHz due to capacitive coupling between the channels, probably between the parallel pcb tracks carrying the signals to the volume control. The HeadRoom's background noise was a little higher than I expected, its S/N ratio measuring about 62dB (unweighted, 22Hz-22kHz, ref. 1V). (An earlier sample of the Supreme offered about 6dB lower noise.) Nevertheless, I was never bothered by audible noise, even with the volume control full up and the CD player in Pause.
The HeadRoom's output impedance was very low, at 0.4 ohms. (Contrary to what the block diagram implies, negative feedback appears to be taken from the socket end of the current-limiting 8 ohm output resistors.) Its maximum voltage gain into a high-impedance load with Process and Filter bypassed was 16.6dB, which, given its measured maximum voltage output into a 150 ohm load of 7.5V, corresponds to a sensitivity of 1.1V. Fig.1 shows the changes in the THD+noise percentage for different RMS output voltages into 150 and 40 ohms. The negative slope of the curve to the left of this graph reveals that the reading is dominated by noise; as the output level increases, the constant level of noise becomes a smaller percentage of the signal. The sharp knee of the traces is when the true distortion starts to rise due to the amplifier running out of the ability to swing sufficient volts. The extra current demanded from the amplifier by the 40 ohm load cuts the clipping voltage (defined as 1% THD+noise) from 7.5V to 4.4V. (These are equivalent to power levels of 375mW and 129mW, respectively.) The latter is still plenty enough to produce deafening sound levels.
Fig.1 HeadRoom Supreme, distortion+noise vs output voltage into 150 ohms, 40 ohms, and 40 ohms in parallel with 0.22礔.
Also shown in fig.1 is a trace produced when I added a 0.22礔 capacitor in parallel with the 40 ohm resistor to see whether the HeadRoom would misbehave with highly capacitive loads. With the exception of a marginally higher distortion level between 2V and 3.6V, this trace exactly overlays the one made with a pure resistive load. Given that this load represents the HeadRoom driving several power amplifiers in parallel at the end of about 700 yards of AudioQuest interconnect cable, this minimal change in clipping behavior would suggest that a good headphone amplifier also makes rather a good line stage (as Melos has discovered).
That the amplifier's intrinsic distortion is very low in level is confirmed by looking at the spectrum of its output when reproducing a low-frequency tone at very high level into 150 ohms (fig.2). The only harmonics popping their heads up above the -100dB level are the second at -95.6dB (0.0015%) and the fourth at -99.3dB (0.001%)梒ompletely inconsequential at these levels, particularly given the sonically benign nature of these two specific harmonics. (As track 21 on Stereophile's Test CD 2 will inform you, even 1% of second harmonic is very hard to hear!) The higher current demand of the 40 ohm load at this 4V level brings up the distortion somewhat (fig.3), the second harmonic rising to -80.5dB (0.01%) and the fourth to -91.8dB (0.0025%). The third and fifth also now make appearances above -100dB [polite applause from the gallery], but I doubt very much if any of this has any impact on subjective quality. "Big volts, little loads梔on'tcha just luv 'em!" says the HeadRoom.
Fig.2 HeadRoom Supreme, spectrum of 50Hz sinewave, DC-1kHz, at 4V into 150 ohms (linear frequency scale). Note that the second harmonic at 100Hz is the highest in level, 95.6dB below the level of the 50Hz fundamental (0.0015%).
Fig.3 HeadRoom Supreme, spectrum of 50Hz sinewave, DC-1kHz, at 4V into 40 ohms (linear frequency scale). Note that the second harmonic has now risen to -80.5dB (0.01%).
This predominantly second-harmonic-plus-noise signature is confirmed by fig.4, which shows the distortion waveform with the amplifier working hard driving 1kHz at 4V into 40 ohms. (I averaged 32 individual measurements to produce this graph, dropping the signal's noise content by 15dB to enable its harmonic content to emerge.)
Fig.4 HeadRoom Supreme, 1kHz waveform at 4V into 40 ohms (top); distortion and noise waveform with fundamental notched out (bottom).
--------------------------------------------------------------------------------
Footnote 1: Unless stated otherwise, all the measurements refer to the HeadRoom Supreme amplifier fitted with the 30k/5600pF EQ circuitry appropriate for the Sennheiser 骜 cans.
Measurements part 2
So, how to examine the HeadRoom's response shaping and crossfeed? In its bypass mode, the amplifier was flat in response up to the mid-treble, with then a gentle ultrasonic rolloff. This can be seen in both the rounding of a 10kHz squarewave's leading edges (fig.5) and in the amplitude response (fig.6). The -3dB point lies at 27kHz, meaning that audio-band response is down 2.2dB at 20kHz. This will be audible as a slight softening of the overall sound梟ot a bad idea with headphones! The effect of the filter is also shown in fig.6; it can be seen to shelve up the mid-treble by a mild 3dB.
Fig.5 HeadRoom Supreme, 10kHz squarewave at 1W into 8 ohms.
Fig.6 HeadRoom Supreme, frequency response at 1V into 100k ohms (bottom trace at 1kHz) and with Filter engaged (top) (right channel dashed, 2dB/vertical div.).
Driving just one channel with Process engaged but looking at the output from both channels shows you the nature of the crossfeed signal. As can be seen from fig.7, the result is not too similar to the Bauer circuit responses (fig.1 in the main text). The driven-channel response is basically the same as in Bypass (fig.6), while the crossfeed signal is nearly 12dB down in level and starts to roll off above 1kHz. (This measurement doesn't reveal the time delay, of course.)
Fig.7 HeadRoom Supreme, frequency response of both channels at 1V into 100k ohms with left channel only driven (right channel dashed, 2dB/vertical div.).
What happens when the HeadRoom is handling signals in both channels? The Audio Precision System One allows you to drive a stereo component with both in-phase and out-of-phase dual-mono sweep signals. The results are shown in fig.8. When a signal is the same in both HeadRoom channels, it is shelved up 2dB below 1kHz, with rolled-off highs and a broad 2dB notch centered on 2kHz. This is for the amplifier customized for use with Sennheiser 骜s; the bass shelf is 1dB higher for the 20k/6800pF version that works best with Beyerdynamics; and 1dB higher still for the 10k/8200pF version for use with Etymotics and as a line-level processor for Stax Lambda Pros. For a signal that is identical but out-of-phase in the two channels, this is equalized in the complementary manner.
Fig.8 HeadRoom Supreme, frequency response at 1V into 100k ohms with the two channels in-phase (top trace at 100Hz) and out-of-phase (bottom at 100Hz) (right channel dashed, 2dB/vertical div.).
Another way of looking at the HeadRoom process's effect is to drive the amplifier first with dual-mono pink noise, then with completely uncorrelated pink noise (ie, the two channels are completely different, not just out-of-phase). Track 15 on Stereophile's Test CD 2 was ideal for this test. When the pink noise was the same in both channels (fig.9), the same sum-channel trace was obtained as in fig.8. (Actually, this test was performed on an earlier version of the HeadRoom; the 2kHz valley is correspondingly about twice as deep.) But when there is no relationship between the signals in the two channels?I>ie, an instrument or voice is panned hard left or hard right梖ig.9 confirms the fig.7 result that each channel is reproduced with a flat response. (The bumps and dips in this curve should be ignored; noise signals only measure truly flat when you can average the power at each frequency over a long period of time.)?B>John Atkinson
Fig.9 HeadRoom Supreme, 1/3-octave frequency response at 1V into 100k ohms using dual-mono pink noise (bottom trace at 1kHz) and completely uncorrelated pink noise (top at 1kHz) from Stereophile Test CD 2 (right channel dashed, 2dB/div.). |
|
|联系我们|有害信息举报：010-60152166 邮箱：zx@jd-bbs.com|手机版|Archiver|黑名单|中国耳机爱好者俱乐部
( 京ICP备09075138号 )
提升卡
变色卡
发表于 2003-6-24 13:45:40
楼主
