Audio Space Reference 3.1 (300B)
ON TEST
Audio Space
Reference 3.1 (300B)
integrated VALVE amplifier
P
20
Australian
output stage itself. The amplifier we were
loaned for review was the lower-powered
triode version of the amplifier, which uses
300B output valves. The higher-powered
version uses KT-88 output valves.
Newport Test Labs
eter Lau, the founder and chief
designer of Audio Space, apparently once considered emigrating
to Australia, but changed his mind,
for which it would appear that Australia is
the poorer, since a recent survey of Japanese
audiophiles voted Lau one of the ten most
influential living valve amplifier designers,
and his company is reputed to be one of the
three largest valve amplifier manufacturers in
The People’s Republic of China.
In addition to manufacturing under the
Audio Space brand name, Lau’s company
manufactures amplifiers to which more
than ten other companies attach their own
brands. David Kan, of 6moons, says that
Lau’s first two products, the Houston 7 and
Houston 9, which kicked off his company
(and one of which is still in Audio Space’s
product line-up as the AS-9) were: ‘obviously
based on the legendary Marantz 7 and 9.’
Although Peter Lau is based in Hong Kong,
and Audio Space has offices and showrooms
in Kowloon, manufacturing takes place on
Power Output: Single channel driven into
8-ohm and 4-ohm non-inductive loads at
20Hz, 1kHz and 20kHz. [Audio Space]
mainland China at Zhuhai, in Audio Space’s
own wholly-owned factory, known as Corilex
Technology Development.
The Audio Space Reference 3.1 (300B)
reviewed here is available in two different
models, the difference being the valves used
in the output stage and the design of the
The Equipment
The Audio Space Reference 3.1 (300B) is a
physically imposing amplifier. And no wonder. It’s big. It’s 475mm deep, 405mm wide
and 215mm deep. It’s no lightweight either,
at 28.7kg. It’s not for nothing that Audio
Space warns in its manual that unpacking
and installing the amplifier is ‘a job for two
persons.’ It’s also as well to watch your fingers,
because the Audio Space Reference 3.1 (300B)
comes with its four very sharply-pointed feet
already installed. The company supplies four
protective chromed furniture cups to prevent
these sharp spikes from damaging your
furniture, but you need to be careful of your
hands too: I accidentally pricked my finger
during the install and drew blood. There’s
no denying that the Audio Space Reference
Power Output: Both channels driven into
8-ohm and 4-ohm non-inductive loads at
20Hz, 1kHz and 20kHz. [Audio Space]
3.1 (300B) is also a very attractive amplifier. I
don’t know if you can admire the shape of a
valve, but I much prefer the curvy look of a
300B valve to the more stubby, ‘businesslike’
look of a KT88. As you can see from our photographs (which show the Audio Space Reference 3.1 (300B) without the protective valve
cover in place, though one comes standard
with each amplifier) there are six valves visible: four 300Bs, and two more valves which
could be either 6SN7s or 6N8Ps (about which
more in a moment).
The other four valves (all 12AX7s) are
hidden inside spring-loaded black-anodised
aluminium tubes that cover them entirely,
except for a small hole at the top, and have
springs that both push down on the valves
themselves, to ensure they can’t vibrate
and simultaneously stop the casing itself
from rattling. My sample, which came
direct from another reviewer, was not fitted
with the 6SN7s that are listed in Audio
Space’s spec sheet, but instead with two
of the Chinese versions of the same valve:
6N8Ps. Moreover, the two 6N8P valves were
physically different, with one valve larger
than the other. I can only assume that the
reviewer before me was experimenting with
different valve types and forgot to re-install
the original matched pair before returning
the amplifier. No problem, because I was
able to quickly rectify this by replacing both
6N8Ps with vintage 6SN7s from my own NOS
valve stock. (I know that many ‘tube-o-philes’
claim that the 6N8P is a superior valve to
the 6SN7—and they’re interchangeable—
but I wasn’t prepared to spend the time
reviewing an amplifier with two apparently
non-matched 6N8Ps when 6SN7s are also
recommended for use in the amplifier and
I had a matched pair on hand, so I used my
own valves for this review, and replaced the
‘original’ valves before returning the Audio
Space Reference 3.1 (300B) to the Australian
distributor, Audio Space Australia.)
While I was installing my 6NS7s to
replace the 6N8Ps, I noted that the 300B
valve sockets are ‘floating’ rather than hard
mounted. This type of valve mount helps
with longevity, since there’s no chance of
cracking any PCB links when you’re pushing
a valve into a socket that’s hard-mounted. I
was talking to the editor about this during
the review process and he told me he’d
recently had to return a non-working valve
amplifier that had been sent in for review,
and it had had hard-mounted valve sockets.
He had noticed that the PC boards on which
the sockets were fixed flexed alarmingly
whenever a valve was inserted or removed,
and suspected that a cracked PCB trace could
have been the reason for the ‘dead on arrival’
review sample.
The front panel is a familiar design with
a raised centre section and ‘shoulders’ at
either end. The main power switch is rotary,
which is a little unusual, but has obviously
been done to make it visually ‘balance’ with
the rotary volume at the opposite end of
the front panel. Symmetry is also obvious
with the other controls, with the rotary bias
select switch on the left of the combination
bias/output meter counterbalanced by the
source selector switch to the right of the
meter. The two front-panel switches are also
symmetrical. The switch on the left allows
Newport Test Labs
Newport Test Labs
Audio Space Reference 3.1 (300B) Integrated Valve Amplifier
Power Output: Single and both
channels driven into 8-ohm and
4-ohm non-inductive loads at
20Hz, 1kHz and 20kHz. [ASpace]
you to adjust global feedback between ‘high’
and ‘low’, while the one at the right selects a
‘direct’ input that bypasses the input selector
switch and the volume control, effectively
converting the Reference 3.1 from an
integrated amplifier into a power amplifier.
Why the feedback switch? Lau maintains
that although negative feedback improves
damping factor, resolution, soundstage width
and depth, it ‘weakens’ harmonics.
Because of this he says he fits most
of his designs with negative feedback
selectors so his customers get to
ON TEST
choose the traits they prefer, rather than have
him dictate the type of sound they should be
listening to. He also says that having a choice
can also help those with small speakers,
because having minimal negative feedback
will give small speaker a bit more ‘kick’ in the
bass, whereas those who own large speakers
might choose to have high negative feedback
to ‘enable the greatest clarity.’
As you have no doubt already gathered
from the previous paragraph, the centrally
located circular meter does double duty,
serving as both a valve biasing meter and a
power output level meter. When operated as
a power output level meter, it can monitor
either the left channel output or the right
channel output, but not both simultaneously.
Again, this unusual implementation seems
to have been done only for reasons of
Audio Space
Reference 3.1 (300B)
Integrated VALVE Amplifier
Brand: Audio Space
Model: Reference 3.1 (300B)
Category: Integrated Amplifier
RRP: $5,480
Warranty: Two Years (Valves: 6 mnths)
Distributor: Audio Space Australia Pty Ltd
Address: 22A Rooks Road
Nunawading
VIC 3131
(03) 9873 0068
sales@audiospace.com.au
www.audiospace.com.au
• Excellent value
• Switchable feedback
• Manual bias
• Sharp feet
• Output meter
• Bias adjust
screwdriver
LAB REPORT
Readers interested in a full technical
appraisal of the performance of the
Audio Space Reference 3.1 (300B) Valve
Integrated Amplifier should continue
on and read the LABORATORY REPORT
published on page 26. Readers should
note that the results mentioned in the
report, tabulated in performance charts
and/or displayed using graphs and/
or photographs should be
construed as applying only to
the specific sample tested.
Lab Report on page 26
avhub.com.au
21
ON TEST
Audio Space Reference 3.1 (300B) Integrated Valve Amplifier
symmetry, so the selector knob would have
an identical number of ‘steps’ on its left and
right-hand sides. In theory you’d imagine
it would be better for the meter to monitor
both channels, and this would be technically
correct, but in practise when you’re playing
back pre-recorded music, monitoring a single
channel will still give a fair indication of
the actual ‘two-channel’ output level and,
more importantly in this case, an excellent
indication of whether you’re overdriving the
Reference 3.1’s output stage.
The input selector switch has positions for
(going clockwise) Phono, CD, Aux and Tuner.
The phono stage is designed for a moving
magnet cartridge with a nominal output of
3–5mV.
The rear of the amplifier reveals that
Audio Space is offering a higher number of
taps on its output transformers than usual, so
you can elect to connect your loudspeakers
via 4Ω, 8Ω or 16Ω taps. (In order to cut
costs, many valve amplifier manufacturers
now offer only a single 6Ω tap or just 4Ω
and 8Ω taps.) I guess designer Peter Lau does
this because with an output of 21-watts per
channel, it’s essential to get maximum power
transfer from the source into the load, and
choosing the tap that best matches the real
(rather than the ‘nominal’) impedance of
your speakers is the best way to achieve this,
and he might be figuring a large number of
people will be using single cone wide-range
loudspeakers, which generally have higher
impedances than multi-way systems. This
means it would be handy to know the real
impedance of your speakers, so they will
have had to have been tested and the results
graphed.
If you don’t know the real impedance
of your speakers, and wish to establish the
correct tap to use, one good method is be
22
Australian
to connect your speakers to the 8Ω tap and
play music and, while the music is playing,
make a mental note of the volume level.
Then, without touching anything at all,
switch the amplifier off, swap the speakers
over so they’re being driven by the 4Ω
terminals, then turn the amplifier on and
play that same piece of music. If the volume
is louder then keep your speakers on the
4Ω tap. If the music volume is lower, then
switch back to the 8Ω tap (remembering to
make connections only while the amplifier
is switched off!). If you have a wide-range
single-driver speaker design (Lowther, Fostex,
E.J. Jordan, Bandor, Feastrex, Tangband et al)
I’d recommend starting with the 16Ω tap,
then trial the 8Ω and 4Ω taps in turn. (If you
don’t own a wide-range single-driver speaker,
the 16Ω taps will not be the best choice for
your speakers, so there’s no need to include
them in your trials.)
The speaker terminals themselves are
very high-quality, gold-plated, multi-way
insulated types. The input terminals are all
high-quality gold-plated RCA types, long
enough to accommodate Lemo connectors.
Alongside the phono inputs is the mandatory
screw-style ground terminal, which is nickelplated. The 240V socket is a standard fused,
IEC type. The ‘shiny’ bits on the chassis are
said to be made from mirror-finished—and
the rest of the chassis from coated—nonmagnetic stainless steel. All three heavy-duty
transformers are also said to be proprietary
to Audio Space, and are claimed to have no
magnetic leakage at all, to prevent hum being
induced in the amplifier’s circuitry.
In Use and Listening
Sessions
First, a big thank-you to Audio Space for delivering this amplifier with all the valves pre-
installed. I have to wonder where the modern
penchant for delivering valve amplifiers
with the valves separately packed ever came
from. After all, it’s not as though valves are
intrinsically fragile. Just ask any lead guitarist
who’s just plugged in his valve-equipped
Fender Twin amplifier, after bumping it up
the stairs to a nightclub for a gig, after first
retrieving the amplifier from the boot of his
car, in which it had been transported—valves
still warm—from an earlier gig a few hours
before. Valve amps are tough, and there’s
absolutely no need to remove them in the
event you have to move the amplifier for any
reason, either from one room to another, or
even to a new home. The fact the valves were
pre-installed (and the amplifier already runin not only by the good folk at Audio Space
Australia, but by a previous reviewer), meant
that all I had to do was warm the amplifier
up, then adjust the bias before starting on
the listening sessions.
It was when I went to adjust the bias that
I realised that I didn’t have an insulated longshanked, spade-headed screwdriver to adjust
the bias screws. To my mind, you’ll need
one because the fact that the bias screws are
located between two very hot valves means
that if you do what I resorted to doing, which
was use a short-shanked screwdriver, you will
end up burning your fingers… unless you’re
very careful indeed. I later discovered that
sourcing a long-shanked screwdriver with a
sufficiently small head wasn’t all that easy, so
it’d be nice if Audio Space had provided one
with the amplifier. Once you have a suitable
screwdriver, setting bias is easy and setting
bias for one valve does not seem to affect any
of the other valves, so there’s no need to set
all four then go back and ‘tweak’ the settings
all over again. And, once set, I found the
bias doesn’t appear to need much resetting, I
suspect a little fine-tuning every two or three
weeks would be all that’s required, and even
that might be gilding the lily. Once you have
finished biasing the valves, switch the combo
meter to either ‘L’ or ‘R’ so the meter is not
in circuit with one or other of the valves
while you are playing music. As for the
reason for the manual bias, it appears
that Lau doesn’t like auto-biasing,
on the basis that it ‘compromises a
power tube’s optimal performance.’
You’d imagine a valve
amplifier would be quick to
warm up, but the extremely
‘open’ nature of the
Audio Space Reference
3.1 (300B) and the
size of the chassis
meant that that isn’t
the case, and I found I
needed a minimum of
14 minutes before the
sound really ‘clicked’…
ON TEST
Audio Space Reference 3.1 (300B) Integrated Valve Amplifier
and that was using the valve cage, which
tends to contain the heat. If you don’t use
the cage, you’ll need to wait a little longer.
Not that you can’t start listening after just a
few minutes, but if you do, you’ll probably
just say ‘Oh’ about the sound quality of this
amplifier, whereas after twenty minutes, it
will be ‘Ooooooohhhhh’. And, just in case it
isn’t obvious, you WILL need to select your
partnering speakers carefully, because the
300B version of the Audio Space Reference
3.1 design is not overly powerful. I’d say that
speakers that are 87dBSPL efficient would be
the least I’d personally consider using and,
if you like your music loud, I wouldn’t even
be thinking of using speakers rated at less
than 90dBSPL. (If you want more latitude in
speaker selection, you could look at the KT88 version of the Audio Space Reference 3.1,
which can be switched between triode pushpull operation, in which mode it delivers
26-watts per channel and ultralinear pushpull, in which mode it’s rated at 48-watts per
channel.)
If you’re listening while the amplifier
warms up, you’ll love it when the sound
‘clicks’ because all of a sudden, notes will
start to float through the room, rather than
being propelled by the speakers, and there’s
instantly a Zen-like sense of calmness to the
sound, even if you’re listening to music that’s
far from calm. The soundstage also opens out
and becomes all-enveloping, so rather than
just being able to hear where instruments
are located, they become tangibly located at
that position… almost visible. I found myself
experiencing an over-arching feeling that
the sound became suddenly totally fluid, yet
all the strands were rendered clearly, and I
could instantly zero-in on any melody—or
countermelody or harmony—to bring its
contribution forward, or zoom-out to hear
the whole and even, weirdly-enough, to
hear both the micro and the
macro at the same time. If
you’ve never experienced this
yourself, the closest I can get to
explaining the moment is that
it’s the aural equivalent of those
3D pictures where you look at
a group of coloured dots and, if
you relax your eyes, suddenly
these dots will become a real
3D image that fills your entire visual cortex.
The bass surprised me by being far tighter
than I usually hear with valve amplifiers,
though I did find that this tightness was a
moveable feast, because when I tried different
loudspeakers with the Audio Space Reference
3.1 (300B), the degree of ‘tightness’ varied…
though it always varied between and ‘tight’
and ‘less-tight’ without ever getting so far
that I would describe it as ‘loose’ or flabby.
The tonality was excellent… OK, perhaps it
was ever-so-slightly warm, but that’s the way
I like it. There was certainly none of that
solid-state steeliness to the sound. Pitching
at low frequencies was absolutely perfect. For
example, when a bass guitar was playing the
same note as another instrument (but a few
octaves lower, of course) the interval between
the two notes was perfect.
High frequencies—treble if you like—
were gorgeous. Stunning. Lush. No, I’m
not going to resort to a thesaurus and keep
going. Instead, just think of the highestpitched note you can, then imagine that
note at its purest and most limpid, and
you’ll have some inkling of the highs you’ll
hear from the Audio Space Reference 3.1
(300B). And that’s with the standard-issue
glass (well, except for my NOS 6SN7s). What
if you swapped out those Chinese 12AX7s
and replaced them with US or UK-made
equivalents? And what of replacing the
standard Audio Space 300Bs with Western
Electric 300Bs? Or perhaps more modern
300Bs from JJ, or KR Audio, or Gold Lion? My
mind is already racing…
‘Hey! What about that feedback switch?
Which setting did you use when writing this
review?’ If that’s you asking these questions,
I fear that I have no easy ‘off-the-cuff’ answer
for you. When I started my auditions, I
thought I’d first work out which mode (high
or low) gave the best sound quality, and
then leave the switch in that position for
the remainder of the review. (I’m a great
believer in the KISS reviewing methodology.)
Naturally, I tried several different pieces of
music when doing this. Unfortunately, the
sound quality was great in both feedback
modes, and the decision of what I personally
thought was the ‘best’ mode depended on
what I was playing and (as I later discovered)
even on my own particular mood when
I was listening. If I had to make a rash
generalisation, I’d say that I liked ‘high’ with
classical music and ‘low’ with all other types.
But equally, I discovered plenty of exceptions
to this rule, where a particular classical piece
sounded better in ‘low’ mode, or a rock track
sounded better in ‘high’ than it did in ‘low’.
One gripe I had with the switch is that it’s
pretty difficult to properly evaluate the two
modes, because whenever you switch from
‘High’ to ‘Low’ the volume level jumps up
dramatically, so unless you compensate for
this, by turning back the volume control
at the same time, you will always judge the
‘Low’ feedback setting to sound better, just
because it’s louder. As I think has been noted
on these pages previously, this is a failing of
the human ear/brain. If you hear two sounds
that are identical except for one being louder
than the other, you’ll always perceive the
louder of the two as being the ‘better’ of the
two.
Conclusion
I loved this amp. In fact I loved this Audio
Space Reference 3.1 so much that I’m going to buy one. Just not this one. You see I
couldn’t live with myself if I didn’t audition
the KT-88 version first (and not least because
it’s almost a grand cheaper, at $4,280). I am
pretty sure that after I have A/B-auditioned
the KT-88 version against the 300B version I
will still end up buying the more-expensive
300B version, but I’m not going to spring the dough until
I am absolutely certain. What
I can tell you is that, either
way, there is an Audio Space
Reference 3.1 in my future.
And if you take the time to listen to one (or the other!) there
may be one in yours, too.
Chris Croft
LAB REPORT ON PAGE 26
24
Australian
Audio Space Reference 3.1 (300B) Integrated Valve Amplifier
LAB REPORT
CONTINUED FROM PAGE 24
Laboratory Test Results
I wasn’t surprised to see the power output figures that resulted from Newport Test Labs tests
on the Audio Space Reference 3.1 (300B).
What is surprising is that Audio Space has
not been tempted to inflate the power output, like most other valve amplifier manufacturers. So, as you can see, Audio Space claims
an output of 21-watts per channel, and that’s
exactly the power output the test sample delivered… though only when a single channel
was driven into an 8Ω load. However, unlike
most valve amplifiers, the Audio Space could
maintain this output level at the frequency
extremes (20Hz and 20kHz). When both
channels were driven, power output dropped
to 15-watts at 20Hz, 16-watts at 1kHz and
17-watts at 20kHz, so to meet the Australian
power output standard, this amplifier would
have to be rated at 15-watts per channel. The
figures into a 4Ω load (driven from the 8Ω
tap) were almost identical to those for the 8Ω
load, except for there being very slightly less
power available at 20Hz.
Note that the power output figures are
all a little imprecise, as they depend on
the visual acuity of the person doing the
test to detect waveform distortion on an
oscilloscope because, in common with all
valve amplifiers, the Audio Space does not
go into ‘hard’ clipping when it reaches the
upper limits of its output power. In practise,
if the technician had allowed a little more
waveform distortion, power output would
be fractionally higher than shown in the
tabulated listing and in the bar graphs.
During testing, Newport Test Labs did a quick
check on the accuracy of the power output
meter and reported that it is calibrated so
that when the amplifier is delivering 1-watt
into an 8Ω load, the needle will be at –10VU.
When the amplifier is delivering 10-watts
into an 8Ω load, the meter will show 0VU.
Note, however, that because the meter
needle’s action is damped, if the needle is
hovering around the 0VU mark when you’re
playing music, it’s likely that musical peaks
will be slipping through that will be taxing
the amplifier’s power output reserves.
This amplifier’s frequency response
also varies depending on the setting of
the feedback control (shown in Graph 4).
With low feedback (red trace) the frequency
Audio Space Reference 3.1 (300B) – Power Output Tests
Channel
Load (Ω)
20Hz
(watts)
20Hz
(dBW)
1kHz
(watts)
1kHz
(dBW)
20kHz
(watts)
20kHz
(dBW)
1
8Ω
21
13.2
21
13.2
21
13.2
2
8Ω
15
11.7
16
17
17
12.3
1
4Ω
18
12.5
21
21
21
13.2
2
4Ω
14
11.4
16
17
17
12.3
Note: Figures in the dBW column represent output level in decibels referred to one watt output.
Audio Space Reference 3.1 (300B) — Test Results
Test
Measured Result
Frequency Response @ 1 watt o/p
Frequency Response @ 1 watt o/p
Channel Separation (dB)
Channel Balance
Interchannel Phase
THD+N
–1dB
3.5Hz – 22kHz
–3dB
69dB / 70dB / 103dB
1.28
4.47 / 0.44 / 4.96
0.44% / 0.93%
Signal-to-Noise (unwghted/wghted)
Signal-to-Noise (unwghted/wghted)
Units/Comment
7.5Hz – 12kHz
(20Hz / 1kHz / 20kHz)
dB @ 1kHz
degrees ( 20Hz / 1kHz / 20kHz)
@ 1-watt / @ rated output
71dB / 84dB
dB referred to 1-watt output
84dB / 92dB
dB referred to rated output
Input Sensitivity (CD Input)
27mV / 128mV
(1-watt / rated output)
Input Sensitivity (Direct Input)
101mV / 485mV
(1-watt / rated output)
Damping Factor
6
@1kHz
Power Consumption
N/A / 179
Power Consumption
182 / 233
watts at 1-watt / at rated output
Mains Voltage Variation during Test
238 – 245
Minimum – Maximum
Heatsink Temperature (Degrees C)
52
26
Australian
watts (Standby / On)
@ 1-watt output on valve cage
LAB REPORT
Audio Space Reference 3.1 (300B) Integrated Valve Amplifier
response of the Audio Space Reference 3.1 (300B) is rather curtailed,
being 1dB down at 9.5Hz and 7.5kHz and then 3dB down at 4.5Hz
and 15kHz. With high feedback (black trace), the response is more
extended, so it’s just 1dB down at 7.5Hz and 12kHz, and 3dB down at
3.5Hz and 22kHz. According to Newport Test Labs’ tests, this puts the
measured frequency response of the Audio Space Reference 3.1 (300B)
at 3.5Hz to 22kHz ±1.5dB when driving a standard laboratory test load
(8Ω non-inductive). This falls a tad short of the claimed response of
20Hz to 30kHz ±1dB. When driving a simulated loudspeaker load the
high-frequency response was further curtailed (as shown in Graph
5). Into this load Newport Test Labs measured the frequency response
(with the high setting of the feedback control) as 3.5Hz to 20kHz
±2.5dB. The degree of variation in the response means that this
amplifier will sound substantially different depending on the speakers
you connect to it—and the transformer tap you use—simply because
of the way the speaker’s own impedance will affect the amplifier’s
frequency response.
Channel separation was good, measuring around 70dB at low to
midrange frequencies, and improving to more than 100dB at high
frequencies, as you can see from the tabulated results. Channel
balance was not so good, with a measured 1.28dB difference between
the left and right channels. However, it’s unlikely that this would
be perceptible and would in any case be swamped by the frequency
response variations caused by the loudspeakers. Inter-channel phase
was just 0.44° at 1kHz and a little more than 4° at the frequency
extremes, which is excellent for a valve amplifier.
dBFS
0.00
Newport Test Labs
As you’d surmise, distortion also varied considerably depending on
the setting of the feedback switch, and the ‘lowest’ distortion levels
were obtained using the ‘high’ switch setting, though irrespective of
the setting, distortion was in all cases higher than I see from solid-state
designs.
Graph 1 (high feedback) and Graph 1a (low feedback) of distortion
into an 8Ω non-inductive test load serve nicely to illustrate. In Graph
1, the second harmonic is at –46dB (0.501%), the third at –52dB
(0.251dB), the fourth at –73dB (0.022%) and the fifth and higherorder components all below –100dB (0.001%). In Graph 1a, both the
second- and third-harmonic distortion components increase to –40dB
(1.0%) and –45dB (0.562%) respectively. The fourth, fifth, sixth and
seventh harmonics descend stair-case-like from –68dB (0.039%) down
to just below –100dB (0.001%) after which there’s nothing much,
save for a few harmonics that are more than 110dB (0.0001%) down.
Overall THD, however, was 0.44%, which is a good result.
Increasing power output to 21-watts (Graphs 2 and 2a) saw all the
higher-order harmonic distortion components rise considerably in
level, but the second and third-harmonic components each stayed
around at –40dB (1.0%).
Newport Test Labs also tested the distortion at rated output when
the amplifier was driving 4Ω loads via the 8Ωtap for both high and
low settings of the feedback switch (not shown). There were small
differences between the performance into 8Ω and 4Ω loads, but
nothing major. Overall THD was measured at 0.93%, just creeping in
under Audio Space’s specification of 1.0%.
dBFS
0.00
-20.00
-20.00
-40.00
-40.00
-60.00
-60.00
-80.00
-80.00
-100.00
-100.00
-120.00
-120.00
-140.00
-140.00
0.00 Hz
4000.00
8000.00
12000.00
16000.00
20000.00
0.00 Hz
Graph 1: Total harmonic distortion (THD) at 1kHz at an output of 1-watt into an 8-ohm
non-inductive load, referenced to 0dB. Transformer tap: 8-ohms. Feedback High.
[Audio Space Reference 3.1 (300B) Integrated Amplifier]
dBFS
0.00
Newport Test Labs
4000.00
8000.00
12000.00
16000.00
20000.00
Graph 2: Total harmonic distortion (THD) at 1kHz at an output of 21-watts into an 8-ohm
non-inductive load, referenced to 0dB. Transformer tap: 8-ohms. Feedback High. [Audio
Space Reference 3.1 (300B) Integrated Amplifier]
dBFS
0.00
-20.00
-20.00
-40.00
-40.00
-60.00
-60.00
-80.00
-80.00
-100.00
-100.00
-120.00
-120.00
Newport Test Labs
-140.00
-140.00
0.00 Hz
4000.00
8000.00
12000.00
16000.00
20000.00
Graph 1a: Total harmonic distortion (THD) at 1kHz at an output of 1-watt into an 8-ohm
non-inductive load, referenced to 0dB. Transformer tap: 8-ohms. Feedback Low.
[Audio Space Reference 3.1 (300B) Integrated Amplifier]
28
Newport Test Labs
Australian
0.00 Hz
4000.00
8000.00
12000.00
16000.00
20000.00
Graph 2a: Total harmonic distortion (THD) at 1kHz at an output of 21-watts into an 8-ohm
non-inductive load, referenced to 0dB. Transformer tap: 8-ohms. Feedback Low. [Audio
Space Reference 3.1 (300B) Integrated Amplifier]
LAB REPORT
Audio Space Reference 3.1 (300B) Integrated Valve Amplifier
Newport Test Labs tested for intermodulation (IMD) effects using
both SMPTE and CCIF test signals, of which only the results for the
CCIF-IMD test are shown here (Graphs 3 and 3a). I was expecting
to see quite a difference here, and there is, with the higher feedback
resulting in less regenerated signal at 1kHz, but it’s only a 5dB
difference, which I don’t think is significant. Perhaps more significant
is the actual level of the regenerated signal, which at around 50dB
(0.316%) is high, but typical of a valve amplifier.
Signal-to-noise ratios were good, again particularly for a valve
amplifier, with Newport Test Labs measuring 71dB (unweighted)
referred to a one watt output, increasing to 84dB with A-weighting.
Referencing the noise to rated output (21-watts), the Audio Space
returned S/N results of 84dB (unweighted) and 92dB (A-weighted).
All these were measured through the standard CD input, so I’d expect
them to improve further if the ‘Direct’ input had been used for testing.
The square wave oscillograms reflect the limited bandwidth of the
Audio Space Reference 3.1 (300B), with the 100Hz wave showing the
expected tilt, but no obvious phase shift, and the 10kHz square wave
exhibiting considerable rounding on the leading edge—even the 1kHz
square wave shows premature rounding on its leading edge. What was
interesting was the amplifier’s performance into a highly capacitive
load. Whereas most solid-state amplifiers would show considerable
overshoot and extended ringing, the Audio Space Reference 3.1 (300B)
overshoots only the tiniest amount when the high feedback mode is
selected, and not at all when the low feedback mode is selected… and
there’s absolutely no ringing visible in either mode. As I have noted
dBFS
0.00
Newport Test Labs
This type of performance is consistent
with amplifiers that are subjectively
assessed as having ‘good sound.’
on previous occasions, this type of performance is consistent with
amplifiers that are subjectively assessed as having ‘good sound.’ On
the technical side, it means the Audio Space Reference 3.1 (300B) will
be completely stable when driving electrostatic loudspeakers, or other
speaker designs that present a ‘difficult’ load to the driving amplifier.
Input sensitivity was measured at 27mV for one watt output, while
128mV is all that’s necessary for the amplifier to deliver its rated
output. This is about par for the course for an integrated amplifier,
so no concerns here. No concerns either if you use the Direct input,
though if you do you’ll need 101mV at the input for the amplifier to
deliver one watt, and 485mV for it to deliver rated output. Again these
are fairly typical figures for any power amplifier.
As you’d expect from a valve power amplifier, the Audio Space will
draw considerable energy from your household mains supply, pulling
around 180-watts all the time, including when you’re playing music
at low-ish levels, and more than 230-watts when being driven hard.
Despite this, the amplifier doesn’t get overly hot, with the top of the
central transformer getting hotter than the valve cage itself, with
Newport Test Labs measuring both temperatures at about 51°C. So to
save on energy bills and extend valve life, I’d recommend turning the
Steve Holding
Audio Space off whenever you’re not using it.
dBFS
0.00
-10.00
-10.00
-20.00
-20.00
-30.00
-30.00
-40.00
-40.00
-50.00
-50.00
-60.00
-60.00
-70.00
-70.00
-80.00
-80.00
-90.00
-90.00
-100.00
-100.00
-110.00
-110.00
-120.00
-120.00
0.00 Hz
6000.00
12000.00
18000.00
24000.00
30000.00
0.00 Hz
Graph 3: Intermodulation distortion (CCIF-IMD) using test signals at 19kHz and 20kHz, at
an output of 1-watt into an 8-ohm non-inductive load, referenced to 0dB. Transformer tap:
8-ohms. Feedback: High. [Audio Space Reference 3.1 (300B) Integrated Amplifier]
dBr
3.00
Newport Test Labs
6000.00
12000.00
18000.00
24000.00
30000.00
Graph 3a: Intermodulation distortion (CCIF-IMD) using test signals at 19kHz and 20kHz, at
an output of 1-watt into an 8-ohm non-inductive load, referenced to 0dB. Transformer tap:
8-ohms. Feedback: Low. [Audio Space Reference 3.1 (300B) Integrated Amplifier]
dBr
9.00
2.50
7.50
2.00
6.00
1.50
4.50
1.00
3.00
0.50
1.50
0.00
0.00
-0.50
-1.50
-1.00
-3.00
-1.50
-4.50
-2.00
-6.00
-2.50
-7.50
Newport Test Labs
-9.00
-3.00
10.00 Hz
100.00
1000.00
10000.00
Graph 4: Frequency response of line input at an output of 1-watt output into an 8-ohm
non-inductive load. Transformer tap: 8-ohms. Feedback High (Black Trace) vs. Feedback Low
(Red Trace). [Audio Space Reference 3.1 (300B) Integrated Amplifier]
30
Newport Test Labs
Australian
10.00 Hz
100.00
1000.00
10000.00
Graph 5: Frequency response of line input at an output of 1-watt into a combination
resistive/inductive/capacitive load representative of a typical two-way loudspeaker system
using high feedback setting (blue trace) and low feedback setting (red trace). Response into
a non-inductive 8-ohm load also shown (black trace). Transformer tap: 8-ohms. [Audio Space
Reference 3.1 (300B) Integrated Amplifier]
O
I
N
I
Z
N
O
E
L
B
A
NOW AVAIL
c
a
M
/
C
P
&
id
o
r
d
n
A
for iPad,
SHOW REPORT High
End Munich
September/October 2013
TOUCH US
PINCH
US
WE’VE GONE ALL DIGITAL
DUAL VALVES!
AVHUB.COM.AU
AVHUB
.COM.AU
A$8.95 NZ$10.99
ALSO
REVIEWED
300B or KT88? Full review and test of
AudioSpace’s innovative amplifier...
Krix
Phoenix V2
Loudspeakers
PSB
HD8
Bricasti Design M1
Digital-To-Analogue
Converter
Subwoofer
HF Sep13_001 Cover.indd 1
SOUND+IMAGE
AUSTRALIAN HI-FI
PRO PHOTO
CAMERA
AUSTRALIAN INCAR
ENTERTAINMENT
8/26/2013 2:46:54 PM
Buy single issues or special
three-issue bargain bundles
through AVHub.com.au or
www.zinio.com/soundimage
AUSTRALIA’S
No.1 AV GUIDE
$7.95 NZ$9.90 Volume 69 Number 6
A L L
Aug/Sept 2013
ISSUE #26/5 $8.95 NZ$10.99
LOCATION REPORT – MOTOREX 2013
T H E
B E S T
F R O M
T H I S
Y E A R ’ S
Head On Festival
ALSO
REVIEWED
NAD C356BEE
Best budget amplifier?
Usher Audio
Diamond tweeter
sings…
D-SLR VIDEO MAKER – EVERYTHING YOU’LL EVER NEED TO KNOW
100% AUSTRALIAN
Sony SS-NA2ES
Amazing
achievement!
MSB Sig. DAC IV Plus
Hand-made heaven
$8.50 /// NZ $10.99
WOOFER WORLD
IN SSV WAGON
THE HIGH-END
ALPINE’S DDLINEAR & PIONEER’S AVIC
LG
Panasonic
Samsung
Sony
Plasma
LED-LCD
4K UHD
65-inch
55-inch
CHOOSE YOUR NEW TV!
SHOOTIN
G
STAR
VW IN ORION
CONSTELLATI
ON
FULL REVIEWS OF THE BIGSCREEN BRANDS
YAMAHA’S ULTIMATE AVENTAGE
EXCLUSIVE TEST OF THE A5000 PRE-POWER
A DS
CANE TO
IN 3D!
Studio 590
bi-radial
horn is JBL’s
best yet…
VAILABL
T’S A
E
IN
HA
W
Studio
Flash
Power
Packs
TRAVEL
PHOTOGRAPHY
Horn avhub.com.au
Power!
MAKE EVERY
SHOT A WINNER
31
NZ $10.99
WONDROUS
WAGON
WHEELS
5#2013
5#2
5#20
Jul/Aug 2013
A$8.95 NZ$10.99
AVHUB.COM.AU
/OCT 2013 $8.50
T H E L AT E S T M O B I L E E L E C T R O N I C S T R E N D S & T E C H N I Q U E S
NIKON D7100
READY TO RELEASE
FIRST LOOK!
PANASONIC’S
ALSO IN THIS ISSUE :
CANON EOS 700D
PANASONIC G6
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising