This review page is supported in part by the sponsors whose ad banners are displayed below

Andreas Hofmann: "The model HP300SE is a natural evolution of the HP300MkII which was based on the concept of a two-stage valve preamp with solid-state output current buffer and switchable gain over two steps to accommodate varying speaker efficiencies. Its XLR output was chip-based, i.e. electronic and the power supply was internal. After a few years I saw potential for improvement in the gain stage, the balanced output and the power supply, particularly its voltage regulation. Further desirable features we didn't yet have were an HT bypass and multiple true balanced inputs all of which the SE accomplishes.


"The XLR output now sports a super-symmetrical low-impedance balancing transformer. By relegating the power transformer offboard it could grow in size and achieve even better suppression of high-frequency AC line noise. The voltage regulation of the SE—all its supply voltages are regulated—is extremely stable to avoid long-term thermal drift and exhibits exceptionally low noise. I modified the signal path to tune sonics for greater smoothness, higher realism, a bit more bass extension and texture; to suppress residual noise even further; and to stabilize the NFB switching. Low-gain mode is obviously achieved with higher feedback which simultaneously maximizes the S/N ratio. NFB switching for gain adjustment naturally affects the response but even in high-gain mode the -3dB down point remains above 300kHz. To vary amplification factor by adding or subtracting gain stages instead would be an interesting challenge though problematic. One would have to successfully avoid altering the sound from one setting to the next. In low-gain mode we retain an amplification factor of 4.4 which is accompanied by residual noise that's down to 7µV across the full audible bandwidth. That equates to minus 108dB relative to a full 2V output. And that is achieved with standard commercial tubes produced today, not costly NOS glass. I'm not a specmanship fanatic but hum rightly bothers many clients. To me too a noisy circuit is nothing but poor engineering.


"In that regard the XLR output behaves identically to the RCAs since our balancing transformer adds no self noise. Power-supply hum is obviously an utter non-issue with relocation to an external 2mm steel casing. Retained from the MkII model is the concept of the solid-state output buffer. Preamps with tube outputs suffer higher output impedance and require big coupling capacitors before they end up with any respectable response and any usefully linear phase behavior in the bass. That's an always compromised solution unless one spent far more money than we allocated for this model or ran with output iron. But even an output transformer as the impedance-matching device isn't an ideal coupling mechanism. At least in my mind the unity-gain transistor buffer really is the best alternative.


"Our buffer couples both the XLR and RCA outputs. In the HP300SE we rely for this on a very special industrial Japanese chip which offers 250MHz bandwidth, distortion figures with three zeroes behind the decimal point and high current gain. We set the RCA output impedance artificially to 50Ω. The buffer would have allowed figures down to about 5Ω but the reason not to do so was threefold. One, the preamp would no longer be immune to shorts. During an inadvertent but prolonged short-circuit condition, too much current would flow to overheat and take out the buffer. To counteract that would then require sizeable heat sinks difficult to accommodate in our chassis since we'd have to dissipate about 8 watts per buffer.


"Two, 50Ω systems exhibit exceptionally wide bandwidth to interface optimally with any feasible ancillary configurations unless someone were to use extremely high-capacitance cables. Three, driving a very fast amp with an ultra-low Z-out preamp could lead to possible oscillations. Back to the XLR outputs, their output impedance is about 35Ω per phase to equate to low-impedance DC coupling. This is very useful for a number of solid-state amps. The RCA outputs are AC-coupled through a 15µF capacitor.


"The improvements in the SE power supply are primarily the actual power transformer. It's no secret that the quality of the mains power impacts the audio quality. More difficult is reducing its negative influence without an AC filter. Our transformers are of our own design and wound in-house to be fully optimized for HF noise suppression, efficiency and power. These factors are all vital in a preamp to minimize noise, maximize dynamic range and assure compliance across a broad range of line-voltage conditions (in our 230V model from 190-260V).


"When properly wound, the balancing output transformer creates perfect phase symmetry on both amplitude and phase. It also allows us to float the ground from the power amp with a rear-mounted switch. This cuts ground loops and increases immunity to utility power influences because certain high-frequency power-line elements sent back to the other electronics are being suppressed. Power-line disturbances caused by the power amps are cancelled out in the counter-phase windings of our balancing transformer. Obviously inferior transformers will restrict the bandwidth. Ours is linear within 1dB from 10Hz to 100kHz. With today's best core materials distortion in transformers has become negligible. Above 300Hz our RCA and XLR outputs measure absolutely identical. Only below 300Hz does the transformer give up just a tiny bit but this is both relative to output voltage and occurs at such low figures as to render any discussion about it truly academic."


A tale of seven preamps
. After years of exploring predominantly low-power single-ended valve amps, I'd joined the dark and quiet side of transistor amps again mostly of the low-power single-ended persuasion. With my initials some things just can't be helped. That being the case, my ongoing fascination with tubes has shifted to low-level applications, i.e. preamps. On balance I favor harnessing glowing bottles once they've been properly isolated from the heavy lifting of wildly fluctuating load impedances, reactive phase angles and the back electromotive forces of manly woofers. For me 2013 would be the year of the tube linestage.


The coincident arrival of Nagra's Jazz and Thrax's Dionysos against my own ModWright LS-100 fitted with Psvane CV-181/6SN7, TruLife Audio Athena with 6H30, Esoteric C-03 transistor linestage and Bent Audio Tap-X autoformer passive made for my first broad opportunity to see how in a hybrid system scheme of valve pre + transistor power various small-signal tube types and circuit ideas might express themselves as flavorants. I'd previously sampled Sasa Cokic's proof-of-concept 101D preamp. That was a one-up sample of a single-stage direct-heated triode variant. With a gloriously textured midband and airily elastic if not ultimately extended treble, it had also been clearly soft in the bass and microphonic enough to have filaments ring after each manual volume knob adjustment. A short-term reader loan of a €24.000 Concert Fidelity CF080LSX introduced me to a hybrid 12AU7 circuit which excelled at clarity, resolution and immediacy, a criminally steep price and most serious lack of featurization. What flavors and particular musical strengths would these five valve decks pursue?