Sunday, April 16, 2017

Nagra BPS Phono Stage - Brief Impressions


As luck would have it, a kind soul lent me a Nagra BPS phono stage for a week to conduct an analog workshop for a small group. The BPS was handed over to me in a small plastic flightcase, which was a refreshing change from the equipment I have had to handle recently (which could double up as weights in the gym !)


The Nagra BPS is slightly larger than the palm of my hand, and operates from a single 9V battery. It has a variety of loading and gain options, which should make it easy to mate with a variety of cartridges. MC gain is partly accomplished using in-house wound transformers.

The top cover of the BPS removes easily using the supplied hex key, giving access to various jumpers and the loading board. A separate hatch allows you to swop the 9V battery easily without having to open the top cover. 

The front panel has a single toggle switch with three positions, "on", "off" and "test" (for battery life). The rear panel has two pairs of RCA sockets and an incredibly small grounding post. A DC socket allows you to use the BPS with an external power supply (7-10 V DC). I used this feature only to keep the circuit warmed-up and switched to the battery for listening sessions.

Gain is a very healthy 53 db for MM and 64 db for MC. Jumpers allow you to attenuate the gain by 16 db if you require. The resistive load settings for the BPS are 100, 150, 220, 330, 470 and 1000 ohms. 

I was a bit concerned about battery life in use, but the manual states that the battery life is 100 hours, and in the 4-5 hours that I used the unit, the dealer installed battery held up fine.

Sound Quality

I tried it first with my high output MC cartridge, a Sumiko Blackbird. The high frequencies were distorted, but using the 16db attenuator fixed all of that. Compared to my Whest Two phonostage, the Nagra had a lighter and faster tone. 

The workshop host was using a midrange Pro-ject deck with an Ortofon 2M Blue MM cartridge. The differences between the Nagra and Whest were clearly audible, with some members of the audience preferring the more linear and extended sound of the Nagra and some preferring the more romantic tone of the Whest. The Nagra had a more neutral tone, with a very tight and fast bass line, and a drier but more extended top end. I also felt that the Nagra's soundstage presentation was more distant, but with better separation and more precise imaging.

Moving the Nagra back home, I tried it with my main rig, a Soulines Kubrick DCX with Jelco SA-750 tonearm. I tried a Hana SH cartridge - a high output MC cartridge (review coming) and a Shelter 5000 MC cartridge. I used my Graham Slee Reflex M and Elevator as a comparison phono stage. The Nagra sounded quite different in this setup. On both cartridges, midrange had a more laidback and subtly creamy quality to it, while both frequency extremes were clean and tightly controlled. The Graham Slee had a much more dynamic presentation, with a more vibrant and tonally dense delivery. Staging wise, the Graham Slee projected a more three dimensional stage, with a greater sense of depth and height. In comparison, you could describe the Nagra as being subtly expressive, but with a more restrained delivery.


Performance wise, I would put the Nagra to be on par with the Graham Slee duo, although their tonal balance are quite different. As a matter of reference, both phono stages are in a similar price bracket locally. Assuming that you like a more neutral tone, the Nagra is an impressive phono stage.

Thursday, March 9, 2017

Hifiman Shangri-La Private Event at AV One


You got to give it to Hifiman. It takes guts to launch a headphone system with a USD 50,000 price tag.

Clearly aimed into Sennheiser Orpheus II territory, the Shangri-La headphone system consists of a fully balanced drive tube amplifier / energizer and a Electrostatic headphone.

AV One, the local dealer for Hifiman arranged a private event to showcase the Shangri-La. Dr. Fang Bian, CEO and founder of Hifiman graced the event and answered any questions posed by the audience.


The Shangri-La headphone amplifier is very large, with four 6SN7 tubes in the input stage, and four 300B tubes in the output stage. The amplifier is designed to offer balanced drive, and the microprocessor controlled volume knob activates a series of relay controlled resistors to set the volume level.

Dr. Fang mentioned that they enjoy a very special relationship with TJ Fullmusic, the tube manufacturer, which is located just a short distance away from them in the Chinese city of Tianjin. A vigorous selection process is employed - only a handful are selected from each batch sent (consisting of a few hundred tubes).

According to Dr. Fang, tube amplifiers are a perfect design for electrostatic headphones due to their high voltage design. Unlike speaker loads, an electrostatic headphone load requires very little current, and an output transformer is not required. 

The driver membrane of the headphones are treated with a nano coating that ensures that performance is not adversely affected by humidity levels (a bane in Singapore with it's tropical humidity). He acknowledged that nano coating is not new, but he felt that the coating employed in the Shangri-La had no adverse effect on the sound, unlike those used by competitors.

I asked Dr. Fang about the cost of the amplifier and headphone. Dr. Fang said that the amplifier cost more to build from a materials point of view. However for Hifiman, the amplifier was relatively easier to design compared to the headphones which required a lot more effort in research and development.

Sound Quality

I spent about 15 minutes alone with the Shangri-La listening to a variety of music that was installed on the supplied notebook computer that was in turned hooked up to a Chord Dave DAC.

The Shangri-La has quite a unique sound signature. It has an immediately recognisable warm and liquid signature. However, it also has a high detailed presentation and very extended top-end. It also has an effortless feel to it, with an easy going sound signature that should make long listening sessions comfortable. Think lightning fast transients with a natural leading edge and decay and that's the Shangri-La for you.

Tonally, it sits somewhere in between the Sennheiser HD600 and 800. Dynamics wise, it is not the most impactful headphone I've heard, but I would say that details wise, it is up there with the best. 

If we disregard the price tag, these are very nice sounding headphones - I could easily live with these. But the price tag on these make these unobtainable for all but a select few. Statement-fi products like the Shangri-La do not follow the usual rules on performance vs cost. They exist simply because they can - there will always be the customer that demands the very best, at any cost.

I asked Dr. Fang whether we could expect any trickle-down products at something mere mortals like myself could afford. He said that something is being planned, and left it at that. Don't keep us waiting too long !


Listening to the Shangri-La was an experience. Let's hope we get a more affordable Shangri-La lite model soon.

Thank you to AV One for inviting me for this exclusive event, and also a big thanks to Dr. Fang of Hifiman for patiently answering my questions.

AV One


Sunday, February 26, 2017

Holo Spring Level 3 DAC - A mini review


I recently spent a few quality days with the Holo Spring Level 3 DAC, courtesy of Sound Affairs, Singapore distributor for Holo. There is plenty of chatter in various audio forums on the Holo Spring DAC, especially on the high level of performance it delivers for it's price.


The Holo Spring is a really nice looking DAC (I adore those copper side panels), and is solidly built. It probably has the largest display I've ever seen, and the dot matrix panel is fully legible even from my listening position. 

The Spring DAC is a R2R DAC design. What differentiates the Spring from the rest of it's competition is that fact that DSD is decoded using a R2R circuit too. This is achieved by having two separate R2R circuits - one for PCM and the other for DSD.

The front panel has four buttons. The first button puts the DAC in standby mode, while the rest of the buttons control the display intensity, oversampling mode and source. There are four oversampling modes, NOS (no over-sampling), OS (both PCM and DSD are oversampled in their native format), OS PCM (oversampled PCM - DSD is converted to PCM), and OS DSD (oversampled DSD - PCM is converted to DSD).

The rear panel has both single-ended and balanced analog outputs, while a total of 6 digital inputs are offered - USB (galvanically isolated), coaxial, BNC, AES, Toslink and I2S via HDMI (PS Audio standard).

In use, the Spring DAC runs quite warm. It was left powered on for the few days I had it in my system. 

Dimensions wise, the Spring DAC measures 430 mm x 300 mm x 55 mm, and weighs 8.5 kg. 

Sound Quality

I was not able to do detailed comparisons of the Spring DAC using various digital inputs or its oversampling modes due to the short evaluation period I had. However, I did find that NOS mode sounded the most pleasing to my ears, while the tone of the unit was similar via SPDIF and I2S. I had a slight preference for I2S (fed via a Singxer SU-1 USB / SPDIF bridge) and stuck to that.

To be continued ...