[Italian version]

Reviewer: Nick Whetstone - TNT UK
Reviewed: January, 2012 (was: June, 2008)

The Chambers online dictionary defines 'hobby' as "an activity that someone does for pleasure when they are not working". Note the word 'pleasure'. And of course pleasurable it should be! I used to find my hobby very pleasurable, with almost nothing negative to say about it. But while the arrival of an Internet connection brought me a wealth of information, and some good friends, recently I have found life on some of the forums shall we say less than pleasurable. Debate on any subject is healthy, and to be encouraged, but it seems that more frequently, the initial respect that people had for each other, and their different opinions, is giving way to rudeness and intolerance. I mention this in the context of this review because the subject of the review has caused a few heated debates as to whether it is even necessary.

It is generally accepted that the best way to enhance a piece of (active) hi-fi equipment is to improve its power supply. The Logitech (formerly Slim Devices) Squeezebox which I shall refer to from now on as the SB3 (although this review applies to the earlier models too) should be a candidate for an upgraded PSU. So why the debate?

The SB3 comes with a small external switched mode power supply (SMPS) so why not substitute it for something better? Well, the argument goes that when the power from the external supply passes into the SB3, it goes through another switching supply that increases the voltage to supply the display, and then the rest of the unit is supplied after the 'switcher'. Still with me? So the argument is that no matter what quality of external supply is used, the quality of the power supply will be 'corrupted', or 'limited' by the internal 'switcher'. And it seems for that reason, there is a group of people who simply won't accept that changing the PSU of the SB3 will make any difference.

My own experience is that in practice the external PSU does make a difference. When I moved from the standard SMPS (that is supplied with the SB3) to my DIY linear PSU, I could clearly hear a difference in the sound, despite only using the SB3 as a digital transport. It would seem that I am far from being the only one to hear a difference and there is now quite a choice of third-party SB3 PSU's on offer. The first one that I have been offered for review is the Paul Hynes Design SR1-5 that I will compare against the stock SMPS, and my own linear supply. Although this review is my first association with Paul Hynes, I remember reading some of his articles on power supply design in a UK DIY hi-fi magazine called Common Ground more than a decade ago. Since then his power supply designs have become highly respected.

The first thing that I noticed about the SR1-5 is that it is well finished with lots of attention to detail. The attractive housing feels solid, and is reassuringly heavy (why do we still associate weight with quality in this age of miniaturization?). The rear panel comes with a switched/fused IEC socket and two connections for the 5 volt supply. The front panel is uncluttered to the point of not even having the usual LED. Instead the power-on indicator LED is mounted on the underside of the case and transmits a soft blue light under the case and onto the shelf or equipment platform. I actually get fed up with the array of LEDs staring at me from my hi-fi rack and think this idea is excellent! The review sample SR1-5 came with a quality umbilical cable, with connectors to connect it to the SR1-5 on one end, and the correct plug to connect it to the SB3. The 5 volt lead uses a quality locking connector to secure it to the SR1-5 and is an optional extra for the SR1-5 at 20 UK pounds. All in all, a very professional looking package.

For the technically minded amongst you, here are a few details taken from the PHD web site.

Specifications - error amplifier used in shunt regulator :- slew rate 5,000 volts per microsecond, gain bandwidth product 1 GHz, rise time < 1 nanosecond, settling time < 10 nanoseconds and wideband noise 2 nanovolts root HZ. Regulator output impedance is less than 0.001 ohm and the power supply line rejection ratio is 110 dB from DC to 200 KHz. Mains input is a standard IEC inlet. The DC output is a Lumberg precision chassis connector and 4mm sockets are also fitted for those who wish to experiment with DIY cables. DC1C DC lead using annealed copper twisted pair with polystyrene insulation, braided screen, terminated with a Lumberg precision connector to fit the power supply connector. Connectors to suit your equipment can be fitted with additional cost. Ask for a quote if this service is required.

After a short session listening to my system using my own DIY linear PSU, I installed the SR1-5. The difference was immediately apparent (instantly deflating my ego although my own supply only uses standard off-the-shelf' regulators, and not the discrete designs used in the SR1-5). Everything was a bit 'tighter'. Bass was better defined, detail was improved, and the sound-stage better defined. It wasn't a huge difference, but enough to not want to go back to my own linear PSU. The improvement over the standard SMPS is even greater, with the SMPS sounding a little 'soggy' after listening with the SR1-5.

At first, I felt that the sound with the SR1-5 was a bit leaner than I am used to with my own linear PSU. I wasn't sure if this was due to the bass being tightened up but in any case, after a couple of days, this 'leanness' had gone, so either I had got accustomed to it, or the SR1-5 had changed slightly as it burned in with use. I guess it was the latter because the SR1-5 continued to get better each day. The bass grew more and more prominent, but not so much as to overwhelm the rest of the frequencies. I also found myself (once more) noticing lyrics that I had not heard so clearly previously. After a few days the small improvement that I heard initially grew into something quite phenomenal! All the little nuances are brought out by the SR1-5 to a degree that some familiar tracks sounded almost new to me.

I should point out that my SB3 has a modified digital output and I don't listen via the analogue outputs. Fortunately I was able to borrow a stock SB3 and try that with its SMPS and the SR1-5 . Although I could clearly hear the same improvements using the analogue output of the standard SB3 through both headphones and amp/speakers, it wasn't quite such a noticeable improvement as with my modified SB3 through an external DAC. What I did notice was that a bit of glare heard with the stock SMPS and analogue output went away when using the SR1-5. How much the SR1-5 will do in your system will depend on your system's ability to show the improvement. However, I'm assuming that if you are considering buying something like the SR1-5, your system is already at a level that is capable of resolving the improvement it will make.

So is that an enthusiastic thumbs up for the SR1-5? Well, as I said, it's considerably better than my own linear supply. After reviewing the Transporter, I was left a little disappointed with the SB3 even with the modified digital output. With the SR1-5, I feel that the sound quality of the SB3 as a digital transport is much closer to (may be even the equal of) the Transporter. I can't say for sure as I no longer have a Transporter here for direct comparison.

I know that many of you will look at the cost of the SR1-5 and realize that it costs slightly more than the SB3. So is it worth it? OK, let's look at it this way. My own DIY effort using a lot of recycled bits like the case from an old PC PSU cost around 25 pounds. But if I used the quality of case and fittings that the SR1-5 uses, I could easily treble that, and still not have the same quality of sound because I wouldn't have the quality of regulators that are in the SR1-5. And I haven't even started to figure in labour and overheads etc. It's true that there are many commercial linear supplies available for much less than the SR1-5 but again, I doubt they will surpass my own DIY effort. It's only when we look at the more 'up-market' alternatives that the price of the SR1-5 is put into context.

One of the best known sources for SB3 modifications is Bolder Cables who will carry out internal modifications on the SB3 as well as supplying an up-rated PSU. At the time of writing, I have not heard the Bolder SB3 PSU which they call the The ULTIMATE Power Supply MK IV so I am not making a comparison on performance. But the price ranges from 950 USD for the basic version, to 1850 USD for the 'top-of-the-range' version, making the SR1-5 look like a positive bargain.

As it comes, the SB3 is not an audiophile device. It's good, remarkably good for the cost but at the price point, compromises had to be made in areas like the PSU. For the improvement it gives at the price, I reckon the SR1-5 is a sound investment that raises the standard of the SB3 to audiophile levels, more so when used with a good external DAC. Of course, go much further than that, and you could be spending almost as much as something like the at-tunes SB+. So again, I would say that the price of the SR1-5 is justified, not only in that the quality and performance is well-worth the money, but in that it won't make upgrading your SB3 uneconomical against buying something like the SB+. In fairness I must remind you that the SB+ offers a lot more than just improved sound quality.

If you are contemplating one of these supplies, Paul Hynes make things a bit easier. That's because he offers a fantastic money-back offer if you don't think that any of his products is worth keeping. And you get a very generous 56 days to take up the offer so that you can wait for the items to burn in, and have plenty of time to audition them at your leisure. Having heard the SR1-5 in my system, I doubt whether anybody is likely to want to return one anyway. As if that wasn't enough, the SR1-5 has a three-year warranty covering the cost of repair and return shipping; and is transferable if the equipment is sold on. No 'fly-by-night' Ebay dealer here then!

For the DIYers amongst you, Paul Hynes also offer modules to enable you to construct your own PSU. What you will need to find is a suitable case and connectors etc. I will cover building up such a DIY PSU using a module in a future article. For now, if you own an SB3 or Duet, and you have a decent system, you owe it to yourself to get an SR1-5. Even if you already have a linear supply, the improvement that you will get from the SR1-5 in a decent system will put a big smile on your face. And if you have a Logitech Duet, fear not because the SR1 also comes in a 9 volt version.

My final thought after this review is that if the SR1-5 makes such a difference even going through the 'switcher' in the SB3, how much improvement will the Paul Hynes replacement regulators make when connected directly to a circuit?

Update January 2012

When Paul Hynes wrote and said that he had an SR3-05 power supply that was better than the SR1, I honestly wondered if I would be able to detect an improvement over the SR1-05. The SR1 sounded so good, and was only supplying a digital source (Squeezebox SB3) anyway. As it turned out, I was pleasantly surprised by just how easily I did hear the improvement that the latest incarnation of this power supply made.

Physically, the SR3 looks almost exactly like the SR1, using the same case. The improvements are internal where the regulated power supply has been refined. The SR3-05 is for use a high performance master power supply for any item of equipment that requires 5 volts at up to 3 amps continuous. It will also deliver 20 amp short duration transients.

The regulator circuit is a proprietary discrete component design that has been optimised for very low noise wide bandwidth regulation. The error amplifier noise spec is 0.5 nanovolts root Hz with a noise corner of 1 Hz. The operating bandwidth exceeds 100 MHz and is >50 times wider than the LM317. The output impedance is typically 3 milliohms from DC to 80 KHz and supply line rejection is >80dB over the same bandwidth and is still >50dB at 1 Mhz.

It took around two days of leaving the SR3 powered up for it to sound its best. Initially the sound was a bit thin. However, even from the start, the improvement in the definition was quite clear. Guitar strings (for instance) sounded much clearer, and richer in tone. Bass was tighter, and everything just a bit more focussed. The music was coming out of a 'blacker' background. It's four years since I first heard the SR1 and in that time, Paul Hynes has obviously worked hard to improve what was already an excellent power supply. Is it worth the price of the upgrade if you have an SR1? I guess that is down to your finances but it does sound clearly better. It also works well with the more recent Logitech Touch, and if it is true that the Touch sounds better than the SB3, it's probably well-worth buying for that product.

The SR3 costs 270 UK pounds and the matching DC3C high performance OFC DC lead is 35 UK pounds terminated for the SB Touch or the SB3. Insured carriage and packing in the UK is 10 UK pounds, Europe 22 UK pounds and the rest of the world 33 UK pounds.