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

S/PDIF and other hiccups. I had set my 160GB AIFF-loaded iPod Classic in Cambridge Audio's iD100 digital-direct dock to endless repeat and leashed up the latter via Chris Sommovigo's very best Tombo Trøn cable. When I returned many hours later the iPod was still streaming but the u:c:a no longer made sound. I shut it down and powered it back up. Instantly the sound was back. Had the S/PDIF input receiver lost signal lock? Listening confirmed intermittent drop-outs already in the first track. Four tracks later those interrupts broke the handshake. Full mute. Something about the protocol wasn't 100% stable.

"I never encountered this problem with our CD drives as data source. As I am an audiophile hardliner I haven’t tried these new media although I did mention it in the manual. The only thing I can imagine is that the sample rate clock of the docking station varies over time. The new Wolfson input receiver works completely different than its Crystal counterparts. Whereas the CS8412/14/16 constantly track the incoming sample rate with an analogue—jittery—PLL, the Wolfson estimates the clock rate only once and then operates with a fixed-frequency stable quartz clock, thus eliminating jitter completely. If the incoming data rate varies too much, the elastic buffer overloads and the hiccups occur. Nevertheless if the operating borders are too tight for operation with less than ideal sources, we have to think again for the full production units. If the hiccups persist with a CD drive, then something must have gone wrong during transit."


Reality can bite on multiple fronts. Engineering a USB DAC with headphone output—two functions predicting quite obvious usages—and not conducting any pre trials for flawless cooperation with various computers and iDevices would seem peculiar. Just how new are those media really? The first iPod was introduced in 2001. Now it was 2012. Who (cough!) would even be interested in a €1.190 DAC and still have a CD transport?


"I am an easy victim of my own prejudices. Therefore it takes time before I jump on new opportunities. This was even the case for USB audio. These days I hardly spin a CD. Shame on me! I did some experimenting and intensive internet research. From that I have to qualify my first remark that it has to do with largely varying sample rates. There are no reports of such things in conjunction with the Cambridge Audio docking station whatsoever. On the other hand there are no complaints about the behavior of Wolfson's WM8804/5 chips either. As a reference one might consider Twisted Pear's widely used Opus DIY board. Their arrangement is very similar to ours except for the input circuitry. Without being able to reproduce the effect it seems clear that it has something to do with the galvanic isolation circuit. Here modifications are easily made." Ivo agreed that I should wait for a replacement to confirm that the hiccups were no more.

"I have another pre-production u:c:a ready. It is just waiting for the back and front panel because we have to make those separately to get them quicker. The S/PDIF input is somewhat modified. It turned out that the transformer for the digital audio data transmission inverted the digital signal. In addition we noticed a slight attenuation and softening of the edges. Although it was impossible to reproduce the described error, the problem is very likely rectified. Note that S/PDIF is bi-phase coded. That means absolute polarity is of no importance. Yet there are violations in the preamble of the S/PDIF signal stream which may cause trouble in certain very rare cases. Result? The u:c:a now works with everything that looks like S/PDIF as long as the sample frequency is within +/-1% of the target frequency. So far I tested the redone S/PDIF input with various CD/DVD players, set top boxes etc. Some produced quite ugly-looking S/PDIF waveforms on the scope but the u:c:a had no problems playing with any of these. I also visited HiFi Studio Liedmann in Bochum. Liedmann is an official Cambridge Audio dealer in Germany. For the test we took a brand new—out of the box—iD100 docking station and an iPod touch. The link to the u:c:a was made with a standard 75Ω digital cable. Analogue out was through a Vincent preamplifier and some other brand tube power amp. Playback was flawless. No problems discovered."


Inspired perhaps by Acoustic Revive, KingRex makes two interesting divide 'n' conquer USB cables whose basic guts are sourced from WireWorld and Oyaide. What the team in Taiwan does then is create a reverse Y-cable with two USB plugs (discrete power and signal) on the computer side which naturally combine at the DAC end. The special KingRex twist comes when the power lead doesn't plug into the computer but their optional uPower battery supply. This neatly eliminates dirty USB power and works like the proverbial charm on my usual Eximus DP-1 whose XMOS USB transceiver runs on bus volts. With the u:c:a however the iMac couldn't see the load. Until I switched to a standard Telos USB cable. This instantly brought up the Linnenberg as USB hiFace. I reverted to the KingRex and again the u:c:a went MIA. On a lark I now moved the USB power lead from battery to iMac. Voilà, connection. Huh?

Neither Ivo nor I had any rational (him) or irrational (moi) explanation. "I looked at the protection circuitry of the M2Tech module but this strange behaviour cannot originate there. Voltage on the USB cable's D+/D- lines is limited to -0.6V to +15V with respect to ground. In our case not a single device on the USB input board connects to the 5V pin of the USB connector. No current can flow in or out of it. Hence it makes no difference if some voltage does appear on that pin or not. Perhaps that cable needs some current flowing out of the battery? That would be a question for KingRex. On the other hand a powered USB cable becomes useless if a component like our u:c:a requires no power." Interestingly the response from KingRex when alerted to this behavior was to revisit the wiring of their cable so that it would "create a proper ground reference" even in u:c:a-type scenarios.


As promised, Ivo's second unit had lost the hiccups. Now signal lock with my docked iPod was solid for 10 non-stop hours of endless random play and listening confirmed that the intermittent mini dropouts had been successfully banished. Linnenberg knew his engineering. He just needs to venture out into the brave new world of portable audio devices more often...
Enlarge!