Audiocom Superclock 4

[leo]

After the great results of upgrading the Duets external psu to a Paul Hynes PR3, adding Bybee's from Audiocom on the Spdif output and adding a few simple tweaks to the decoupling and spdif buffer I thought it would be worth upgrading the stock clock. The main problem with the stock clock is that the XO isn't the best quality and it also shares the same Hex inverter IC as the spdif out circuit. The plan was to use a better quality XO with low noise regulation and then run straight to the main Xilnx IC direct

I started off using a Burson clock which I've had a while, I added attenuation on the output because it was set for 5v where as the one needed for the Duet is 3v3
I then upgraded the onboard XO to one from Trichord, I then changed that to one from Tent labs which was also tried with a PFM flea module for regulation (set to 3v3).
The improvement with this mod was worthwhile . It improved details and focus of the sound, the stock unit sounded almost compressed in comparison.

I've now had chance to try an Audiocom superclock 4 just to see how it compares against my clock. I felt this clock deserved mentioning on here which is why I started this thread its also easier than faffing about trying to cobble your own clock together.
Details can be found here http://www.audiocominternational.com...products_id=31

Anyway fitting the Superclock 4 is pretty easy, it comes with instructions which are easy to follow, I've added a few pics underneath to give an idea, its a snug fit with a little care and the lid of the Duet can be used.

Mark from Audiocom also sent me a piece of mini 50ohm co-ax to use on the output which I wired underneath the pcb. The screen of the co-ax was fitted to the clock ground only, the main ground of the clock was ran to one of the Duet's VIA's (blue wire) as seen in the pic . DC input came from a pad running straight to the DC input socket of the Duet which is supplied by the Paul Hynes PR3 based PSU.







Dacs used was Modified Buffalo32s (ES9018 Sabre) , Beresford Caiman and Audial AYAII

The Superclock 4 has only had a few days run in but already the difference this brings is more than noticeable. Mark advised the clock reaches 80% of performance after 160 hours, the last 20% can take as long as 300 hours.
In comparison to the other clock I was using the Superclock 4 soundstage is wider , bass is noticeably deeper but the biggest difference is the mids the other clock seemed to have a slight dip in the mid-range where as with the Superclock 4 theres more fullness and voices sound more organic without any loss of detail, the other clock seemed more mechanical (this was noticeable with all dacs)

I'm going to allow to run in some more but so far this is proving to be a very good quality unit , the Duet used as a transport performs seriously good after some work and effort

The stock Touch would have to be bloody good to better this on normal 44.1 material, I can say with confidence it trounces both stock Duet and SB3 (I have both here to compare)

[Gazjam]

Given the good write ups (and the sound hitting my ears!) of how good the Touch is, a comparison between your modded Duet and the Touch would be a great read!

[leo]

Yes, I hope to do that soon Gaz although honestly it'll have to be damn good to better this

If you heard the dynamics coming out of this Sabre dac you'd shat yourself

[John]

The superclock 4 is great I got Mark to upgrade a mf nu-vista and it gave it so much more resolution

[Audiocom AV]

Hi Leo

Thanks for your feedback on the Superclock 4.

An important factor with clocks is the method used to generate the ‘square wave’ output and the ‘phase noise’. We feel that the oscillator used in Superclock is the most ideal because it generates a Pure Sine Wave (analogue) which is then converted to square. Using a Pure Sine Wave oscillator allows circuit and parts to be optimised for lowest phase noise (jitter). This is different to the method used by a TCXO, or canned oscillator which digitally generates the clock signal.

Also, we realised with earlier generation of Superclock that it is the low frequency noise which is crucial to audio performance. This is the sub 10 Hz noise, even as low as 0.1 Hz! The problem with this band of noise is that no commercially available regulator can provide low noise below 10 Hz. The Superclock 4 uses a MAX6126 at the heart of the regulation, this with some circuit tweaking gets down to 1.3µV @ 0.1 to 10 Hz.

OK, enough technical talk; great that you like the sound.

I purchased a Squeezebox Touch for the purpose of evaluation, both with the stock PSU and Teddy Pardo power supplies. The Touch was running for a week 24/7 before I did any serious listening. Used only via the S/PDIF out to a few different DAC’s I was not overwhelmed with the sound quality.

Certainly compared to a fully modified Duet the sound of the Touch playing the same files was tonally harder, grainer, with less depth and flatter sound-stage. I must buy another and try again, perhaps 1 week of burning-in was not enough.

Best,
Mark

[Ali Tait]

Will you be offering mods to the Touch Mark?

[Gazjam]

now THAT got my attention, hope so!

and Leo, this Sabre Dac of yours...was there a thread on AOS about it, its piqued my curiosity.

[Ali Tait]

It's a nice dac,we heard one at Owston with a 6n1p output stage using Sowter trannies.Very good.I'll probably end up making one myself.

[Audiocom AV]

The superclock 4 is great I got Mark to upgrade a mf nu-vista and it gave it so much more resolution

Thanks John.

[Audiocom AV]

Will you be offering mods to the Touch Mark?

Hi Ali

When I initially looked at the Squeezebox Touch I asked, "How can we mod this, it's so thin".

I expect we can do certain upgrades like Bybee, capacitors, connectors, maybe even a Super-regulator, but not the clock. There is not enough space.

We need to revisit the Touch and work out what can be done.

Best,
Mark