CHANGSTAR: Audiophile Headphone Reviews and Early 90s Style BBS

  • December 31, 2015, 09:39:17 AM
  • Welcome, Guest
Please login or register.

Login with username, password and session length
Advanced search  

Author Topic: Stoner Acoustics UD120 Measurements  (Read 3929 times)

0 Members and 1 Guest are viewing this topic.

firev1

  • Cynophobic Puss
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +52/-0
  • Offline Offline
  • Posts: 490
Stoner Acoustics UD120 Measurements
« on: December 13, 2014, 09:50:58 AM »

inb4 crappy phone pix
stoneracoustics.blogspot.com

I did a review of the UD110mk2 a while back and while performance was pretty okay and sound was good, I thought it could use more refinement. The DAC with shipping was really cheap and worked right out of the box with android devices and all that other nonsense. Only thing was that it had mediocre power supply rejection and significant low frequency jitter but hey, its really cheap. Stoner Acoustics took the feedback they got from customers and came up with the UD120 which boasts better performance and again at a really great price of 79USD WITH shipping all in.

Design wise there really is absolutely nothing to look at, just a DAC with an ugly plastic mold case with a short wire out for the socket. Plug and play with Macs and Linux, Windoze with Drivers. Installation is fuss free. On Win8 I think you would see a 32 bit(LOL) option for audio as well.

For the nerds and dorks, the UD120 sports a PCM5102A which has selectable interpolation filters.  TPS7A4700 LDO is used for regulation and it has an output Z of 470ohms(from website). It supports all PCM rates under the sun.

Subjectively, the DAC seems to do well on first listen. Unlike the GEEK on TCM filter, seems acoustically lightweight. Nothing to hate about so far. I think I prefer the GEEK to it though on first listen it might sound slightly holographic in terms of imaging.  Need to do more direct comparisons.

For me so far on first impressions, GEEK-> UD120 -> ODAC Been exclusively listening to it on the Wyrd because of practical considerations rather on a sonic basis.

Copy Pasta Rig:
Acer Win8 Laptop-> ASIO -> UD120-> Short 3.5mm to RCA cable-> EMU0404 via RCA to XLR adapters-> Fujitsu Laptop
Software: Foobar with RAM buffering, WaveGene, WaveSpectra, ARTA

DISCLAIMERS: Not science, YMMV. Also do take note of the rising noise with wide bandwidth measurements of the 0404.

The UD120 sports a minimum phase filter with what it seems to be less post ringing than the GEEK OUT IEM's filter. Funny glitch at the rising slope, no idea why, does not come up with all other DACs on hand.



Reconstruction filter test. White noise and a spectrum of a 19khz tone played at 44.1khz were played and captured at 192khz on the ADC side to characterise the DAC's interpolation filter. Probably as a price for the lower post ringing the filter has significantly less stopband rejection so there is some aliasing and later you will see, poorer high frequency performance. If I were to venture a guess, I would think that this is the 4X interpolation filter.



With the low level resolution test with a 1khz sine wave at -90dbFS, the UD120 showed pretty decent performance similar to the GEEK albeit with a little more spuraie. Both are better than the ODAC in this case and it shows in the waveforms, slightly noisy but otherwise good performance.



-90dbFS 1khz sine, Above: 16 bit Below: 24 bit

EDIT: Unlike the last round, 50hz component does not modulate the low level signal.


1khz THD tests, 2nd harmonic distortion profile with perfectly decaying harmonics. Tests as well as the GEEK on TCM filter but with significantly less low level spuraie.
« Last Edit: December 13, 2014, 11:41:36 AM by firev1 »
Logged
Time spent on enjoyment is not time wasted. - someone

firev1

  • Cynophobic Puss
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +52/-0
  • Offline Offline
  • Posts: 490
Re: Stoner Acoustics UD120 Measurements
« Reply #1 on: December 13, 2014, 09:54:00 AM »
With Twin tone testing, the UD120 performs similarly to the GEEK but with less low level spuraie. Its a significant improvement from the last DAC, smashing the ODAC.


In the high frequency test though particularly with the difficult 19khz 20khz tests, the UD120 suffers a dip in performance. Probably due to the choice in filtering. Still not too bad though it performs slightly worse than the ODAC in the 19khz 20khz HF test but better in the 13khz+14khz one.


In the 24 bit Jtest, the UD120 shows pretty lowish jitter but with some sidebands far from the centre frequency. Implementation is not ideal but still good considering the price as the UD120 improves significantly with the addition of the Wyrd in the test chain. The UD120 does not show any sign of warmup issues as well.

24 bit JTest, comparison with and without the Wyrd.
Conclusion: The UD120 is a real no bullshit DAC at a no bullshit price.
Logged
Time spent on enjoyment is not time wasted. - someone

tomscy2000

  • The Bill Nye of CIEMs
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +46/-1
  • Offline Offline
  • Posts: 190
Re: Stoner Acoustics UD120 Measurements
« Reply #2 on: December 13, 2014, 03:33:04 PM »
Good stuff.  :money:

The PCM5102A has a built-in PLL but runs the oversampling at a lower factor; does the UD120 use the PLL mode or does it use an external MCLK?
Logged

firev1

  • Cynophobic Puss
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +52/-0
  • Offline Offline
  • Posts: 490
Re: Stoner Acoustics UD120 Measurements
« Reply #3 on: December 14, 2014, 06:03:00 AM »
Hmm, I'm not too sure but I do know some sort of PLL or clock synthesizer has to be there to feed MCLK. My guess would be at the Saviaudio usb chip they use. The one oscillator on the board is pretty far away from the PCM5102a chip. EDIT 27/12/2014: Yes, the Saviaudio's PLL supplies the MCLK needed for the PCM5102a
« Last Edit: December 26, 2014, 04:20:07 PM by firev1 »
Logged
Time spent on enjoyment is not time wasted. - someone

firev1

  • Cynophobic Puss
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +52/-0
  • Offline Offline
  • Posts: 490
Re: Stoner Acoustics UD120 Measurements
« Reply #4 on: December 18, 2014, 03:08:08 PM »
Stoner Acoustics never got back to me sadly. Some pics though. No idea what the funny numbered parts are aside from one of the regulators. Awfully simple layout though with the outputs resistor coupled.
Logged
Time spent on enjoyment is not time wasted. - someone

tomscy2000

  • The Bill Nye of CIEMs
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +46/-1
  • Offline Offline
  • Posts: 190
Re: Stoner Acoustics UD120 Measurements
« Reply #5 on: January 09, 2015, 02:19:19 PM »
Found out the UD120 is running the SA8133 clock generator, so the MCLK, BCLK, and LRCK are all clocked from that component. No idea how well it theoretically performs, but it looks good from here.

The UD120 serves as a good base for an ultra-tiny, thumbstick DAC/amp --- imagine an "UD130" that replaces the PCM5102 with a PCM5242 and directly couples the output into an ES9601.

The PCM5242 is not all that different from the PCM5102 except that it's now differential Vout, so it can couple directly with differential input headamp ICs, like the TPA6120A2, MAX97220A, and ES9601. Because of differential output, a little bit of the charge pump noise is cancelled out and the part gains a couple dB of SNR as well. A simple change leading to improvement by TI. The "4" in the PCM5242, much like the PCM5142, means it also has some embedded ROM inside, so people can code in a custom digital filter, like an IIR minimum phase, for instance.

My assumption is that such a design would require minimal extra BOM cost (a few more linear regulators for the headamp IC, plus the IC itself), so the cost can definitely be at ~$99. Something like this would run rings around the Dragonfly and ALO Key.
Logged

firev1

  • Cynophobic Puss
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +52/-0
  • Offline Offline
  • Posts: 490
Re: Stoner Acoustics UD120 Measurements
« Reply #6 on: January 09, 2015, 02:32:33 PM »
Stoner has pretty much confirmed that the UD120 would be their last SE DAC, its safe to assume that their next few products are gonna be differential outputs. Measuring a few of these BB dacs tells me that the probable main compromise is indeed using the internal digital filtering. Does not seem to provide enough image rejection. A custom filter would really improve things.

Jitter is good but not optimal compared to say the GEEK based devices. Power supply rejection was a huge improvement but from the measurements with the Wyrd, has yet to be fully optimal. Still have to give them for getting out the performance with so little parts though.

Another thing that Stoner has in works is their ESS9061 integrated amp with I think balanced inputs and a stepped SMD volume pot.
Logged
Time spent on enjoyment is not time wasted. - someone

tomscy2000

  • The Bill Nye of CIEMs
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +46/-1
  • Offline Offline
  • Posts: 190
Re: Stoner Acoustics UD120 Measurements
« Reply #7 on: January 09, 2015, 02:34:36 PM »
Sounds like I'm thinking along the same lines as Stoner, then! I'll look forward to those developments.

Indeed the integrated filters on the TI chips kind of suck. They're one of the reasons why the Meridian Explorer sucked.

Can't do too much about the jitter performance with off the shelf clock generators, I suppose.
Logged

firev1

  • Cynophobic Puss
  • Able Bodied Sailor
  • Pirate
  • ***
  • Brownie Points: +52/-0
  • Offline Offline
  • Posts: 490
Re: Stoner Acoustics UD120 Measurements
« Reply #8 on: January 09, 2015, 02:42:13 PM »
From what I gathered they are still learning how to test their products though. At least their product's performance is somehow a step above other Chinese outfits.
Logged
Time spent on enjoyment is not time wasted. - someone