[ogodei]

Introduction

This thread will be a discussion of the Questyle CAS192D DAC.  Perhaps a very short discussion as it doesn’t seem to be widely used in the Head Fi community yet.
 
The CAS192D is a fully balanced, pure dedicated DAC (no volume control, no amp) built around the Wolfson Micro WM8741 chip (which is also used in the PS Audio PerfectWave and Rega DACs).   For lots of pictures and an in depth review (not by me), see here. 
 
The factors that led me to try the CAS192D are:

1. I’m on an unofficial mission to try out DACs with chipsets other than the Sabre ESS9018, to see if I can hear any differences among the chips out there
2. I’m on a separate but related unofficial mission to compare DSD playback from multiple DACs
3. Questyle Audio Engineering has been a firmware partner with JRiver since 2012, so I knew I could count on DSD playback working with my software of choice
4. The quality of the product and technology was obvious from John’s review and from the materials available at the Questyle web site

and

1. I had money and it was new

 

Who is Questyle?

Questyle Audio Engineering is a Chinese company offering audiophile products that I hadn’t heard of until a few weeks ago.  That seems the norm these days but Questyle’s been around a couple of years.  They launched their first headphone amp (the CMA800, designed to complement the Senn HD800) in 2011, and their first DAC (the CAS192) in March of 2012.  Since then they have focused on upgrades to those two products under a confusing array of similar names (CMA800R, CMA800i , CAS192D, CAS192S) and creating one DAC\AMP combo (the also confusingly named “Q192 DAC”).   After announcing the latest versions of their DAC and headphone amp earlier this year they now seem to be focusing on class A and D amplifiers with a ‘Current Mode Amplification’ design which I haven’t really looked into yet.
 
So if they’re any good why did I just now hear about them?  Seems they didn’t have a presence in the U.S. until recently and were selling directly over Amazon and EBay with no advertising.  That’s all over with their forming “Questyle Audio North America”, which I think is one guy in California.  Questyle now appears at the high-end shows, on Facebook, and product reviews are appearing at Inner Fidelity (and here at Head-Fi) with a regularity that screams forced march new marketing effort.
 

Enough of the background stuff

From the manufacturer:
 

Digital-to-Analogue converting
▪  Wolfson flagship DAC chip WM8741，with which the innovative application circuit design is officially affirmed by Wolfson.
▪  DAC power supply utilizes 4 levels voltage-regulating.  
▪  Specially designed OPA627+BJT hybrid LPF stage eliminates TIMD (Transient Intermodulation Distortion).
▪  WIMA FKP2 capacitors are used in LPF and buffer to ensures a clean sound .
▪  IIR (MP) Digital filter eliminates the pre-echo from pulse wave; it is switchable for 5 different kinds of filters and is convenient for users to compare different sound to find their favorite one.
▪  Dual Clock integral upsampling technology: two ultra-low phase noise clocks makes upsampling integrally; the upsampling is optional for start or shut.
▪  Specially designed toroidal transformer by Plitron has the separate digital and analog power supply.  (Plitron is a Canadian partner, I want to read up on this)
▪  Schottky rectifier, 22 PCS of 2200uF Nichicon FG audio capacitances and 20 sets of separate regulators are utilized in the whole device.

Input and output section
▪ Standard Coaxial, Optical and USB input, PCM 192KHz/24Bit. 
▪ 3X clock structure for USB asynchronous transmission, which refers to3 independent ultra-low phase clocks respectively control USB data transmission, and data of integral multiples of 44.1KHz and 48KHz transmission, together with asynchronous technology, perfectly reduce Jitter.
▪ The dedicated USB driver guarantees the stability and compatibility of the whole system.  (WASAPI and ASIO drivers provided)
▪ XLR and RCA Output is optional. 
▪ XLR output utilizes Neutrik connector.
 
Specification

Digital Input USB Input	Optical, Coaxial,, PCM 44.1K-192K/16-24bit signal PCM 44.1K-192K/16-24bit and True DSD Source Code (N.B: “True DSD” is marketing for native DSD playback with JRiver)
Digital Filters in PCM mode Digital Filters in DSD mode	5 groups of switchable IIR (MP) and FIR (LP) digital filters (The filters available are different based on the source file bitrate) No
DAC Chip	WM8741 ( Wolfson chip)
PCM mode up sampling	Dual clock integral up sampling; optional for open or shut
Supportive OS @ USB Input	Win XP,Vista,Win7,Win8 and MAC OS
Music Player @ USB Input	Any player
Kernel Stream Mode Supported by USB Data	WASAPI，ASIO，KS
Output  Interface	RCA , XLR
Output Amplitude	RCA :2.44Vrms,XLR :4.88Vrms (The output amplitude is actually adjustable to two levels, essentially a “Hi” and “Low” toggle.)
Frequency Response	DC-77.5 kHz（192K sample rate, +0,-3dB）
THD+N	< 0.005%  (This actually measured lower, based on tests the lab did specific to my unit. See below.)
SNR	RCA:115 dB,XLR:116 dB
Max External Dimension	330*300*55MM

 

Not from the Manufacturer

The review referenced above indicates the output stage uses D8599 opamps from Analog Devices but the user manual indicates "4 pcs OPA627 operational amplifier dedicated to process signals, not drive any load’.  I am probably confused or reading that wrong, I will look into that at some point.
 

From the unboxing

   
 
 
Shipped from B&O (by a well-intentioned but not too knowledgeable sales clerk) in the original box as packed at Questyle’s labs.  Questyle claims they do a final test of all units at their own facility for quality control, and sure enough a copy of the handwritten test list with THD+N measurements was included along with the unit:
 


 
The manual is printed and hand-sleeved page by page into a binder.  This and the generic DVD-R the drivers arrived on indicate to me a small(ish) company trying to offer a good level of service.
 


 
Well packaged, no shipping damage.  The unit itself is heavy-ish (the power supply as usual), nicely milled aluminum.
 

 
The final “D” left off the front lettering. They may still be working through cases manufactured for the previous model.   Boot up and a peek at the board inside in verifies it is in fact the DSD model.
 

Enough of all that, how does it sound?

That part will need to wait for now.  I fired the unit up for initial testing, it works, and it is playing away by itself for a while. 
 
Will be posting additional notes here as I find them.

 

[ogodei]

Reserved.

 

[project86]

Subscribed. I'm curious to hear if others like this thing as much as I do.
 
Btw, the confusion about the output stage - apparently the original CAS192 (non DSD capable) uses the OPA627+discrete transistor (BJT) for the output stage. They made some tweaks to the CAS192D and discovered the AD8599 sounded pretty much identical in this particular circuit. They mentioned to me something about these particular AD8599 chips being "special" but I'm not clear if that meant specially designed or tweaked by Analog Devices for their purposes (much like Knowles and Sonion will make special versions of a balanced armature driver for JH Audio, UE, etc), or maybe just hand selected and tested for quality. 
 
I'm not too concerned about the distinction because to my ears they sound pretty much identical. 

 

[MLGrado]

I was interested until I read the absolutely incorrect propaganda about DoP.  In no way whatsoever is DSD 'processed as PCM' when using DoP.  PCM is simply the container.  Upon receiving the DoP signal, the DSD file is reconstructed fully and completely.  There is no actual conversion to PCM, processing or otherwise.  When the actual DAC receives the signal, contrary to what is implied by whoever is making and selling this DAC, it knows no difference between DSD transfered via ASIO native vs DoP.  It is the same file, period.  
 
If you look at it on a scope, DSD over DoP looks EXACTLY the same as DSD over native ASIO.  I have a DAC that can do both, as well.  Subjectively there also is absolutely no difference whatsoever.  Furthermore, DSD over DoP could actually be HIGHER quality on the test bench, due to less jitter sensitivity.  

 

[ogodei]

Yeah, the Questyle marketing material focuses on "it's technologically different" rather than "it's a great sounding DAC".  That leads to a few dubious sounding statements about the tech.  I am rather charitably attributing their DSD "processing" and "converting" statements to zealous advertising and a not-so-great translation into English.
 
As you state, in DoP the DSD data is removed from the PCM container, reconstructed perfectly by the DAC, then played.  Questyle is calling out that because they are pulling native DSD rather than the packaged signal they skip the "reconstructing" step, and that they have an entirely separate DSD processing circuit which avoids all the PCM digital filters (duh) and the gain control. From the website:
 
"There are two individual circuits in CAS192D, one is PCM signal channel circuit with 4-pins, the other is True DSD channel with 3-pins. Once the Questyle Audio CAS192D detects the input signal is DSD, its hardware circuit will be automatically switched to the True DSD channel to process DSD signal."
 
I get that the statements could be considered as FUD, but they are by no means alone in jumping on the "native DSD is better than DSD-over-PCM" train.  There seem to be positions being taken about that all over Head-fi.  I haven't heard it myself, that’s one reason I wanted to try this DAC.
 
I have no idea about whether having a separate 3-pin channel to process DSD signals makes a difference, or even if that is unique.  Maybe that’s the way every manufacturer does it?  If it is really a separate process chain it might make a difference.
 
edit:spelling

 

[project86]

Yeah I'm skeptical about the whole True DSD thing too.... except that the CAS192D sounds freakin' amazing with DSD. I have no idea if their Native DSD angle has anything to do with it, or if it's more mundane than that (analog stage well suited to the DSD sound, for example). 
 
I do wonder if they mean Native DSD sounds better in their particular application, compared to what a DoP implementation would sound like. Perhaps they tried both methods, and DoP required some unacceptable (to them) compromise in the signal path.... I know of some designers who don't like to split their signal to feed both line-out and headphone amp sections, because they feel it degrades the sound. Something like this might explain how Questyle finds DoP to be "inferior" despite ending up with an identical file once the unpacking is complete. 

 

[wisnon]

  Yeah I'm skeptical about the whole True DSD thing too.... except that the CAS192D sounds freakin' amazing with DSD. I have no idea if their Native DSD angle has anything to do with it, or if it's more mundane than that (analog stage well suited to the DSD sound, for example).
 
I do wonder if they mean Native DSD sounds better in their particular application, compared to what a DoP implementation would sound like. Perhaps they tried both methods, and DoP required some unacceptable (to them) compromise in the signal path.... I know of some designers who don't like to split their signal to feed both line-out and headphone amp sections, because they feel it degrades the sound. Something like this might explain how Questyle finds DoP to be "inferior" despite ending up with an identical file once the unpacking is complete.

Its all Obfuscation.
 
Lampizator uses completely separate DSD circuit, so in a typical box, you have 2 separate Dacs. Lampi uses NO opamps, just tube output stage and no DAC chip, just low pass filtration and that gives the purest DSD playback imaginable. DSD LOVES tubes. I have found.
 
There is NO compromise to DoP whatsoever, it is Pure DSD in a packaged transport shell.
 
Now, I cant say anything about this Dac's DSD playback capabilities and it SEEMS to be quite good, so that is a welcome addition to the marketplace. too many Dacs out there mess up DSD and its no wonder so manyt are anti-DSD. Poor playback is all they ever heard and tus the basis for tehir wrong conclusion.

 

[allanallan]

Hello guys,

I have Questyle CAS192D DAC and just like other user with CMA800, I also had issues with both fuses.
The question is... I don't know the fuse's amperage. It's not written on the DAC, as it should be. It's only written to use 250V fuse, but not the ampere.
Could someone help me please?

Thank a lot!

 

[ogodei]

  Hello guys,

I have Questyle CAS192D DAC and just like other user with CMA800, I also had issues with both fuses.
The question is... I don't know the fuse's amperage. It's not written on the DAC, as it should be. It's only written to use 250V fuse, but not the ampere.
Could someone help me please?

Thank a lot!

 
 
HI allanallan:
 
The Questyle takes a 250V, 1 Amp, 5 x 20mm fuse.  I suggest glass tube, fast blowing.
 
(EDIT: Per Questyle this actually should be a 250V, 100 mili-amp, 5 x 20mm fast blow fuse)
 
I didn't realize other people were having issues with the fuses.  Or that anyone else on Head-Fi even owned this DAC. 

 

 

[allanallan]

Oh, thanks a lot @ogodei! =)
I appreciate your fast help.
I couldn't find this information, even talking directly to the company.
I've found the same issue from a Questyle CRM800 amp user (page 2 here on head-fi).

I know nothing about fuses.
Is there a brand you suggest? Do you thing is much better to get a Furutech fuse?

Thanks again!

 

[ogodei]

NP.
 
I saw the posting on the CRM800 thread.  The poster there states Questyle indicates use of a 300 milli-amp, slow-blow fuse with the CRM800.   I told you above the CAS192D uses a 1 amp, fast blow fuse, which is a little different. Essentially a 300 milli-amp fuse will blow out sooner than a 1 amp fuse and therefore protect the unit a bit more in case of a power surge.
 
I could have sworn the original fuse in mine was 1 amp, which is what I replaced it with.  It does work but I sent a question off to Questyle here on Head-Fi, maybe he can get an answer from the engineers as to which they prefer.  In the mean time, you can use the 300 milli-amp, slow-blow fuses recommended on the CRM800 thread without issues.  You may need to replace them a bit sooner but that's exactly what they're there for.

I don't have a brand name suggestion.  These fuses suggested by grs108 will work fine.   I don't buy brand name fuses because I don't understand how they could affect the sound or performance of the equipment.  I personally just buy a bunch of cheap ones and even if I get a bad batch I've saved money.   If purchasing a brand name would help with your enjoyment of the equipment I would say go for it but technically I don't believe it will affect the hardware's performance.

 

[allanallan]

I was a bit confused about it.
Before I've written it here, I contacted Questyle on FB and was told me to use the fuse "T313MAL/250V", with 313mA, slow blow.
I didn't find info about it on Google.
I could be wrong, but I searched about it and 313 seems to be the fuse series, but not the amperage (they have many different amparage on this series): http://www.littelfuse.com/products/fuses/cartridge-fuses/3ab_3ag-6_3x32mm-fuses/313.aspx
I'm waiting for an answer too.

So I will get 300mA fuse in the mean time.
Thanks again for trying to help me!

 

[ogodei]

  I was a bit confused about it.
Before I've written it here, I contacted Questyle on FB and was told me to use the fuse "T313MAL/250V", with 313mA, slow blow.
I didn't find info about it on Google.
I could be wrong, but I searched about it and 313 seems to be the fuse series, but not the amperage (they have many different amparage on this series): http://www.littelfuse.com/products/fuses/cartridge-fuses/3ab_3ag-6_3x32mm-fuses/313.aspx
I'm waiting for an answer too.

So I will get 300mA fuse in the mean time.
Thanks again for trying to help me!

 
Actually the 313 is the amp rating in milli-amps.   That spec calls for for 250V, 313-milli-amp, 5 x 20mm slow blow fuse.  Closest thing would be these from Amazon which will work absolutely fine, I just ordered some myself.  You could also use the 300 mA fuses from the link above with absolutely no problems.
 
Thanks for posting the question, now I know better!

 

[Questyle]

Hi,
Just to clarify the fuse issue. The Questyle CMA800R uses a T313mAL/250V which is a little hard to find here, a 300ma 250V slow-blow will work fine. The Questyle CAS192D uses a 100ma 250V fast-blow fuse. It would be very unusual to blow a fuse on the CAS192D, but every unit comes with a spare fuse.

 

[ogodei]

Thanks for the clarification, updated the thread above.