AliExpress Cirmech WM-013 CS4398 DAC - review and measurements

[IVX]

oh, there is some old-school muting that also may add tons of distortions. CS43198 has muting too, hence, LPF-opamp is really redundant

 

[audio_tony]

What is DBT, please?

Double Blind Testing

 

[audio_tony]

Here's another variant of these cheap CS4398 boards. This one has been hacked around a bit in an attempt to improve the performance, but it's just as bad as the one in the OP.
I tried taking the output directly from the DAC into a separate filter stage with no improvement.

 

[pma]

oh, there is some old-school muting that also may add tons of distortions. CS43198 has muting too, hence, LPF-opamp is really redundant

Yeah but the board has no muting. The scheme is from Cirrus EVAL board.

 

[pma]

I have put the DAC into the preamp case:

There is a lot of wiring, both SE signal gnd and PSU gnd loops, so the mains noise background is now worse. Still, not unacceptable, with peaks below -120dBr. SINAD has rather improved , and SNR is now higher (= better) at 106 dB. It will serve for the S/PDIF signal from the TV set, so good enough for the purpose.

 

[Sokel]

What is DBT, please?

Double Blind Test.

 

[pma]

RightMark Audio Analyzer Report attached below.

 

[sabih]

I have the same DAC board. Is there any tweak available to improve SINID of this DAC board?

 

[fordiebianco]

Many thanks for this Pavel. Super interesting and enlightening. Please keep them coming.

 

[jamescarter1982]

has anyone ever measured the xkitz active crossover boards ?

 

[EJ3]

Useful hint: Try to be a little more polite in the future.

That is a SKILL that many need to work on (myself included).

 

[DVDdoug]

What is DBT, please?

Double Blind Testing

Just adding to that in case you don't know about these things -

This isn't necessary unless you are hearing a problem, or if you think you are are hearing a problem, or think that maybe you're hearing a problem. If there is nothing to listen for, there's no need for a double blind listening test!

You'd need to compare it to something. Hopefully to something with "better specs". If you compare it to a DAC with a buzz or some other defect, there's no point.

What is an ABX Test?.

And Amir has a very-good 40 minute video: Controlled Audio Blind Listening Tests.

 

[Hayk]

How about 16bit 48khz -60db ?
I suppose these measurements are with 24bit ??? Khz

 

[sabih]

@pma are you sure this is a cs4398 board. Because the chip in your board does not seems to be cs4398

CS4398 actual chip size is different. It has horizontal layout. (28 TSSOP)

you chip says: 4398CN.
This chip package is not mentioned anywhere in the datasheet.

I think it is chinese version of the chip and not the original chip.

 

[JSmith]

@pma are you sure this is a cs4398 board. Because the chip in your board does not seems to be cs4398

CS4398 actual chip size is different. It has horizontal layout. (28 TSSOP)

Fair point... can see what you mean on this one;

CS8416 + CS4398 Decoder Board Audio DAC 24Bit 192KHz w/ Coaxial Optical Ports | eBay

CS8416 + CS4398 Decoder Board Hifi Audio DAC Board Module 24Bit 192KHz w/ Coaxial Optical Ports. - Original chip: Original chip for CIRRUS LOGIC 24BIT 192KHz specification, and compatible with lower specification, no driver required, super anti-interference ability.

www.ebay.com

But that board says V1.3 and the one in the OP says V1.4... which has this in it;

https://www.taobao.com/list/item/678054057405.htm

This is the V1.4 board;

HIFI发烧级CS8416CS4398芯片24BIT192KHz同轴光纤DAC解码板：音质巅峰，耳朵盛宴！-解码器-淘宝好物网

音响迷们注意啦！这款HIFI发烧级CS8416CS4398芯片24BIT192KHz同轴光纤DAC解码板，让你在家就能享受到录音棚级别的音乐盛宴！无论是古典乐还是摇滚，都能还原每一个细节，仿佛置身现场。现在入手，享受音乐的极致魅力，成为朋友圈里的听音大师！

goods.taobao.com

Which specifically refers to WM-013

Not genuine, counterfeit?

4398CN | 4398CN | 4398CN Stock | OMO Electronic

Global distributor of + 4398CN + , Stock: Yes, shipping: Can Ship Immediately. OMO Electronic, Your trustworthy partner.

www.omo-ic.com

https://www.mouser.com/c/semiconductors/audio-ics/audio-d-a-converter-ics/?m=Cirrus%20Logic&series=CS4398

JSmith

 

[pma]

I have the same DAC board. Is there any tweak available to improve SINID of this DAC board?

Yes, definitely, but not simple. The ceramic capacitors in the yellow circles must be exchanged for the NP0/C0G types. Lot of filigran job with high probability of PCB damage. And the new capacitors would be larger.

CS4398 actual chip size is different. It has horizontal layout. (28 TSSOP)

Cirrus produced it in both packages, you may find it in older datasheets. QFN package was removed in 2021. I would not speculate on op-amp or DAC fake. Op-amp fake suspicion was debunked, the op-amp measures identically as a genuine part with known origin. I assume the same for the DAC. The issue is in high K ceramic capacitors used in the op-amp LPF balanced-to-SE output stage.

On QFN package, see release note F3 below:

As I wrote I placed the DAC into one of my old preamps. The preamp has 3 RCA inputs and one XLR input. The DAC now occupies IN3 RCA input. Below is the comparison of the IN1 (analog linear input) and IN3 (S/PDIF Cirmech DAC) shown in THD vs. frequency plot. One more note - distortion with the preamp is lower than that measured by E1DA Cosmos ADC directly at DAC output. This is because of low input impedance of the Cosmos (about 1k5) compared to 10k of the preamp. The DAC output is less loaded with the preamp and the output coupling electrolytic capacitor distortion is lower then.

The measured distortion degradation is huge, but is it audible?

I made some auditioning yesterday, to me it sounds like any "normal" DAC.

 

[pma]

As a matter of fact, I am satisfied with this DAC. At a cost of a common lunch, we get a 24bit/192kHz DAC that works immediately after connecting to the power supply and input S/PDIF signal. No need for software updates and changes, that often end up in a SW crash. I made level matched A/B tests (easy as the 2.3V output is quite standard) with a 112dB SINAD DAC and with Vincent CD-S3 player and I hear no differences.

 

[Sokel]

has anyone ever measured the xkitz active crossover boards ?

If their specs are right (THD+N: 0.0086% but at what conditions, etc? ) for classic 1kHz/20Hz-20Kz BW, etc, even this one is better.

 

[sabih]

As a matter of fact, I am satisfied with this DAC. At a cost of a common lunch, we get a 24bit/192kHz DAC that works immediately after connecting to the power supply and input S/PDIF signal. No need for software updates and changes, that often end up in a SW crash. I made level matched A/B tests (easy as the 2.3V output is quite standard) with a 112dB SINAD DAC and with Vincent CD-S3 player and I hear no differences.

Are you sure that you are unable to hear any difference between this DAC and high SINAD DAC?
Don't you think high SINAD DAC will sound more cleaner and clearer?

 

[pma]

Are you sure that you are unable to hear any difference between this DAC and high SINAD DAC?
Don't you think high SINAD DAC will sound more cleaner and clearer?

Only a DBT test would tell and I am sceptical for the reason that I have performed dozens of level matched ABX tests with equipment with much higher distortion that did not result in audible difference in a DBT test. Any sighted test is pointless. Those parameters wars are also a kind of marketing.