|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
98.217.204.116
If you are looking for a DAC with very low distortion in several standard basic measurements, then this DAC is not for you.It has somewhat high 2nd and 3rd order harmonics at low end frequencies, though it has very linear chips: PCM1792, OPA627, and OPA604.
I asked the manufacturer whether mine is defective or whether it is so by design. They wrote that they were sorry that they could not answer.
The graph shows the output of the DAC into a 9.2 kOhm pot (Scosche passive volume control purchased at Amazon) at 20Hz pure sinusoidal tone (Pierre Verany CD track 19 (20Hz, 0dB) one of the tracks for measuring harmonic distortion).
The A/D is Behringer UCA202 (27 kOhm input impedance, max input level 2dBV, according to the manual). The 9.2 kOhm pot was used to attenuate the output signal of the DAC to about 6.5 dB below the max input level of the A/D.
The 2nd and 3rd harmonics in the graph are not products of the A/D. I check it by using other DAC with the same measurement setting.
The harmonic distortion becomes lower as the frequency becomes higher. It also becomes lower as the load impedance becomes higher.
The minimum burn in time of 200 hours did not pass yet. The DAC does not sound fatiguing.
Edits: 12/10/14 12/10/14Follow Ups:
The manufacturer wrote to me that their test result wasSNR : 102dB
THD+N : 0.004%They did not specify test tone frequency, but they know that my question is related to the result of my Sapphire LE at low frequency in the audio band. So I assume the TDH figure is valid for low frequency too.
According to them, the product has various filters. Occasionally, not-so-good measurement results occur. They said sorry about what I have.
Edits: 12/23/14
Due to the digital filtering essentially removing distortion components above 22 kHz (for CD playback), a "high distortion" player could in essence "restore" some of those harmonics (distortion or otherwise) that were removed in the digitization process. And if done in an artful way, it can sound more "analog" than a typical DAC with infinitesimal distortion.
What they "restore" may or may not correspond to the missing harmonics. Different instruments have different amounts of high frequency distortion but after the ADC filters out this information there will be nothing left for the DAC to work with. For example, what is a DAC supposed to do with an 8 kHz sine wave? Was the musical instrument really like a sine wave, or was it more like an 8 Khz square wave? The audio file will be identical for both of these sounds, so if the DAC "improves" the square wave by adding in the clipping, it was harm the sine wave.
If people value "analog" like distortion, then perhaps they need to go unamplified live acoustic concerts. This sets the standard for good sound and it does not sound "analog" or "digitial". It sounds like music.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Fortunately, there are no 8-kHz fundamental frequencies in music so your example is irrelevant. Fundamental frequencies in music rarely go higher than 4-kHz. Consequently, even Redbook CD can reproduce the 2nd, 3rd and 4th harmonics of a 4-khz fundamental.
Best regards,
John Elison
It is necessary to use simplified examples to help people learn. It is not possible to teach calculus to first graders who are learning to count.
Music is what the composers and performers call music. If someone likes to play around with audio oscillators or computers to compose electronic music, who are you to say that it's not music? Does your record library include any electronic music? Mine does.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
The sound of "some" music instruments can be described as fundamental tone and integer multiple overtones.Only "some,", not all.
Anyway, even for music with only instruments with fundamental tone + integer multiple overtones,
Tony Lauck's argument about restoring frequency components above 22kHz holds.
The play back system does not know how much above-22kHz creation is correct.
The psychoacoustic effect of above-20kHz frequencies is not yet proven, as far as I know. Some published test result turned out to be the effect of IMD component below 20kHz.
Edits: 12/14/14
In this product (Styleaudio Sapphire Limited), the harmonics above 22kHz are not likely to be created. The large amount of 2nd and 3rd harmonics appear at low frequencies only.Anyway, the manufacturer should be able to tell whether this large amount of harmonic distortion, which is likely to be accompanied with large amount of intermodulation distortion (IMD), is within their specification after they did "tuning" for subjective sound quality in their design stage.
Edits: 12/13/14
Pierre Verany CD tracks for harmonic distortion measurement
0dB sinusoidal signals
Apple mac mini
optical cable
Style audio Sapphire Limited
analogue output
9.2kOhm passive volume control
signal level attenuated to approximately 4dB below the clipping level of the A/D
A/D
Behringer UCA202
20kHz pic
15kHz pic
10kHz pic
5kHz pic
1kHz pic
500 Hz pic
200Hz pic
100Hz pic
50Hz pic
20Hz
If you purchase this DAC, you may see low THD 20Hz - 20kHz.
Below is the copy of a part of my return request.
The manufacturer refused to answer my question. If the distortion were the result of their careful design for excellent subjective sound quality, they should have proudly answered me that the distortion I observe was within their quality control specification. They did not do so. So the only remaining possibility is that the product I received is defective.
It's a $30 interface, sold through retailers, so Behringer probably gets $18-20 for each sale. A couple minutes on the phone or email with you wipes out their margin completely. Remember, they paid to design the circuit, design and manufacture the populated PCB, the enclosure, and the packaging, *then* paid for the marketing that got retailers to stock it. All this for something that gets them maybe twenty bucks, gross. Would you invest in that? I know I wouldn't, and I admire them for making it happen.
Behringer has been successful making gear that works kinda sorta barely OK for a *tiny* fraction of the cost of the good stuff. Their model has gotten thousands of people into (mostly) pro audio, enabling them to get started far sooner than otherwise. That's good, even if their gear isn't. But a DAC/ADC for $30?? It's pretty amazing if it does anything but burp.
If you want good support for an interface, Metric Halo will respond instantly, either via email or phone, and they know their stuff. But their cheapest interface, the ULN-2, goes for $1,600 and up. If you want that kind of support, go with the good stuff - MH is great, as is Manley with amplifiers. There are others; these are just outfits I have personal experience with.
Someone's got to pay for the time support people (or designers, or company owners) spend conversing with customers. You're just not going to get much for thirty bucks. So enjoy your DAC, and save up for a good one.
WW
"A man need merely light the filaments of his receiving set and the world's greatest artists will perform for him." Alfred N. Goldsmith, RCA, 1922
The Behringer product was used as a measurement device (A/D converter).
I was talking about high 2nd and 3rd harmonics of Styleaudio Sapphire Limited. It is a DAC (without headphone amp, without USB input) costing over $460 at eBay.
To show that the distortion is not from the A/D, I present the output of Audioengine D1 with the same measurement setting (9.2 kOhm pot).
The graph with lower 2nd and 3rd harmonic distortion is the output of Audioengine D1.
42 kOhm pot vs 9.2 kOhm pot
Edits: 12/10/14
Hi,
Looks to me like an undersized coupling cap of the electrolytic persuasion somewhere. Or maybe just one that distorts a lot (any Black Gate in sight?).
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
I cannot see anything wrong from outside. I did not open it because opening the product will void warranty, or void my right to return it to the seller.
I also thought about the possibility that something is wrong in power supply.
Unfortunately, the manufacturer (Styleaudio) refused to answer whether the distortion I have is within their specification or not.
There were products with magazine reviews with similar defect or similar intentional tuning (high 2nd and 3rd harmonics at low frequency). See for example YBA 2 preamp review in Stereophile.
Hi,
I would not class anything below around 0.3% H2 and 0.1% H3 at full scale in a DAC as "defect" (in the light of for example the distortion in speakers or microphones), unless it is very frequency selective.
I suspect it could be something in the PSU, but this is less likely as the feedback would likely kill it. My bet is on a coupling cap.
With electrolytic's (and or certain ceramic's) you must keep the AC voltage across the cap very low, or distortion will result. If the cap is too small in value, you will get distortion only at very low frequencies, improving quickly as the frequency rises, even though there is no obvious LF rolloff.
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
There is a possibility that the power supply circuit has an active ground drive circuit, and its internal series resistance is high and non-linear due to some defect.
Then, even with the feedback in the output stage, the DAC will show nonlinearity in low frequencies? Just my rough guess.
Hi,
Active ground drive problem, maybe. I doubt it, but we are going now deep into design details. My money is on coupling Caps...
Ciao T
At 20 bits, you are on the verge of dynamic range covering fly-farts-at-20-feet to untolerable pain. Really, what more could we need?
The harmonic distortion and IMD is low in mid and high frequencies.See the 19kHz + 20kHz IMD measurement (Pierre Verany CD track 51). The 1kHz ( 20 - 19 = 1) distortion component is below the noise level of the Behringer A/D converter. Such a low distortion is expected in the frequency range 0Hz - 20kHz from PCM1792, OPA627, and OPA604. (Ignore the spikes below 300Hz in the graph. They are noise on the A/D side.)
The roll off at 20Hz (ref. 500Hz) into about 9kOhm load is approximately 0.4 dB.
Anyway, I think the manufacturer should be able to say whether the high harmonic distortion at low frequencies is within their spec or not. They refuse to say anything like that.
Edits: 12/11/14
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: