• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Audio measurement gear

edechamps

Addicted to Fun and Learning
Forum Donor
Joined
Nov 21, 2018
Messages
910
Likes
3,620
Location
London, United Kingdom
I just received my QA401. It's working well (aside from the fact that the experimental ASIO driver doesn't seem to work with REW), especially for noise and distortion measurements, but I'm wondering about frequency response tests. The bundled QA401 software seems to only have rudimentary support for FR tests (Impulse or white noise only, chirps don't seem to be supported on the analyzer side). My situation is made more complicated by the fact that I want to test things like smartphone audio outputs where there is no clear way to get the test signal in aside from storing it in some WAV file and playing it. Any suggestions on how to make such measurements?
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,370
Likes
234,411
Location
Seattle Area
My situation is made more complicated by the fact that I want to test things like smartphone audio outputs where there is no clear way to get the test signal in aside from storing it in some WAV file and playing it.
That is the way I do it. Play white noise, and then configure the FFT in the analyzer to average a ton (16+ samples). That way, the line doesn't look as noisy. It still won't be a flat line and not good for fractional dB measurements but is the only option.

Actually a more tedious method and that is have stepped frequencies play on the mobile device and watch the level meter in the analyzer and write down the values, then plot them in excel.

With Audio Precision you can run external chirps for frequency response measurements. My older AP was actually nicer in that you could put logic in there to analyze the incoming signal and look for tags for start and stop of the measurements.
 

edechamps

Addicted to Fun and Learning
Forum Donor
Joined
Nov 21, 2018
Messages
910
Likes
3,620
Location
London, United Kingdom
I did a full suite of loopback measurements on my QA401, so that I have a baseline for actual measurements. Here are my findings with QA401 1.720, ASIO401 1.1 and REW 5.20 Beta 6:

  • I get the best THD+N numbers with the QA401 directly connected to one of the front panel USB3 ports on my desktop computer. Using another USB port makes the number worse. Using a powered USB hub makes the number worse. Directly connecting it to my laptop makes the number worse, even if the laptop is on battery. The QA401 refuses to work with a HifimeDIY USB Isolator, so I couldn't get a number with that. YMMV, of course.
  • Inspired by the Hypermilling post on the QA401 blog, I measured THD+N for every combination of input and output. It ranges from -97.1 dB (L- → L-) to -94.8 dB (R+ → L+).

Here's an archive containing all the raw measurements. All measurements are made from the L- output to the L- input, single-ended (except for input-output-combinations.mdat). All unused inputs were properly terminated. Single-tone measurements are made at 1 dBFS. Multi-tone measurements are made at -3 dBFS. Frequency response measurements are made at -20 dBFS. Summary:

Maximum output voltage: 0 dBFS = 1.95 V (5.80 dBV)
Maximum input voltage: 0 dBFS = 2.53 V (8.06 dBV) (without attenuator)

  • 48 khz sample rate
    • Noise floor: -115.0 dBV, -118.5 dBVC, -117.9 dBVA, -122.8 dBFS
    • Frequency response: +/- 0.04 dB from 10 Hz to 10 kHz; 0.18 dB down at 20 kHz
    • 997 Hz pure tone: THD -106.0 dB THD+N -97.1 dB
    • IMD: SMPTE -94.4 dB DIN -94.7 dB CCIF -106.1 dB AES17-MD -112.3 dB AES-17 DFD -103.6 dB
    • AES17/IEC 61606-3 dynamic range: 119.0 dB CCIR-RMS, or 19.8 bits (protocol)
  • 192 khz sample rate
    • Noise floor: -102.8 dBV, -112.8 dBVC, -111.7 dBVA, -110.9 dBFS
    • Frequency response: +/- 0.003 dB (!) from 25 Hz to 7 kHz; +/- 0.02 dB from 20 Hz to 10 kHz; 0.18 dB down at 20 kHz; 1.3 dB down at 50 kHz; 10 dB down at 92 kHz
    • 1 kHz pure tone: THD -107.0 dB THD+N -96.2 dB
    • IMD: SMPTE -94.4 dB DIN -95.9 dB CCIF -105.8 dB AES-17 MD -109.8 dB AES-17 DFD -105.2 dB

Linearity (48 kHz, protocol): -121 dBFS (0.5 dB error)

linearity.png


THD+N ratio and absolute N+D vs output level (48 kHz, protocol):

thdn.png

nd.png


UPDATED 2019-03-19: re-did all measurements; added THD+N/N+D vs output level.

UPDATED 2019-03-24: fixed THD+N and N+D plots to show the correct reference.
 
Last edited:

DanB

Member
Joined
Feb 21, 2019
Messages
9
Likes
0
Location
Perth WA
Hi, I'm new to this forum. I'm a hearing researcher, in need of a quality ADC/DAC device for otoacoustic emission testing.
Specifically, I'm after an ADC with an SNR of >125dB, and a noise floor of less than 4uV RMS.
Basically, I think the QA401 will do the trick, but I just wanted to get some feedback on it before purchasing.
Specifically, does it simply show up as a windows sound device (i.e. does it use Windows API)? I need to use it with LabVIEW.

Also, I see from above a few of you have tried the HiFime isolator without success.. I have a few other USB3 isolators that are much more reliable than the HiFime isolator. I will try them.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,524
Likes
37,057
Hi, I'm new to this forum. I'm a hearing researcher, in need of a quality ADC/DAC device for otoacoustic emission testing.
Specifically, I'm after an ADC with an SNR of >125dB, and a noise floor of less than 4uV RMS.
Basically, I think the QA401 will do the trick, but I just wanted to get some feedback on it before purchasing.
Specifically, does it simply show up as a windows sound device (i.e. does it use Windows API)? I need to use it with LabVIEW.

Also, I see from above a few of you have tried the HiFime isolator without success.. I have a few other USB3 isolators that are much more reliable than the HiFime isolator. I will try them.

The biggest issue with the QA is noise floor modulation. In presence of high signal levels the normally low noise floor rises with signal level. If you only care about the low noise level when only lower signal levels are present it still might work.

The RME ADI 2 pro might come closer to what you want. It will cost more money. It is rated at 120 db snr by them, but they usually are honest. It has variable settings on the input and one of them will likely get the low noise you look for or very close. What you are asking for is near state of the art for an ADC.
 

DanB

Member
Joined
Feb 21, 2019
Messages
9
Likes
0
Location
Perth WA
The biggest issue with the QA is noise floor modulation. In presence of high signal levels the normally low noise floor rises with signal level. If you only care about the low noise level when only lower signal levels are present it still might work.

The RME ADI 2 pro might come closer to what you want. It will cost more money. It is rated at 120 db snr by them, but they usually are honest. It has variable settings on the input and one of them will likely get the low noise you look for or very close. What you are asking for is near state of the art for an ADC.
Thanks for the above advice. I had been looking at the RME ADI 2 pro... But at >4 times the price, I was hoping the QA401 would suffice. I also appreciate the changing noise floor issue in the 401.

I was just hoping to find a simple ADC/DAC that uses the AKM VERITA chips, but without the additional features (e.g. headphone amp and multi in out) that go along with pricey sound cards like the RME.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,524
Likes
37,057
Thanks for the above advice. I had been looking at the RME ADI 2 pro... But at >4 times the price, I was hoping the QA401 would suffice. I also appreciate the changing noise floor issue in the 401.

I was just hoping to find a simple ADC/DAC that uses the AKM VERITA chips, but without the additional features (e.g. headphone amp and multi in out) that go along with pricey sound cards like the RME.

Apogee Element 24 is $595. However it is connected by Thunderbolt and Mac only to my knowledge. Not quite the specs of the RME, but fairly good. Maybe the best I know for this amount of money.

https://www.apogeedigital.com/products/element/specifications

Claimed SNR of 119 A wtd. The RME was 120 unweighted.

The Focusrite Clarett 2pre is a bit cheaper and similar specs.
https://focusrite.com/usb-c-audio-interface/clarett-usb/clarett-2pre-usb

I can't say which would be better. Either is going to be close to the best for the $400-600 price range. Unless there is something I'm unaware of which is possible.
 

SIY

Grand Contributor
Technical Expert
Joined
Apr 6, 2018
Messages
10,386
Likes
24,749
Location
Alfred, NY
You might consider the combination of a less expensive sound card with a separate front end like Jan Didden's Autoranger.
 

helloworld

Active Member
Joined
Jul 10, 2018
Messages
220
Likes
194
Thanks for the above advice. I had been looking at the RME ADI 2 pro... But at >4 times the price, I was hoping the QA401 would suffice. I also appreciate the changing noise floor issue in the 401.

I was just hoping to find a simple ADC/DAC that uses the AKM VERITA chips, but without the additional features (e.g. headphone amp and multi in out) that go along with pricey sound cards like the RME.
why not considering an evaluation board for ak5578?
 

March Audio

Master Contributor
Audio Company
Joined
Mar 1, 2016
Messages
6,378
Likes
9,317
Location
Albany Western Australia
I think it might even be 123dB snr as they can paralell adc inputs. I would agree this is a better option than the 401

RME ADI 2 Pro - its 120 dB unweighted 124dB A weighted.

• Signal to Noise ratio (SNR) @ +13/19/24 dBu: 120.x dB RMS unweighted, 124.x dBA
• Signal to Noise ratio (SNR) @ +4 dBu: 119 dB RMS unweighted, 123 dBA

@DanB As Blumlein says the specs you have asked for are state of the art, so it wont come cheap.
 

edechamps

Addicted to Fun and Learning
Forum Donor
Joined
Nov 21, 2018
Messages
910
Likes
3,620
Location
London, United Kingdom
Specifically, does it simply show up as a windows sound device (i.e. does it use Windows API)?

No, it won't. The QA401 will only work with the bundled software and ASIO applications (through ASIO401). It will not appear as a Windows audio device.

Also, I see from above a few of you have tried the HiFime isolator without success.. I have a few other USB3 isolators that are much more reliable than the HiFime isolator. I will try them.

AFAIK, the QA401 doesn't really need USB isolation - it is already isolated internally.
 

DanB

Member
Joined
Feb 21, 2019
Messages
9
Likes
0
Location
Perth WA
Thanks all for the input. I know I'm asking for cutting edge tech., and I guess the next step (given the lack of other akm VERITA devices out there) is to see if I can get the AK5397EQ evaluation board - but as pointed out, it will be tricky.

I have (in my lab) 12 different USB & firewire soundcards. All fall 'just' short of what I need (though in fairness I don't have an RME ADI 2 pro)
This is because I use a Knowles LOW noise microphone (for measuring echoes from the ear canal - noise floor of -20dB SPL), that is also used to measure sounds at 100dB SPL. For most of the soundcards I've tested, their noise floor is about 10dB above that of the mic's EIN.

The solution has been to use low-noise mic pre-amps, but you run into headroom issues.
I'm hoping that with modern tech (VERITA ADCs) I can get rid of the preamp, with sufficient headroom to simplify the whole process.

That said, I will see if I can get hold of an RME ADI 2 pro.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,524
Likes
37,057
Thanks all for the input. I know I'm asking for cutting edge tech., and I guess the next step (given the lack of other akm VERITA devices out there) is to see if I can get the AK5397EQ evaluation board - but as pointed out, it will be tricky.

I have (in my lab) 12 different USB & firewire soundcards. All fall 'just' short of what I need (though in fairness I don't have an RME ADI 2 pro)
This is because I use a Knowles LOW noise microphone (for measuring echoes from the ear canal - noise floor of -20dB SPL), that is also used to measure sounds at 100dB SPL. For most of the soundcards I've tested, their noise floor is about 10dB above that of the mic's EIN.

The solution has been to use low-noise mic pre-amps, but you run into headroom issues.
I'm hoping that with modern tech (VERITA ADCs) I can get rid of the preamp, with sufficient headroom to simplify the whole process.

That said, I will see if I can get hold of an RME ADI 2 pro.
Do you think the Knowles is picking up ambient noise which is why it is above the microphone pre EIN? BTW, the RME ADI 2 Pro doesn't have microphone inputs. Also which Knowles microphone are you using? We might be of more help with the specs to it.
 
Top Bottom