Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Equipment & Tools From test equipment to hand tools

Audio grade ADCs, PCM4222, AK5572, CS5381, etc...
Audio grade ADCs, PCM4222, AK5572, CS5381, etc...
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 21st May 2020, 12:51 PM   #291
JensH is offline JensH  Denmark
diyAudio Member
 
Join Date: Jul 2009
Audio grade ADCs, PCM4222, AK5572, CS5381, etc...
If you want something with a low noise up to 96 kHz you should take a look at the AK5397.
I used it initially, but changed to the AK5394A because the distortion of the AK5394A is lower. But in terms of noise, the AK5397 is better.
  Reply With Quote
Old 21st May 2020, 05:25 PM   #292
syn08 is offline syn08  Canada
diyAudio Member
 
syn08's Avatar
 
Join Date: Aug 2005
Location: Toronto
Quote:
Originally Posted by JensH View Post
If you want something with a low noise up to 96 kHz you should take a look at the AK5397.
I used it initially, but changed to the AK5394A because the distortion of the AK5394A is lower. But in terms of noise, the AK5397 is better.
I did, and the AK5397 is IMO not up to the task of distortion performance for instrumentation. -108dB is a non starter.

AK5394 is slightly better (still not good enough, IMO), but it's no longer in production, anyway.
  Reply With Quote
Old 22nd May 2020, 08:39 AM   #293
phofman is offline phofman  Czech Republic
diyAudio Member
 
Join Date: Apr 2005
Location: Pilsen
syn08:

Quote:
If you can disable the STATUS byte with an SPI register write before entering frame-sync slave mode, then there is no issue with using a 24.576-MHz SCLK to daisy-chain two ADS127L01 devices. This would allow you read exactly 48 bits per frame at 512 kSPS. The note in the data sheet assumes that most customers will exclusively use frame-sync mode and will not have the ability to read/write registers prior to data collection.
https://e2e.ti.com/support/data-conv...354268#3354268

Apparently there should be a way, though a bit more complicated.
  Reply With Quote
Old 22nd May 2020, 03:03 PM   #294
syn08 is offline syn08  Canada
diyAudio Member
 
syn08's Avatar
 
Join Date: Aug 2005
Location: Toronto
Quote:
Originally Posted by phofman View Post
syn08:



https://e2e.ti.com/support/data-conv...354268#3354268

Apparently there should be a way, though a bit more complicated.
Never thought of setting for strict 48bit per frame in SPI mode, then switching to frame sync mode. I have to look closer to this scenario, my gut is telling me it's not a breeze to set up and operate since, to my experience, each START re-asserting is resetting the 32bit mode. Since such a re-assert is required for each parameter change in frame sync mode (oversampling rate, high speed/low power, etc...) it may that this configuration would not be a one time per session thing. Ultimately, it's doable, of course. BTW, in this mode you are stuck in Left Justified mode, since there is no room for I2S without dropping the LSB. No a big deal for instrumentation, but not really an option for strict audio ADC where I2S is the standard.

But otherwise, IMO 384KHz sampling is not a big deal, actually for all I can tell, software like ARTA doesn't even support more than 192KHz. Nice to have, but I wouldn't sweat a bit over it. If >192KHz is really required, there are better options like the 20bit SARs in SPI mode, with the addition of a small CPLD to convert to I2S. I have here an LTC2378 clocked at 100MHz (ok, pushing it a little over the data sheet) and a 4nS MAX V CPLD that spits out 784KHz 20bit I2S without a sweat.

Last edited by syn08; 22nd May 2020 at 03:10 PM.
  Reply With Quote
Old 22nd May 2020, 03:33 PM   #295
zfe is offline zfe
diyAudio Member
 
Join Date: Dec 2014
e.g. audiotester (audioTester) supports up to 784KHz sampling rate.
  Reply With Quote
Old 22nd May 2020, 07:33 PM   #296
phofman is offline phofman  Czech Republic
diyAudio Member
 
Join Date: Apr 2005
Location: Pilsen
REW supports 1.5MHz (on current linux build, JohnPM has reported a windows REW build to run fine at 768kHz with the RME soundcard), Visual Analyzer tested to work up to 2.3MHz in linux wine (i.e. would run even better natively in windows), octave's (i.e. open source matlab) playrec (ASIO build) and every linux analyzer can run much higher than 192kHz (either out of the box or after trivial change in code). Many soundcards support 384kHz IN/OUT, some do 768kHz. Unfortunately they use the AK5xxxx ADCs whose gain drops at 190kHz (as the figure I posted shows) - but the existing drivers can feed that data rate, even in windows.

It's only a matter of (IMO short) time before every analyzer with healthy support/development will allow this. Adding higher samplerates is simple IF the software is coded soundly. It took a few hours between my question to John and him releasing the build which performed smoothly up to 1.5MHz (and likely higher, had John included a higher option in his java code, likely just adding a number to a static list).

I will explore the options as I need to sample several harmonics for the digital distortion compensation (ADS127L01 should be able to precisely sample H2, H3 for 75kHz fundamental). The ADCs and supporting chips (the balanced buffer, voltage reference, all the LDOs, clock-generator chip) are quite expensive (esp. in smaller quantities) so if a software (i.e. zero variable cost) can allow using them to their maximum capabilities, I want to explore that. I assume a relatively simple python code developed and debugged in a comfortable IDE will take care of all the control.

The I2S (format notwithstanding) support would be a major breakthrough for me because the PCM peripheral on RPi has 32 frames-long input and output buffers, significantly easing timing constraints for servicing IRQs (I2S flow is continuous, unlike the DRDY-timed SPI). The RPi I2S driver is well tested and requires just changing the hard-coded max_rate number from 384000 to 512000.

These are my objectives, everybody has them different which is perfect.

Last edited by phofman; 22nd May 2020 at 08:01 PM.
  Reply With Quote
Old 22nd May 2020, 10:13 PM   #297
syn08 is offline syn08  Canada
diyAudio Member
 
syn08's Avatar
 
Join Date: Aug 2005
Location: Toronto
Quote:
Originally Posted by phofman View Post
REW supports 1.5MHz (on current linux build, JohnPM has reported a windows REW build to run fine at 768kHz with the RME soundcard), Visual Analyzer tested to work up to 2.3MHz in linux wine (i.e. would run even better natively in windows), octave's (i.e. open source matlab) playrec (ASIO build) and every linux analyzer can run much higher than 192kHz (either out of the box or after trivial change in code). Many soundcards support 384kHz IN/OUT, some do 768kHz. Unfortunately they use the AK5xxxx ADCs whose gain drops at 190kHz (as the figure I posted shows) - but the existing drivers can feed that data rate, even in windows.

It's only a matter of (IMO short) time before every analyzer with healthy support/development will allow this. Adding higher samplerates is simple IF the software is coded soundly. It took a few hours between my question to John and him releasing the build which performed smoothly up to 1.5MHz (and likely higher, had John included a higher option in his java code, likely just adding a number to a static list).

I will explore the options as I need to sample several harmonics for the digital distortion compensation (ADS127L01 should be able to precisely sample H2, H3 for 75kHz fundamental). The ADCs and supporting chips (the balanced buffer, voltage reference, all the LDOs, clock-generator chip) are quite expensive (esp. in smaller quantities) so if a software (i.e. zero variable cost) can allow using them to their maximum capabilities, I want to explore that. I assume a relatively simple python code developed and debugged in a comfortable IDE will take care of all the control.

The I2S (format notwithstanding) support would be a major breakthrough for me because the PCM peripheral on RPi has 32 frames-long input and output buffers, significantly easing timing constraints for servicing IRQs (I2S flow is continuous, unlike the DRDY-timed SPI). The RPi I2S driver is well tested and requires just changing the hard-coded max_rate number from 384000 to 512000.

These are my objectives, everybody has them different which is perfect.
I'm sorry Pavel, you lost me shortly after the beginning... What java code did John (I presume the TI guy?) release, what is 1.5MHz (I presume it's not the ADS127L01 sample rate), which sound cards are doing 768KHz I/O, what is this "ADC gain drops at 190kHz", I thought the 24bit AK ADCs sampling rate is 200k max.

Otherwise, I forgot you are an RPi afficionado, I must admit I'd rather include in a design a CPLD/FPGA/STM32 chip then a generic RPi board. The only scenario I would use an RPi would be a stand alone instrument, with it's own display and entry devices, when the whole UAC2 **** is no longer required. But that's just me...
  Reply With Quote
Old 22nd May 2020, 11:47 PM   #298
1audio is offline 1audio  United States
diyAudio Member
 
Join Date: Mar 2004
Location: SF Bay Area
Audio grade ADCs, PCM4222, AK5572, CS5381, etc...
I believe he is referring to John at REW REW - Room EQ Wizard Room Acoustics Software not TI.

AKM has ADC's including the AK5397 that will sample at 768 KHz however their frequency response rolls off below 200 KHz. The QA401 uses the AK5397 and while its not as low distortion residual as the AK5394 its pretty good still and very cost effective.

Pavel is talking about is software distortion correction process that can correct the distortion of a DAC-ADC chain to a remarkable degree.

I still have not fully conquered REW. I'm still working through Virtins, which is also very capable and also supports NI DAQ cards with higher sample rates. Too many software options to choose among.
__________________
Demian Martin
Product Design Services
  Reply With Quote
Old 24th May 2020, 08:43 AM   #299
phofman is offline phofman  Czech Republic
diyAudio Member
 
Join Date: Apr 2005
Location: Pilsen
Quote:
Originally Posted by syn08 View Post
Otherwise, I forgot you are an RPi afficionado, I must admit I'd rather include in a design a CPLD/FPGA/STM32 chip then a generic RPi board.
I perfectly understand your position, a hard-coded MCU or FPGA is technically the best way for mere bridging to UAC2. My requirements are different - before the UAC2 transmission I need a rather complicated multithreaded double-precision DSP plus some means of controlling the process from the connected PC. Since I use linux for the processing, RPi4 is basically the only available option. The four 1.6GHz cores will likely be used to max, I would employ more if such a hardware with complete, continously updated, and long-term software support was available. The USB gadget code allows multidevice configuration - the plan is an UAC2 sound device and a USB network card which via autodiscovery and a web server running on the RPi would offer detailed control in web browser on the work PC. USB-audio features cannot provide the needed control plus I would have to create and maintain specific drivers for all major OSes (= nonsense).

Quote:
The only scenario I would use an RPi would be a stand alone instrument, with it's own display and entry devices, when the whole UAC2 **** is no longer required.
I think people prefer running the actual analysis on their existing PCs with working environments already customized to their individual needs. Moving the analysis to the device would go against this logical trend. Plus RPi4 does not have enough power to run the whole chain including high-resolution display reliably (i.e. with sufficient timing margins).
  Reply With Quote
Old 24th May 2020, 05:23 PM   #300
syn08 is offline syn08  Canada
diyAudio Member
 
syn08's Avatar
 
Join Date: Aug 2005
Location: Toronto
Quote:
Originally Posted by phofman View Post
I think people prefer running the actual analysis on their existing PCs with working environments already customized to their individual needs. Moving the analysis to the device would go against this logical trend. Plus RPi4 does not have enough power to run the whole chain including high-resolution display reliably (i.e. with sufficient timing margins).
That's an interesting conclusion which may, or may not be always true. Personal example: logic analyzer, I got one of those Kingst Chinese USB devices, running at 500MHz timing/100MHz state; comes with a decent PC software and works perfectly for this project. Then I realized I don't have room on my bench for a laptop, powered USB hub, etc... the laptop screen was too tiny to read all the details at high resolution (or I had to spend time and a mouse to scroll through the screen) so I ended up spending a lot of time and effort cruising between the bench and the computer screen 10ft. away. So then I realized an Agilent logic analyzer not only would provide 4GHz timing 600MHz state (not really required here, but nice to have, anyway), 68 channels (instead of 16 with the Kingst device) but it would also not take more space than the USB setup, plus less cable salad on the bench. So I got a second hand 16903A logic analyzer (which is just an old PC running Windows XP, plus the Agilent software) with a 16950A card, for less than the Kingst USB toy. Figure out...

The same (for me) with an audio analyzer, I find much more convenient to use an all-in-one Rohde UPD in the rack than a sound card and cruising to the computer screen to watch the results (plus the advantage of a calibrated instrument). One may argue I need a larger bench, which is indeed true .

Last edited by syn08; 24th May 2020 at 05:25 PM.
  Reply With Quote

Reply


Audio grade ADCs, PCM4222, AK5572, CS5381, etc...Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Current thinking :) on Audio-grade vs Standard grade Toroids in PSU!! tonewheelkev Power Supplies 40 21st January 2020 06:30 AM
AK5572, a new hope for diy Audio Analyzer? mtoc Equipment & Tools 0 5th July 2016 07:38 AM
FS: Digital Audio ICs - ADCs, DACs, SRCs, other stuff.. len_scanlan Swap Meet 0 29th December 2011 08:40 AM
Audio humor. Wattgate 381 Audio Grade Duplex Socket Pano Parts 18 17th March 2010 07:04 PM
Rebuilding a DAC Audio grade or Computer grade caps? gto127 Digital Source 0 28th January 2009 08:17 PM


New To Site? Need Help?

All times are GMT. The time now is 04:14 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2020 DragonByte Technologies Ltd.
Copyright ©1999-2020 diyAudio
Wiki