Hmm...
BTW: I also noted Tenor (made by GFEC), which is used in quite a few commercial/audiophile devices. Some of the chip models are TE8802 and TE7022
Really? Last i checked, XMOS was the most expensive (????)Hi,
I think that now there is no perfect chip, each has some drawbacks.
XMOS:
Pros:
- low price
Hmmm ... didn't know about this one. Seems interesting, and the C-MEDIA website is much more informative than many others.C-MEDIA CM6631A
This one is priced high? Really? It's found in some of the lowest-priced gear: Stoner Acoustics UD120 ($60-70) and one of the popular hifimediy Sabre DACs. I also saw a $40.00 SaviAudio SA9227 USB Audio 32Bit 384KHz USB to I2S Converter Daughter Board from yuan-jing.com.SaviAudio SA9227
Cons:
- price
BTW: I also noted Tenor (made by GFEC), which is used in quite a few commercial/audiophile devices. Some of the chip models are TE8802 and TE7022
Last edited:
Hi,
I wrote about same chip prices.
XMOS XS1-U8A (integrated USB PHY ) about 11$
XMOS XS1-L8A (without USB PHY) about 8$
XHRA-2HPA about 8$
They are more expensive than expected.
I suggested advertising the XMOS web:
"$2 Chips Deliver High Resolution Audio to consumer applications"
CM6631A cost about 11$
SA9227 cost about 20$.
TAS1020B costs abut 6$.
Prices are for small volume (DIY).
I think that the prices for Chinese manufactures are completely different (high volume and Asia market).
TE8802 has status End of Live (there was some problems with drivers)
TE7022 it is only USB Audio Class 1.0
Manuactures like C-MEDIA, SaviAudio GFEC are not very friendly to DIY.
Regards
JarekC
I wrote about same chip prices.
XMOS XS1-U8A (integrated USB PHY ) about 11$
XMOS XS1-L8A (without USB PHY) about 8$
XHRA-2HPA about 8$
They are more expensive than expected.
I suggested advertising the XMOS web:
"$2 Chips Deliver High Resolution Audio to consumer applications"
CM6631A cost about 11$
SA9227 cost about 20$.
TAS1020B costs abut 6$.
Prices are for small volume (DIY).
I think that the prices for Chinese manufactures are completely different (high volume and Asia market).
TE8802 has status End of Live (there was some problems with drivers)
TE7022 it is only USB Audio Class 1.0
Manuactures like C-MEDIA, SaviAudio GFEC are not very friendly to DIY.
Regards
JarekC
Compatibility with legacy DACs and DFs
I prefer digital filtering (oversampling) using SAA7220B or PMD100.
I assume that your orig. USB converter also is okay feeding these legacy devices?
What about more-modern USB decoders (XMOS, SAVITECH, C-Media) and compatibility with legacy DACs and DFs? Is it "simply" a matter of the USB decoder spitting out data at or below a certain maximum speed? Even if a modern decoder is too fast, I would think a buffer would allow compatibility???
So I assume your original project -- as outlined in the first post of this thread -- has been tested in the legacy DACs (like 1541, etc.)?I wrote the firmware myself.
I sterted this thread 4 yaers ago, since then a lot has changed
TAS1020B is still in production but it is not recommended for new design.
Ti don't have replacment for this chip.
If you don't need more than 96 kHz (for example old chips like TD1541A), it is a good chip.
I prefer digital filtering (oversampling) using SAA7220B or PMD100.
I assume that your orig. USB converter also is okay feeding these legacy devices?
What about more-modern USB decoders (XMOS, SAVITECH, C-Media) and compatibility with legacy DACs and DFs? Is it "simply" a matter of the USB decoder spitting out data at or below a certain maximum speed? Even if a modern decoder is too fast, I would think a buffer would allow compatibility???
For USB to I2S converter make special 16 bit version 16bit frame-16bit data for old DACs
Some people from this forum use this version, for example:
http://www.diyaudio.com/forums/digi...pdif-converter-24bit-96khz-9.html#post3285677
If you use I2S bus there should not be compatibility problems with new decoders.
New decoders use 32 bit frames.
Regards,
JarekC
Some people from this forum use this version, for example:
http://www.diyaudio.com/forums/digi...pdif-converter-24bit-96khz-9.html#post3285677
If you use I2S bus there should not be compatibility problems with new decoders.
New decoders use 32 bit frames.
Regards,
JarekC
BCK (for I2S)
The TDA1541 can accept only:
BCK=6Mhz (max)
For classic Philips CDPs, the XO is 11.2...MHz and the SAA7220 digital filter divides this by two internally (so BCK of ~6MHz is sent to the TDA1541).
(Note: whether or not your converter is still for sale is unimportant)
Will the version you offered work with TDA1541.For USB to I2S converter make special 16 bit version 16bit frame-16bit data for old DACs
Some people from this forum use this version, for example:
The TDA1541 can accept only:
BCK=6Mhz (max)
For classic Philips CDPs, the XO is 11.2...MHz and the SAA7220 digital filter divides this by two internally (so BCK of ~6MHz is sent to the TDA1541).
(Note: whether or not your converter is still for sale is unimportant)
TAS1020B
The TAS1020B also has the XTALI (pin 47) "Crystal input: Input to the on-chip oscillator from an external 6-MHz crystal".
So, JarekC, I'm not sure how you're using MCLKOx, if you're using it at all??? This may be how a CDP or DAC's XO is configured to the TAS1020B?
I just noticed MCLKOI, MCLKO1 and MCLKO2 pins on the TAS1020B datasheet; these are described as: "Master clock input. An input that can be used as the master clock for the codec port interface..."Will the version you offered work with TDA1541.
The TDA1541 can accept only:
BCK=6Mhz (max)
For classic Philips CDPs, the XO is 11.2...MHz and the SAA7220 digital filter divides this by two internally (so BCK of ~6MHz is sent to the TDA1541).
(Note: whether or not your converter is still for sale is unimportant)
The TAS1020B also has the XTALI (pin 47) "Crystal input: Input to the on-chip oscillator from an external 6-MHz crystal".
So, JarekC, I'm not sure how you're using MCLKOx, if you're using it at all??? This may be how a CDP or DAC's XO is configured to the TAS1020B?
Last edited:
Converter: I2S 16bit frame MCLK 256 Fs
TDA1541A BCK max = 6MHz, SCK max = 12MHz
44.1kHz BCK = 2*16*44100 = 1.411200 MHz (< 6MHz) SCK = 11,2896 MHZ (<12MHZ) OK
48,0kHz BCK = 2*16*48000 = 1.536000 MHz (< 6MHz) SCK = 12,2880 MHZ (<12MHZ) OK
88.2kHz BCK = 2*16*88200 = 2.822400 MHz (<6 MHz) SCK = 22,5792 MHZ (>12MHZ) Outside TDA specification
96,0kHz BCK = 2*16*96000 = 3.072000 MHz (<6 MHz) SCK = 24,5760 MHZ (>12MHZ) Outside TDA specification
It is possible to reduce MCLK to 128Fs and then all configuration will be ind TDA specification.
JarekC
TDA1541A BCK max = 6MHz, SCK max = 12MHz
44.1kHz BCK = 2*16*44100 = 1.411200 MHz (< 6MHz) SCK = 11,2896 MHZ (<12MHZ) OK
48,0kHz BCK = 2*16*48000 = 1.536000 MHz (< 6MHz) SCK = 12,2880 MHZ (<12MHZ) OK
88.2kHz BCK = 2*16*88200 = 2.822400 MHz (<6 MHz) SCK = 22,5792 MHZ (>12MHZ) Outside TDA specification
96,0kHz BCK = 2*16*96000 = 3.072000 MHz (<6 MHz) SCK = 24,5760 MHZ (>12MHZ) Outside TDA specification
It is possible to reduce MCLK to 128Fs and then all configuration will be ind TDA specification.
JarekC
Does your converter offer 11.2MHz XO?
In your converter with two XOs, why not jut make one of them 11.2...MHz (is this what you offer?).
For TDA1541, you still need to /2, but that is done automatically in the SAA7220 or a simple /2 circuit.
Do you mean in the firmware or using a separate clock (e.g., 11.2...MHz XO crystal commonly found)?It is possible to reduce MCLK to 128Fs and then all configuration will be ind TDA specification.
In your converter with two XOs, why not jut make one of them 11.2...MHz (is this what you offer?).

For TDA1541, you still need to /2, but that is done automatically in the SAA7220 or a simple /2 circuit.
just to clarify ...
Just to clarify my last reply ...
... if someone had one of your converters (as pictured below), soldered out the 22.xx MHz XO, and replaced it with 11.2xxx MHz XO, would it work (or would you have to also adjust in firmware)?
Just to clarify my last reply ...
... if someone had one of your converters (as pictured below), soldered out the 22.xx MHz XO, and replaced it with 11.2xxx MHz XO, would it work (or would you have to also adjust in firmware)?
Do you mean in the firmware or using a separate clock (e.g., 11.2...MHz XO crystal commonly found)?
In your converter with two XOs, why not jut make one of them 11.2...MHz (is this what you offer?).
![]()
For TDA1541, you still need to /2, but that is done automatically in the SAA7220 or a simple /2 circuit.
TAS1020B has internal programable MCLK clock divider so output MCLK frequency is changed by firmware.
First XO 22.579200 MHZ is for 44.1kHz and 88.2Khz sampling frequency
Second XO 22.5760 MHz is for 48kHz and 96Khz sampling frequency
If you will use only 44.1 (88.2kHz) then you need one XO.
After changing firmware to 128Fs, you don't need to divide MCLK.
Best Regards
JarekC
First XO 22.579200 MHZ is for 44.1kHz and 88.2Khz sampling frequency
Second XO 22.5760 MHz is for 48kHz and 96Khz sampling frequency
If you will use only 44.1 (88.2kHz) then you need one XO.
After changing firmware to 128Fs, you don't need to divide MCLK.
Best Regards
JarekC
GUI
Does your converter come with a GUI "app" that allows the user to control/tweak certain parameters?After changing firmware to 128Fs, you don't need to divide MCLK.
Suggestions...
-Most of the latest Chinese USB-to-I2S converters have some GUI, so that's good to have because DIYers love to tweak for some reason! If the TAS1020B is keeping you from doing this, dump it (use Tenor, Savitech, etc.)
-Keep clocks (XO) as clean as possible -- have separate LDO, use ferrite beads, etc.
- Incorporate a decent DAC (ESS, Wolfson, Cirrus, TI, AD, AKM)
- Offer at low price (it's really hard to beat Chinese eBay/Alibaba deals, so watch out).
Not sure your converter is still avail for sale, but some suggestions:No.
Firmware must be recompiled and programed to EEPROM.
-Most of the latest Chinese USB-to-I2S converters have some GUI, so that's good to have because DIYers love to tweak for some reason! If the TAS1020B is keeping you from doing this, dump it (use Tenor, Savitech, etc.)
-Keep clocks (XO) as clean as possible -- have separate LDO, use ferrite beads, etc.
- Incorporate a decent DAC (ESS, Wolfson, Cirrus, TI, AD, AKM)
- Offer at low price (it's really hard to beat Chinese eBay/Alibaba deals, so watch out).
Not sure your converter is still avail for sale, but some suggestions:
-Most of the latest Chinese USB-to-I2S converters have some GUI, so that's good to have because DIYers love to tweak for some reason! If the TAS1020B is keeping you from doing this, dump it (use Tenor, Savitech, etc.)
-Keep clocks (XO) as clean as possible -- have separate LDO, use ferrite beads, etc.
- Incorporate a decent DAC (ESS, Wolfson, Cirrus, TI, AD, AKM)
- Offer at low price (it's really hard to beat Chinese eBay/Alibaba deals, so watch out).
The converter has been designed in accordance with good engineering practice.
There are 4 low noise LDO, all high speed lines are teminated.
I'm DIYer 😀
I designed this converter to my satisfaction and I do not have to compete on price with the Chinese
Regards,
JarekC
Hi,
I don't have any spare boards. TAS1020B chip is now obsolate and difficult to buy.
I designed new converters besed on Microchip PIC32MX and PIC32MZ, but PCB design is in prototype stage.
My newest firmware (PIC32MX) is used 😀 by Feniks Audio in Essence Speakers. Feniks Audio
Regards,
JarekC
I don't have any spare boards. TAS1020B chip is now obsolate and difficult to buy.
I designed new converters besed on Microchip PIC32MX and PIC32MZ, but PCB design is in prototype stage.
My newest firmware (PIC32MX) is used 😀 by Feniks Audio in Essence Speakers. Feniks Audio
Regards,
JarekC
- Status
- Not open for further replies.
- Home
- Source & Line
- Digital Source
- TAS1020B-based Asynchronous USB to I2S/SPDIF converter 24bit/96kHz