Just came across this relatively new part and haven't found any discussion, finally a low cost solution for USB isolation with high speed devices, basically all modern DACs at this point.
Should allow an external isolator to be easily added to any USB DAC, might not hurt to have this extra layer of isolation in DACs with internal post-USB isolation either.
Should allow an external isolator to be easily added to any USB DAC, might not hurt to have this extra layer of isolation in DACs with internal post-USB isolation either.
This is not new, I made a board with a chip in November 2021 - just when the samples (named XISOUSB211DPR) became available.
Works fine!
Unfortunately, this chip is still in "Preview" status, not mass production.
So, the only way to get it - to order or to buy the samples directly from TI.
But, even there - Out Of Stock now 🙁
Works fine!
Unfortunately, this chip is still in "Preview" status, not mass production.
So, the only way to get it - to order or to buy the samples directly from TI.
But, even there - Out Of Stock now 🙁
Do you have a USB DAC? Any changes in sound with the ISOUSB211 High/Full/Low Speed Isolated USB Repeater?
My USB DACs (all are XMOS) are always have galvanic isolation on I2S bus between the USB part and DAC itself, so additional isolation is not required.
But I used ISOUSB211 with non isolated ADC (Cosmos ADC) to avoid the ground loop. The results were very good.
Of course I've checked few DACS with ISOUSD211, "double isolation" gave nothing, it was just to check functionality.
The board was equipped also with DC-DC to provide 5V to the ADC or DAC: https://www.altor.co/products/test-equipment/usb-isolator
But I used ISOUSB211 with non isolated ADC (Cosmos ADC) to avoid the ground loop. The results were very good.
Of course I've checked few DACS with ISOUSD211, "double isolation" gave nothing, it was just to check functionality.
The board was equipped also with DC-DC to provide 5V to the ADC or DAC: https://www.altor.co/products/test-equipment/usb-isolator
Last edited:
before isolator maybe the best option is to put some USB redriver.
TI also has it. I cant remember a part marking...
.
I used to try silanna chip for the usb isolator and i can say that sound wasclearly better wih. More "analog", less grainy, more fluid and with stronger body. As I remember well 🙁
.
And that is the only isolator that can work with contemporary USB intefaces like amanero etc.
I rtied almost every other AD, LT - not working. Probably about speed?
.
I am glad to gear that TI has new chip and working. Because Silanna does not produce these isolators any more...
Thanks
Cheers
.
BTW Wurth has very good collection of app notes as parts as well for managing USB input. ESD, etc.
that goes in the package with USB input.
.
Take e look at the USB fround shield issues too - google
.
And for the cable i can recommend Phoenix Contct USB cable. Not too expensive (much much cheeper than audio spec.) but better.
TI also has it. I cant remember a part marking...
.
I used to try silanna chip for the usb isolator and i can say that sound wasclearly better wih. More "analog", less grainy, more fluid and with stronger body. As I remember well 🙁
.
And that is the only isolator that can work with contemporary USB intefaces like amanero etc.
I rtied almost every other AD, LT - not working. Probably about speed?
.
I am glad to gear that TI has new chip and working. Because Silanna does not produce these isolators any more...
Thanks
Cheers
.
BTW Wurth has very good collection of app notes as parts as well for managing USB input. ESD, etc.
that goes in the package with USB input.
.
Take e look at the USB fround shield issues too - google
.
And for the cable i can recommend Phoenix Contct USB cable. Not too expensive (much much cheeper than audio spec.) but better.
Maybe.
As I remember, AD/LT does not have High Speed isolators, only Full Speed, so they are not suitable for UAC2 operation (XMOS, Amanero, CM6662, Comvia, SaviTech).
I had this isolator: https://hifimediy.com/product/hifime-high-speed-usb-isolator/
If I remember correctly - it was on Silianna chip. Works, but was not so good and sometimes had a problem with USB Audio devices.
I used it mostly with high speed USB2COM (FTDA based) and JTAG interfaces.
I never had any problem with silanna (Silanna boards DC-DC converter, and no with my own with same chip but separate PS)
.
Never mind now we have TI new chips...
Take a look at the TI usb redriver ICs. They are so small to be packed into little pcb inside the USB connector at the computer usb out?
.
Never mind now we have TI new chips...
For that purpose industial ing-s using these Phoenix contact usb cables and propper ground shields. (Because we can just peplace interface for 60-70eu they loosing much more money when machinery not working 🙂 )
Take a look at the TI usb redriver ICs. They are so small to be packed into little pcb inside the USB connector at the computer usb out?
It was for the test purposes in the lab, not for production.
Ohhh, thanks!
Will try.
DK promises 3266 at December and 3165 at February. Price is good!
Last edited:
I use the iso after usb/digital audio interface too - but sound was clearly better with addition of usb isolator.
Yes indeed
but i think that worths
Thanks 🙂
it seems they have "recklocking" feature. Eye patterns in time looks real good. Price for the complet evaluation board is close to the IT chip if i dont get it wrong?
To be honest, I don't really understand why they are needed here, except that someone needs a 10-meter USB cable.
I may be a little out of the topic, but I tried a redriver chip from TI. Of the two ICs shown in the photo, the large is the USB hub chip for the retimer function, and a very small IC on top of it is the tusb217 redriver chip. (Has function of signal equalization and edge, DC boost)
The redriver has improved sound quality depending on the setting of the signal conditioning. The setting value varies depending on the nature of the cable, but it had a good effect with medium-intensity signal conditioning for cables over 1 meter.
.
I made it assuming that the redriver can improve the bit error because the music USB signal is an Isochronous method that does not fix the error.
That's right, but the frequency of these errors are very small. I made some test few years ago - a digital loop was implemented in XMOS USB-interface and after the "send' and "receive' data were compared. There were only few errors per hour(!), and sometimes there were no errors at all for 3-4 hours.
While this error is possible in theory, it is negligible in practice (using good cable and other components).
Good evidence of this error absence is DSD in DoP - with error DoP markers will be corrupted, which is vert listenable.
"Retimer" function, with async. sync in isochronic transfer should have zero effect.
I have tested asynchronous UAC2 with I2S (i.e. digital) loopback at 768kHz/32bit without any errors (bit perfect to 32 bit) in the measurement period of 15 minutes. This was without USB isolator and using USB power. Even if there were a random error now and then I doubt a USB redriver would be of help. USB isolator may be useful if there are severe noise or grounding issues with the USB connection.
Take a look for eye pattern measurements in time domain in the datasheet. For TUSB211 2m cable used
Attachments
Last edited:
I've been told by other engineers that there is actually quite a bit of data loss on USB cables. I really want to see real test results I can see it with my eye.
And from TI’s datasheet of tusb217 , when signal has ISI jitter , signal integrity can’t be guaranteed.
And from TI’s datasheet of tusb217 , when signal has ISI jitter , signal integrity can’t be guaranteed.