I2S multiplexer / switch
Part Number: TS3A44159
I have read a couple of posts that indicate the TS3A44159 switch would be suitable as an I2S MUX. I used the part in my PCB and it turns out the BCLK input switch is crosstalking. Obviously the BCLK is critical for timing in I2S when no MCLK is used.
I am using a 3.3V signal with single switch of the chip and BCLK1 being NC and BCLK2 NO,w/ output to a audio amp.
BCLK1 and BCLK2 are at different frequencies (different sources)
When in NC, (IN1-2 = low) there are no issues. Audio from BCLK1 is fine
When in NO, (IN1-2 =3.3V) Audio from BCLK2 is being effected by BCLK1 (digital noise).
I verified this by cutting the trace of BCLK1 and noise was eliminated
Any suggestions on how to eliminate the crosstalk?
As an alternative, if I use two of the switches instead of 1, would the crosstalk remain?
Switch 1 NC = BCLK1
Switch 1 NO -= OPEN
Switch 2 NO = BCLK2
Switch 2 NC = OPEN
So when IN1-2 is toggled, the source changes.
My concern is that if I leave one leg of the switch OPEN, it will create problems since your datasheet recommends 50ohm to GND. Obviously if I tie the open to GND it will kill the signal.
What do you recommend in this situation?
thank you fr reaching out to us. Which frequencies are the I2S BCLK on? At around 48kHz there should be very low crosstalk (-70dB).
Have you measured the crosstalk according to Figure 24 in the datasheet?
If two devices would be used, the crosstalk would be an non-issue, but I think there could be another issue.
Let us know how much crosstalk you see in your implementation.
Thank you and best regards,
We are glad that we were able to resolve this issue, and will now proceed to close this thread.
If you have further questions related to this thread, you may click "Ask a related question" below. The newly created question will be automatically linked to this question.
In reply to Ambroise Suter:
Well the BCLK is a result of the following..
44.1 kHz × 16 × 2 = 1.4112 MHz and can be as high as 48 kHz × 32 × 2 = 3.072 MHz
Since I don't have a test setup (breakout board), doing this experiment on my finished PCB would be near impossible (6mil traces..), so I can't really answer that part of the response.
In terms of using two devices, that's not what I was suggesting. I was suggesting using 2 of the 4 switches within a single TS3A44159 device. The idea being, instead of using the shared transistors within a single switch, using switch 1 for BCLK1 and another for BCLK2, as you can see in the picture below. I'm wondering if the crosstalk is a result of the shared transistors and therefore not using the same transistor path would help with the issue?
In reply to John Heaton:
with 3Mhz, the crosstalk isolation is at -45dB. With a signal of 3.3V the crosstalk signal should be below 0.02V. That's why I think there could be another source.
Your suggested solution should improve it anyway though, maybe even use NC1 and NO2 to maximize the distance of the BCLK traces on the board.
All content and materials on this site are provided "as is". TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with regard to these materials, including but not limited to all implied warranties and conditions of merchantability, fitness for a particular purpose, title and non-infringement of any third party intellectual property right. No license, either express or implied, by estoppel or otherwise, is granted by TI. Use of the information on this site may require a license from a third party, or a license from TI.
TI is a global semiconductor design and manufacturing company. Innovate with 100,000+ analog ICs andembedded processors, along with software, tools and the industry’s largest sales/support staff.