This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TCA9517: TCA9517 B side connect to TCA9546A

Part Number: TCA9517
Other Parts Discussed in Thread: TCA9546A, , TCA9617

Hi Sir,

We just found if we connect TCA9517 B side to TCA9546A, the I2C can't work, as below topology.

But if we swap TCA9517 A-side to connect with TCA9546A, the I2C work well.

Could you please help to clarify the reason?

Thanks

  • TCA9517 has a higher VOL on the B-side, which may be a factor here.  By the way, you can read more about this "static voltage offset" in this app note:

    https://www.ti.com/lit/an/scpa054/scpa054.pdf

    In order to debug this specific case, though, I think it would be useful to take a look at the I2C signals (SDA/SCL) on both sides of the buffer on an oscilloscope for the working and non-working cases.  That way we could see specifically what may be causing the communication issue.

    Regards,
    Max

  • Hi Max,

    We have measured the TCA9517 B side( between the TCA9546A I2C MUX and TCA9517B side).

    It seems TCA9546 or PCA9546 I2C MUX port have SVO(Static voltage offset) about 0.4V~0.5V, and our TCA9517 B side(have SVO) connect to it I2C MUX port causing two SVO connection issue?

    Am I comprehend it correct?

    For I2C MUX output port, we measure the waveform and shown that have offset about ~0.3V, which mean that there are different voltage offset level for I2C repeater(~0.5V) and I2C mux out port(~0.3V).

  • Hi Anne,

    Actually, the mux will not add an SVO.  I just acts as a low-impedance switch that would activate when the TCA9617 (or device on the other side of the switch) pulls low.  I was just thinking that since the main difference between the A and B side of the TCA9617 is the higher VOL due to the SVO, that maybe the low-level voltage seen by the device on the other side of the switch might be too high (i.e., higher than its "VIL" specification).  If that were occurring then one solution may be to lower the low-level voltage by using weaker (higher resistance) pull-up resistances.  Please let me know if that doesn't make sense.

    Thanks for the waveforms.  I'm trying to understand them.  Is the first one showing the I2C lines at the "end device" / TCA9617 "A" side?  And, is the end device an I2C slave?  Then, the bottom two waveforms (in green and magenta) are on the TCA9617 "B" side, which is connected through the mux to a master?  In this case, is communication failing?

    Regards,
    Max

  • Hi Max,

    Thank you for your reply! May I ask for your help to double check whether the following application suitable or not?

    From the document you provide , TCA9517 seems have no VCC rules applied on it. It seems I can connect VCCA to 3.3V_AUX and VCCB to 3.3V.

    We use I2C repeater to separate power rails 3.3V and 3.3V_AUX as below diagram.

    Please provide comment and suggestion!

    Thank you

  • Hey Anne,

    What you have shown "can" work but you need to keep in mind that the B side of TCA9517 is seen through the 9546s you have so the static voltage offset is still seen past the switches/muxes when the channel is enabled. This means the TCA9517 should not be connected to another static voltage offset buffer like another TCA9517 through an I2C switch/MUX.

    Do you have a more detailed I2C block diagram you may be able to show us offline? duynguyen@ti.com

    Thanks,

    -Bobby