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.

ISO6763-Q1: Difference between ISO7763 and 6763

Part Number: ISO6763-Q1
Other Parts Discussed in Thread: ISO7763-Q1, ISO6763, ISO1640, ISO7763

Tool/software:

Hello Team,

We are using ISO7763-Q1 isolator in our project, VCC1 and VCC2 is connected to 5V regulator. 

INB is connected to a clock signal from Micro with clock frequency max 1kHz through 2K ohm resistor.

with this no issues. 

Now, for the same application if we use ISO6763 instead of 7763, we measured that the clock from micro is pulled HIGH for 180-190 ms constantly, aft that we can see oscillations on clock (refer image attached)

Note; No other components or test conditions varied except change of isolator. In image attached , Pink is CLK and Yellow is reference point.

Why ??

  • Hello,

    Thank you for your inquiry. Please give me an additional day to respond.

    Regards,
    Aaditya Vittal

  • The ISO676x is an older device an has a longer power-up delay.

  • Hello,

    Thank you for your inquiry. Please give me time until Monday to respond.

    Regards,
    Aaditya Vittal

  • How much is the delay time ? could you please elaborate the detailed difference on power up delay between these 2 isolators with data and with regards to 25 deg to 70 deg junction temperature ?

  • Yes, please. No issues Mr. Vittal.

  • Hi Kavin, 

    Thanks for your patience.

    Please allow me some additional time to inquire some experts in my team.

    In the meantime, can you please explain a bit more regarding the waveform you have provided? and labelling of the waveform would also be helpful.

    Regards,
    Aaditya Vittal

  • Hi Aaditya,

    I am not allowed to disclose more info. Product is an Automotive. Isolator is an interface between FS32K MCU and ADS1015AQDGSRQ1 (TI ADC with I2C)

    One mistake in my info before, I2C clock frequency from MCU is at 50 kHz.

    CLK connected to INB, 2K ohm pull up resistor is connected between VCC1 and INB. INC is bidirectional data from I2C to MCU  and Viceversa. 1.5K ohm connected from INC to VCC1.

    VCC1 and VCC2 is at constant 5 V from internal regulator.

    Graph:

    1. Yellow (VSUP_EN) comes from SBC of product.

    2. 2 ms after VSUP_EN turned on VCC1 and VCC2 is constantly supplied with 5 V from 2 different sources (hence use this as a reference point)

    3. CLK from MCU started Microsecs after VSUP_EN is turned on.

    4. Pink(CLK) is the clock measured closest to INB

    Using 7763 ISO--> CLK starts 2.87ms after VSUP_EN (expected and working okay)

    Using 6763 ISO--> CLK starts 196.3ms after VSUP_EN (NOK)

    Nothing else changed in product except this ISO, old ISO i can expect a longer propagation time, clock stretching, longer power up time. but 190ms is too much and I don't understand why. First suspect was CLK from MCU (SW) itself, but we tested in a same part with same SW , same environmental condition, expect change in ISO.

  • Hi Kavin, 

    Thank you for the additional clarification. Please allow me to review this and I will respond shortly.

    Regards,
    Aaditya Vittal

  • Hello Aaditya,

    This one is kind of high prio from our side, Please consider and possible a quick feedback is much appreciated.

    If you need any further info, please let me know.  

    If any meeting required, let me know, we can connect remotely, with our Hardware experts (I am not one of them).

    Thank you 

  • Hello,

    Thank you for your patience.

    Isolator is an interface between FS32K MCU and ADS1015AQDGSRQ1 (TI ADC with I2C)

    It looks like the digital isolator that is being used is being used as a bidirectional buffer for I2C communication. Also, it seems the ADC device is an open-collector device while the digital isolator is a CMOS device.

    Considering this, I do not think the "power up delay" or other timing related specs are of concern, but rather a question of compatibility. The digital isolator does not support bidirectional communicated which I2C requires. Another concern is that the open collector ADC may also have combability issues with the CMOS digital isolator. 

    With that said, I recommend the ISO1640, a bidirectional I2C isolator, which may be more appropriate with your application.

    Regards,
    Aaditya Vittal

  • Hi Aaditya,

    Regarding compatibility, what is the impact? and why 7763 works without any issues.

    Data line is Bidirectional only.

  • Hello Kavin,

    ISO7763 may work for certain unidirectional transmissions, however, each channel is uni-directional. Data will not transmit in the reverse direction, and this could be the cause of the issue. 

    ISO1640 can handle bidirectional signals like I2C and would be the recommended solution for isolated bidirectional communication.

    Best,
    Andrew

  • Sorry for wrong info, its unidirectional,  refer the schematics.

  • Hi Kavin, 

    Thanks for the schematic.

    Please see my markup below:

    As you can see in the boxed section, the INC and OUTD pins are connected together. Even if a diode is used, this does not mean that the digital isolator can be used for bidirectional communication. As Andrew highlighted in the previous post, the ISO6763 and ISO7763 both only support unidirectional channel. These channels cannot be tied together directly to support bidirectionality.

    There are several other components and considerations to allow for bidirectional communication, which is explained in more detail here: Designing an Isolated I2C Bus interface by using digital isolators (Rev. A) (ti.com)

    This E2E FAQ also highlights some more details as well: ISO7742: Could ISO7742 be used in I2C isolation? - Isolation forum - Isolation - TI E2E support forums

    As we mentioned before, we recommend the ISO1640, a bidirectional I2C isolator, which may be more appropriate with your application.

    Regards,
    Aaditya Vittal