Can the MMA8452 I2C Interface run at 1.8V while VDDIO is tied to 2.5V?

[jweganalogic]

Hello,

We've recently been informed that the MMA8450 has been marked for EOL, with the recommended replacement being the MMA8452.  The main difference between the two parts is that the MMA8452 uses a core voltage of 2.5V, whereas the MMA8450 could use a core voltage of 1.8V.  In our particular application, the MMA8450 was chosen because it meant that it could operate at the same I/O voltage as the microprocessor with which it was interfacing.  Also in the current design, the interrupt signals interface with some level shifting circuitry that then wake up the system in the event of a drop.  There is a separate regulator on the boards that operates at 1.8V to keep the MMA8450 alive while the rest of the system is powered down.  If the I2C interface on the MMA8452 can run at 1.8V while the rest of the sensor subsystem ran at 2.5V, then we would not have to change our board design; we could simply replace the 1.8V fixed regulator with the 2.5V equivalent.  According to the datasheet, the input voltage thresholds fall within range for a 1.8V system, but there are some notes in the datasheet about supply sequencing that make me think that this might not operate correctly if the I2C lines are using a different voltage than the rest of the chip.

Any insight is much appreciated; we have to make a decision on whether or not to respin the board quite soon.

Thanks,

Jake

[reyes]

Hi Jake,

You can use the MMA8452 for your design, and as mentioned, you should only separate the VDD (2.5V) voltage from the VDDIO (1.8v) voltage, this way you can maintain the interface voltage with the mcu at the same 1.8V, and you will only need to create a new 2.5V voltage for the VDD analog supply.

You can find more details about how to make these connections and the differences between the MMA8450 with the MMA8452 in section 3 of the following application note: http://cache.freescale.com/files/sensors/doc/app_note/AN4077.pdf

Have a great day,
Jose Reyes

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------

[jweganalogic]

Hi Jose,

Thank you for the response, but that does not actually address my question.  On the current board fab, VDDIO and VDD both tie to what would be 2.5V if we were to replace the EOL'ed 8450 with the 8452.  The I2C interface connects to a 1.8V supply.  There is no way on the board fab to separate VDDIO and VDD.  The datasheet would seem to indicate that one could operate the I2C interface at 1.8V even though VDDIO is 2.5V (Vih = 0.7VDDIO, which, at 2.5V, is equal to 1.75).  Is this true, would this work?

Thanks,

Jake

[reyes]

Hi Jake,

Our recommendation is to have the same voltage at the mcu communication interface and at the VDDIO of the sensor or to add an I2C level shifter to avoid communication issues, however, based in that note at the datasheet and the configuration (Vih = 0.7VDDIO, which, at 2.5V, is equal to 1.75), we can say that it should (and will probably) work fine, but it also important to mention that it is not is not reliable or recommended by Freescale.

Have a great day,
Jose Reyes

-----------------------------------------------------------------------------------------------------------------------
Note: If this post answers your question, please click the Correct Answer button. Thank you!
-----------------------------------------------------------------------------------------------------------------------