• State
  • Locked Locked
  • Replies 17 replies
  • Subscribers 32 subscribers
  • Views 2402 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TM4C1294NCPDTI3: Impedance of IO pin while MCU is OFF

ikon
Genius 5355 points

Hi Support,

What will be the impedance of IO pin while the MCU is OFF condition.

Facing some weird behavior in one of the IO pin.

PQ4 was used as output pin and connected to a voltage divider resistor network .

This resistor some time may have a voltage of 3.00V (from charger input voltage) even when the MCU is OFF.

The issue is  voltage across the resistor is 0.6V where as expected voltage is 3.00V( Input voltage was 33V and resistor divider circuit made using 100K and 10K combination) .

Expected Voltage @ 10K resistor is 3V but getting less voltage because that resistor connected to MCU Pin PQ4. 

When 33V applied there Is no 3.3V so MCU is not up.

Thanks.

  • 0
    Guru 117855 points

    It appears that you seek to verify the presence of that (voltage divided) 33V. While this MCU (may) survive the effects of voltage introduction upon GPIO pins when the MCU is "not" under bias - I don't believe the impedance of such GPIO is listed - or consistent.

    Classically it has been bad form to allow signals/voltages to arrive when the MCU is "off." The typical work-around is to employ a, "blocking" signal FET which only allows signal introduction when the MCU is "on." (A GPIO is sacrificed to cause the external FET to enable signal passage to the MCU.)

    Final sentence "When 33V applied - there Is no 3.3V so MCU is not up" confounds! This suggests that there (may) be no need for this (voltage scaled) signal to be applied to MCU's PQ4.  (this due to the fact that 33V is only applied when the MCU is unpowered - thus the MCU is (powerless) to react to this logic level - is it not?)    The "justification" for this voltage input into PQ4 has not been made - as best I can tell...

  • 0
    TI__Guru**** 244400 points
    Hello Shaun,

    As you can see in the data sheet, TM4C129x IO's are not full fail safe. So when a voltage is applied on the IO without having supplied on the VDD and VDDA power rail, the device would parastically power up by sourcing current from the IO. Since the current would be sourced from the resistor divider network, the current draw would cause a voltage drop across the resistor and this would in turn result in lower voltage on the IO. But since the current would not be sufficient to power the device, the IO state would continue.

    Now the second part is what is 33V? How are you converting it to 3.3V for VDD and VDDA power rails of the MCU

    Regards
    Amit
  • 0 in reply to cb1_mobile
    Guru 27665 points

    cb1_mobile said:
    (A GPIO is sacrificed to cause the external FET to enable signal passage to the MCU.)

    Or possibly the supervisory IC, but you need to be carful about ensuring the power off behaviour.

    Robert

  • 0 in reply to Robert Adsett
    TI__Guru**** 244400 points
    Hello Robert

    The poster has not clarified as to how the VDD and VDDA rails are being powered?

    Regards
    Amit
  • 0 in reply to Amit Ashara
    Guru 27665 points
    Other than they are apparently not powered in this situation. So as long as the supervisor is monitoring the power chain for the micro, as it must to be effective, it should be available to consider.

    Robert
  • 0 in reply to Robert Adsett
    Guru 117855 points

    May I ask you two - what is the point of routing this voltage divider (33V into 10:1 divider) into the MCU - when the MCU is stated to be OFF when the 33V is supplied?

    Post is undisciplined - although avoids, "Does not Work" - yet replaces that w/No description whatsoever for the introduction of the divided 33V to the MCU.

  • 0 in reply to cb1_mobile
    Genius 5355 points

    Hi Support,

    Thanks for detailed response. 

    Isolate the IO line is best option I agree, one more thing is same system with M3 MCU working fine no issues but in M4 we are seeing problem.

    Attached the image for reference.

    Net nCHRSW_CNT connected to MCU PQ4, Charger+ is 33V power input.

     

    Thanks.

    TM4C1294NCPDTI3.docx

  • 0 in reply to ikon
    TI__Guru**** 244400 points
    Hello Shaun

    And what is the supply to TM4C?

    Regards
    Amit
  • 0 in reply to Amit Ashara
    Genius 5355 points

    Hi Amit,

    Supply voltage to TM4C is 3.3V.

    Thanks.

  • 0 in reply to ikon
    TI__Guru**** 244400 points
    Hello Shaun,

    Thank you. And how are you generating the 3.3V to the MCU?

    Regards
    Amit
  • 0 in reply to Amit Ashara
    Genius 5355 points

    Hi Amit,

    Highlighted in Red previously.

    From 24V battery/33V charger generating 8.25V then using LDO generate 5V and from 5V generate 3.3V.

     

    Thanks.

  • 0 in reply to ikon
    TI__Guru**** 244400 points
    Hello Shaun

    You meant the attachment!
    As I mentioned, the current drain into the TM4C129X which causes the voltage drop on the 10K would be the cause for the supply not to start up. If you can disconnect PQ4 from the resistor divider network, then does the supply come up?

    Regards
    Amit
  • 0 in reply to Amit Ashara
    Genius 5355 points

    Hi Amit,

    Yes, once PQ4 isolated from below mentioned connection then board is booting up (Power up all voltages available).

    We did following things.

    1. Change the resistor divider combination to 22K and 2.2K then it worked even the PQ4 was in loop. Same scenario not working for 33K and 3.3K combination.

    2. Introduce and NPN transistor between PQ4 and resistor divider combination.

    Please suggest.

     

    Thanks.

  • 0 in reply to ikon
    TI__Guru**** 244400 points
    Hello Shaun

    I believe is the current being sourced via the pin which is causing the voltage drop on the 3.3K. With the 2.2K resistor most of the current makes it way to the FET. Did you check the voltage level on the 2.2K resistor in the working case?

    Regards
    Amit
  • 0 in reply to Amit Ashara
    Genius 5355 points

    Hi Amit,

    Yes , voltage across 2.2K resistor is around 3V. But voltage require to make the Q1 to ON is 1.5V max.

    Thanks.

  • 0 in reply to ikon
    TI__Guru**** 244400 points
    Hello Shaun,

    What is the function of PQ4? As you mentioned earlier in #2, there must be some form of isolation circuit in the path if the pin is only to be an output for control.

    Regards,
    Amit
  • 0 in reply to Amit Ashara
    Guru 117855 points
    Post (not poster) reads like, "scratching chalk on a chalk-board." Endless cross examination to "tease out" needed facts.