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.

TUSB4020BPHPEVM: Eval board fails automated test for DCP mode

Part Number: TUSB4020BPHPEVM
Other Parts Discussed in Thread: TUSB4020BI

While using the automated USB testing (USB-PET) for DCP mode, I get 2 test failures:

[30.566,146] Checking leakage from D+ or D-

[30.566,146] 7. Connect D+ via 100k to 0V

[30.566,153] 8. Wait 2 seconds to eliminate capacitive effects.

[32.566,153] 9. Check that voltage at D+ is below 1.44V (Two RDAT_LKG (300k)

in parallel, VDAT_LKG = 3.6V).

[32.566,158] - Leakage sourced by D+/D- in spec (V = 0.014V).

[32.566,166] 10. 10. Connect D+ via 100K to 3.6V. Connect D- via 100K to 3.6V.

[32.566,190] 11. Wait 2 seconds to eliminate capacitive effects.

[34.566,190] 12. Check that D+ is not less than 2.65V (Two RDAT_LKG (300k) in

parallel, VDAT_LKG = 0V).

[34.566,197] FAIL: Excessive leakage source by D+/D- (V = 0.636V)

and also

[34.566,216] Checking Capacitance of D+ or D-

[34.566,216] 13. Discharge Standard 1nF capacitor and Capacitance under Test

[34.566,216] Connect 0V to D+ via 1nF test capacitor. Connect 0V to D- via

200R resistor. There is a tested, <200R, resistor between DM

and DP. This will discharge the standard 1nF capacitor and

the capacitance under test to 0V. Wait 10ms.

[34.576,240] 14. Isolate Capacitances

[34.576,240] Disconnect 0V from test capacitor to isolate it. Disconnect

0V from 200R resistor.

[34.576,252] Note: We will now use the D+ voltage watch-block to determine

whether, during the charge-sharing process, D+ rises

above 1.65V. The watch-block amplifier has a limited

band-width by design, so that the watch-block voltage set

is lower than 1.65V. The actual value can be found in the

test script.

[34.576,252] 15. Share Charge Between Capacitances

[34.576,253] Set DP watch-block to be testing for voltage less than the

value required. Connect 3.3V to D+ via 1nF test capacitor.

Wait 1ms. This allows for charge sharing between standard 1nF

capacitor and capacitance under test.

[34.577,276] 16. Read watch-block to see if voltage on DP went above 1.65V. If

it did, then the capacitance under test is less than 1nF and

therefore in specification.

[34.577,283] FAIL: Capacitance exceeds 1nF.

[34.577,295] End of Test

[34.577,309] FAILED TEST - (Does not prevent further tests).

  • Hello Olan,

    The capacitance failure is expected. The TUSB4020B was designed prior to the addition of the capacitance test to the USB Battery Charging specification. The other test failure is not expected, can you confirm that automode is disabled via EEPROM or OTP EFUSE? If Automode is enabled, the TUSB4020B will attempt to enter other battery charging states that are not compliant to the USB BC 1.2 specification.

    Regards,
    JMMN
  • Hi JMMN,

    Strapping mode and EEPROM mode both yield the same test failures.  Each disables the AutoMode.

    The EEPROM contents are:

    00 55 01010101 'ROM Signature - Do NOT modify!
    01 51 01010001 'Vendor ID LSB - Can be modified
    02 04 00000100 'Vendor ID MSB - Can be modified
    03 25 ‭00100101‬ 'Product ID LSB - Can be modified
    04 80 10000000 'Product ID MSB - Can be modified
    05 10 00010000 'Device Configuration - Can be modified
    06 03 00000011 'Battery Charging (1 = enabled) - Can be modified
    07 03 00000011 'Device Removable- Can be modified
    08 00 00000000 'Port Used - Can be modified
    09 00 00000000 'Phy Custom - Can be modified
    0A 22 00100010 'Device Configuration 2 - Can be modified
    20 09 00001001 'Language ID - Can be modified
    21 04 00000100 'Language ID -  Can be modified
    22 18 00011000 'Serial Number String Length -  Can be modified
    23 00 00000000 'Manufacturer String Length
    24 00 00000000 'Product String Length
    F0 00 00000000 'Additional Features
    F2 00 00000000 'Charging Port Control Register

    Thanks - Olan

  • Hi Olan,

    Ah, I didn't realize that the automode is disabled by default on this hub.

    The other reported failure also occurs in the DCP Resistance and Capacitance Test module that was added after this device was designed. This error is caused by excess leakage due to the single ended receivers being enabled during DCP, not incorrect charging. We found no functional issues with DCP mode due to this leakage.

    Regards,
    JMMN
  • We get the following additional DCP test failure on our custom board (with TUSB4020), in addition to the two failures listed earlier:

    [28.404,352] Checking Resistance between D+ and D-

    [28.404,353] 6. Check that resistance from D+ to D- is less than RDCP_DAT max

    (200R). i.e. Connect 2.0V via 200R resistor to D+, connect

    0.05V via 200R resistor to D-. Measure voltages at D+ and D-.

    The difference must be less than 0.67V.

    [28.414,391] FAIL: Voltage difference 0.706V exceeds 0.67V

    Is this also expected?

     

  • No, that should pass (see below). You may want to try retesting, I noticed that the values on this test can vary.

    [23.279,210] Checking Resistance between D+ and D-

    [23.279,211] 6. Check that resistance from D+ to D- is less than RDCP_DAT max
    (200R). i.e. Connect 2.0V via 200R resistor to D+, connect
    0.05V via 200R resistor to D-. Measure voltages at D+ and D-.
    The difference must be less than 0.67V.
    [23.289,248] - Voltage difference 0.650V is in spec.
  • Hi JMMN,

    I have retested several times on our custom board (3 separate board samples), and continue to fail that test portion ([23.279,210] Checking Resistance between D+ and D-).

    Since we do not have any significant resistance between our USB connector and the TUSB4020BI chip ...

    Can you suggest anything for me to check, that can affect this? 

    Thanks

  • Olan,

    Sorry for the delay, our USB_PET is being sent out for calibration. I will retest the TUSB4020B on the Allion UCPET to see if the resistance is marginal or not.

    Regards,
    JMMN
  • JMMN,

    Results from the Allion UCPET would be very helpful.

    This DCP sub-test (Checking Resistance between D+ and D-) may be a test timing issue between the USB-PET and our custom board.

    The USB-PET measures approx. 211 ohms, and the test spec requires less than 200 ohms.  Measuring by hand with a DMM results in 80 ohms, which would be a “PASS”.

    Thanks - Olan

  • Hi Olan,

    I was able to see the same failing resistance during DCP mode with the TUSB4020B on the Allion UCPET tool that you reported with the MQP tool. I checked my test logs archive and this problem was indeed seen on some units during earlier testing - the failure appears to be marginal and varies from unit to unit. The root cause of the failure is that the USB 2.0 15K pulldowns were left enabled during DCP mode. Again, we did not see any functional charging issue with this, but it was addressed on our next generation of hubs.

    Regards,
    JMMN
  • JMMN,

    Thanks for your testing and explanation.  Please confirm that TI does not have a next generation hub that is pin-compatible with the TUSB4020BI.

    Olan

  • Olan,

    We do not have a next generation hub that is compatible with the TUSB4020BI pin out at this time or on the immediate roadmap.

    Regards,

    JMMN

  • JMMN,

    Thanks for your responses.  Have a good day.

    Olan