• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 104 subscribers
  • Views 336 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TDA4AP-Q1: Bring up no stable switch on for TPS6594 PMIC possible

Thomas Paula
Prodigy 60 points
Part Number: TDA4AP-Q1

Hello,

we're in the bring up phase for our TDA4AP design. We followed for the Power supply implementation SLVUIC2 PDN 3A solution.

1) When switching on the PMIC with PMIC enable pin (ENABLE pin 20) the PMIC starts to ramp up the voltage rails and after around 10ms we also get High signals for GPIO8 and GPIO10 from the external PWRGRP Pins.
2) after around 12ms the PWRGRP signals on GPIO8, GPIO10 switch back to low and stay low.
3) From that stage the PMIC retries to enable its rails again until the rmax retry count has reached
4) We identified a higher current than expected on BUCK3 VDD_RAM_0V85 (max. 5Amp) but should be fine due to limit at 5.5A
5) We also see in the log see end of the mail that 0x5D register shows an ILIM on BUCK3

6) For further investigation we disconnected GPIO8, GPIO10 and put there two pull ups instead. Then the retry behaviour is no longer visible and the rails come up without retring and also the POR reset signals from PMIC to SOC (NRESETOUT and GPIO11) are released.
Unfortunatelly trough the high on GPIO8 during power up the  BUCK5 is obviously not switched on and therefore also not LDO_C (missing enable from VDD_MCU_0V85).

What could be the root cause for the
a) high current flow there on BUCK3 (we did an x-ray of the board so there should be no soldering issues)
b) why the problem with BUCK3 obviously does not exist with the option described under 6)

What else can we check or which other signals can may influence the power up of the PMIC?

Thanks for you answer in advance.

Kind regards
Thomas Paula

PMIC register dump from address 0x48. (offset in decimal)

[0] '\0' (0x00)
[1] '‚' (0x82)
[2] ':' (0x3A)
[3] '.' (0x03)
[4] '"' (0x22)
[5] ',' (0x2C)
[6] '"' (0x22)
[7] ',' (0x2C)
[8] '"' (0x22)
[9] ',' (0x2C)
[10] '"' (0x22)
[11] '+' (0x2B)
[12] '"' (0x22)
[13] '.' (0x1C)
[14] 's' (0x73)
[15] 's' (0x73)
[16] 's' (0x73)
[17] 's' (0x73)
[18] 'A' (0x41)
[19] 'A' (0x41)
[20] '²' (0xB2)
[21] '²' (0xB2)
[22] 'A' (0x41)
[23] 'A' (0x41)
[24] '.' (0x1B)
[25] '.' (0x1B)
[26] '.' (0x1B)
[27] '.' (0x1B)
[28] '.' (0x1B)
[29] ' ' (0x20)
[30] ' ' (0x20)
[31] ' ' (0x20)
[32] ' ' (0x20)
[33] '\0' (0x00)
[34] '\0' (0x00)
[35] '8' (0x38)
[36] 'ô' (0xF4)
[37] '.' (0x10)
[38] '8' (0x38)
[39] '.' (0x1B)
[40] '.' (0x1B)
[41] '.' (0x1B)
[42] '.' (0x1B)
[43] '.' (0x01)
[44] '?' (0x3F)
[45] '\0' (0x00)
[46] '\0' (0x00)
[47] '\0' (0x00)
[48] '\0' (0x00)
[49] ' ' (0x20)
[50] '@' (0x40)
[51] 'P' (0x50)
[52] 'Ð' (0xD0)
[53] '.' (0x01)
[54] '.' (0x03)
[55] '0' (0x30)
[56] '.' (0x18)
[57] '.' (0x01)
[58] '.' (0x10)
[59] 'C' (0x43)
[60] '.' (0x01)
[61] '\0' (0x00)
[62] '\0' (0x00)
[63] '\0' (0x00)
[64] '\b' (0x08)
[65] '¥' (0xA5)
[66] '•' (0x95)
[67] '.' (0x05)
[68] '.' (0x1E)
[69] '.' (0x01)
[70] 'U' (0x55)
[71] 'Õ' (0xD5)
[72] '.' (0x1D)
[73] '.' (0x03)
[74] '3' (0x33)
[75] '.' (0x03)
[76] '3' (0x33)
[77] '3' (0x33)
[78] '\0' (0x00)
[79] 'ÿ' (0xFF)
[80] 'ÿ' (0xFF)
[81] '?' (0x3F)
[82] '.' (0x11)
[83] '.' (0x02)
[84] '0' (0x30)
[85] '\0' (0x00)
[86] '\0' (0x00)
[87] '\0' (0x00)
[88] '\0' (0x00)
[89] '\0' (0x00)
[90] '»' (0xBB)
[91] '.' (0x06)
[92] '\0' (0x00)
[93] '\b' (0x08)
[94] '.' (0x02)
[95] '.' (0x02)
[96] '\0' (0x00)
[97] '@' (0x40)
[98] '\0' (0x00)
[99] '\0' (0x00)
[100] '\0' (0x00)
[101] '.' (0x02)
[102] '.' (0x01)
[103] '\n' (0x0A)
[104] '\0' (0x00)
[105] '.' (0x06)
[106] '\0' (0x00)
[107] '\0' (0x00)
[108] '\0' (0x00)
[109] '\0' (0x00)
[110] '\0' (0x00)
[111] '\0' (0x00)
[112] '\0' (0x00)
[113] '\0' (0x00)
[114] '\0' (0x00)
[115] '.' (0x02)
[116] '\0' (0x00)
[117] '\0' (0x00)
[118] '\0' (0x00)
[119] '\0' (0x00)
[120] '\0' (0x00)
[121] '\0' (0x00)
[122] '\0' (0x00)
[123] '\0' (0x00)
[124] '\0' (0x00)
[125] '\0' (0x00)
[126] '\0' (0x00)
[127] '\0' (0x00)
[128] '\0' (0x00)
[129] '\f' (0x0C)
[130] '.' (0x18)
[131] '.' (0x0F)
[132] '.' (0x0F)
[133] '\0' (0x00)
[134] '\0' (0x00)
[135] '.' (0x1F)
[136] '\0' (0x00)
[137] '\0' (0x00)
[138] '.' (0x1F)
[139] '\v' (0x0B)
[140] 'ÿ' (0xFF)
[141] 'ÿ' (0xFF)
[142] '\0' (0x00)
[143] '\0' (0x00)
[144] '\0' (0x00)
[145] '\0' (0x00)
[146] '\0' (0x00)
[147] '\0' (0x00)
[148] '\0' (0x00)
[149] '\0' (0x00)
[150] '\0' (0x00)
[151] '\0' (0x00)
[152] '\0' (0x00)
[153] '\0' (0x00)
[154] '\0' (0x00)
[155] '\0' (0x00)
[156] '\0' (0x00)
[157] '\0' (0x00)
[158] '\0' (0x00)
[159] '\0' (0x00)
[160] '\0' (0x00)
[161] '\0' (0x00)
[162] '\0' (0x00)
[163] '€' (0x80)
[164] '\0' (0x00)
[165] '\0' (0x00)
[166] '\b' (0x08)
[167] '\0' (0x00)
[168] '\0' (0x00)
[169] '\0' (0x00)
[170] '\0' (0x00)
[171] '\0' (0x00)
[172] '\0' (0x00)
[173] '\0' (0x00)
[174] '\0' (0x00)
[175] '\0' (0x00)
[176] '\0' (0x00)
[177] '\0' (0x00)
[178] '\0' (0x00)
[179] '\0' (0x00)
[180] '\0' (0x00)
[181] '\0' (0x00)
[182] '\0' (0x00)
[183] '\0' (0x00)
[184] '.' (0x01)
[185] '.' (0x01)
[186] '\0' (0x00)
[187] '\0' (0x00)
[188] '\0' (0x00)
[189] '\0' (0x00)
[190] '\0' (0x00)
[191] '\0' (0x00)
[192] '\0' (0x00)
[193] '\0' (0x00)
[194] '\0' (0x00)
[195] 'à' (0xE0)
[196] '€' (0x80)
[197] '\0' (0x00)
[198] '\0' (0x00)
[199] '\0' (0x00)
[200] '\0' (0x00)
[201] '\0' (0x00)
[202] '\0' (0x00)
[203] '\0' (0x00)
[204] '.' (0x01)
[205] '-' (0x2D)
[206] '' (0x9D)
[207] '\0' (0x00)
[208] '\0' (0x00)
[209] '\0' (0x00)
[210] '\0' (0x00)
[211] '\0' (0x00)
[212] '\0' (0x00)
[213] '\0' (0x00)
[214] '\0' (0x00)
[215] '\0' (0x00)
[216] '\0' (0x00)
[217] '\0' (0x00)
[218] '\0' (0x00)
[219] '\0' (0x00)
[220] '\0' (0x00)
[221] '\0' (0x00)
[222] '\0' (0x00)
[223] '\0' (0x00)
[224] '\0' (0x00)
[225] '\0' (0x00)
[226] '\0' (0x00)
[227] '\0' (0x00)
[228] '\0' (0x00)
[229] '\0' (0x00)
[230] '\0' (0x00)
[231] '\0' (0x00)
[232] '\0' (0x00)
[233] '\0' (0x00)
[234] '\0' (0x00)
[235] '\0' (0x00)
[236] '\0' (0x00)
[237] '\0' (0x00)
[238] '\0' (0x00)
[239] '\0' (0x00)
[240] '\0' (0x00)
[241] '\0' (0x00)
[242] '\0' (0x00)
[243] '\0' (0x00)
[244] '\0' (0x00)
[245] '\0' (0x00)
[246] '\0' (0x00)
[247] '\0' (0x00)
[248] '\0' (0x00)
[249] '\0' (0x00)
[250] '\0' (0x00)
[251] '\0' (0x00)
[252] '\0' (0x00)
[253] '\0' (0x00)
[254] '\0' (0x00)
[255] '\0' (0x00)

  • 0
    TI__Guru**** 155180 points

    Hi,

    I have looped in the hardware expert. We will get back to you in a day or two.

    Thanks,
    Keerthy

  • 0
    TI__Mastermind 23645 points
    Thomas Paula said:
    2) after around 12ms the PWRGRP signals on GPIO8, GPIO10 switch back to low and stay low.

    If GPIO10 goes low, the PMIC will react the same as if there was a MCU_PWR_ERR. It will do a recovery attempt.

    Also from the register dump, there are faults on BUCK5 and LDO4 as well. BUCK3 ILIM would not cause a recovery attempt but the BUCK5 and LDO4 faults would.

    Thomas Paula said:
    What could be the root cause for the
    a) high current flow there on BUCK3 (we did an x-ray of the board so there should be no soldering issues)
    b) why the problem with BUCK3 obviously does not exist with the option described under 6)

    Is it possible that with GPIO10 and GPIO6 both pulled high, something else in the system doesn't have as big a pull on BUCK3? In either case, the BUCK3 ILIM fault would not be reason for shutdown.

    Can you get a scope capture of BUCK5 and LDO4 right before the first power down? 

    The PWRGRP signals come from the voltage monitors in this PDN. Have you checked the voltages that they monitor? Do both go low at the same time?

  • 0 in reply to Michael Gambrill
    Prodigy 60 points

    Hello Michael,
    we did a scope capture for BUCK5 and LDO4 when the first real pulse occurs.

    Start of recovery:
    Before the first pulse goes to 0.85V there is also a small one, but could be also caused from switching on the supply for the board.




    Zoom in to first high pulse, here we saw this short drop around 835mV first and then overshoot to 1.12V after that on Buck5.



    Zoom in second high pulse, the drop is not really visible any more:


    The subsequent ones are same as the second one.

  • 0 in reply to Thomas Paula
    Prodigy 60 points

    In addition also measured at the feedback resistor for BUCK5, and PMIC_ENABLE signal as reference.
    On feedback signal the drop is also visible, around 780mV for maybe 10us. Quality of scope is limited here.

    The first low pulse is from powering on the board and does not affect the others (see l/h change on PMIC_ENABLE).


    Original shot:



    Zoom in into first high pulse voltage drop

  • 0 in reply to Thomas Paula
    Prodigy 60 points

    Update to the issue:

    we change the sensing point for BUCK5 from the point close to the SoC to the point close to indcutor, we need to check here if there could be a layout issue.
    The recovery of the rails disapeared with that. We need to do additional measurements for other rails and IO signals from PMIC.

    Two topics are still left:
    -The error logging on LDO4 register 0x61h  LDO4_SC_INT is still there. Where is the sampling point for detecting this error?
    -On the BUCK5 rail below the SoC at caps there voltage is stable, but voltage is only ~831mV. We assume that the voltage is high enough for further bring up?

  • +1 in reply to Thomas Paula
    TI__Mastermind 23645 points

    Issues resolved with customer via conference call.

  • 0 in reply to Michael Gambrill
    TI__Guru**** 155180 points

    Michael,

    That's awesome! For my understanding what was the root cause and fix?

    Best Regards,

    Keerthy 

  • 0 in reply to Keerthy J
    TI__Mastermind 23645 points

    BUCK5 FB measured an under voltage event. Experiment showed BUCK measured 0.78V near the SOC, which is close to the UV threshold. Recommendation is widen trace of BUCK5 to reduce voltage drop from local point to SOC.