• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 33 subscribers
  • Views 1209 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.

TPD4S012: VBUS pin on MCU gets shorted plugging USB cable into hub port.

Genatco
Guru 54077 points
Part Number: TPD4S012
Other Parts Discussed in Thread: EK-TM4C1294XL, TM4C1294KCPDT,

Randomly seeing several EK-TM4C1294XL and TM4C1294KCPDT GPIO (PB1) +5 volt tolerant VBUS pin get shorted, DMM has very low 39-70 ohms to ground after short occurs.

The MCU pin (PB1) eventually gets wacked after several times plugging USB cable into host powered hub. The issues is more slow to reveal if the MCU is DC powered from the same source as the USB Hub. The situation is bleak in a custom PCB (TM4C1294KCPDT) powered via different DC source than USB hub DC from computer.  VBUS power randomly surges during plugging cable, must be well above 5v5 if it shorts GPIO 5v tolerant input pin (PB1).

The case markings on the TPD package are not described in the datasheet where the bottom side angled pad indicates being pin 1 but the plastic case corner dimple appears to be pin 4 and not pin 1 as typical in most all plastic cases..

1. EK-TM4C1294XL has two installed TPD4S012 plastic case dimple pin 4 to ground, is that correct?  

2. Why does it seem the VBUS zener is not protecting USB from over powering the MCU 5v tolerant pin (PB1) on either PCB?

  • 0
    TI__Expert 3500 points
    BP101-

    I see you posted in two places, apologies for splitting the conversation. Can you try to remove the TPD4S012 and see if the problem continues to occur?

    Thanks,
    Alec
  • 0
    TI__Genius 12410 points
    Hello,
    1. I am not sure about a dimple on the packaging. I have ordered samples and will let you know. From the datasheet it only shows the pin 1 marking being on the bottom side with the slanted pin.
    2. The Vbus zener diode has a breakdown voltage of 24V

    Please let me know if you have more questions I apologize if I did not answer them enough.
    Regards,
    Cameron
  • 0 in reply to Alec Forbes
    Guru 54077 points

    Not I posting two places must be forum glitch. The point is the TPD is not protecting the MCU 5v tolerant VBUS pin from over voltage/current. Notice the VBUS zener clamping voltage recommend operating voltage is 0-15v and roughly 13.5v for ESD event fig 4 in data sheet

  • 0 in reply to Cameron Phillips
    Guru 54077 points

    Cameron Phillips said:
    2. The Vbus zener diode has a breakdown voltage of 24V

    I think that Figure 4 graph implies an ESD must be present for the zener to clamp any voltage spike starting at 13.5v, no where near VBUS DC range. And 20v Min BVD perhaps is a bit high to protect an MCU pin that is tolerant just up to 5v5 before it shorts the internal GPIO pin rail diodes?

    Is the TPD zener not protecting the MCU 5v VBUS pin from any DC supply over voltages unless an ESD in the KV range caused the voltage spike? 

  • 0 in reply to Cameron Phillips
    Guru 54077 points

    Cameron Phillips said:
    I have ordered samples and will let you know

    Double checked the TPD case recessed corner dimple is on pin 4 (GND) makes it seem like the subcontractor installed the case 180 degrees out? Also verified on another out of circuit TPD, the pad with slanted corner aligns with top of case printed white dot under the number 38 is on pin 1 so pin 6 is thus VBUS. Datasheet figures of case layout does not confirm top case markings are relative to slanted pad of pin1. That datasheet clarification is important so installer QC can check for proper installation against a datasheet if questionable.

    It seems there could be an issue where a micro mini USB port metal plugs chassis ground suddenly is made common to DC ground of USB hub plug in event. That could cause a voltage surge or excessive current flow in the elevated voltage if VBUS pin is allowed to rise above 5v5dc. Most computers that have external USB ports the jacks metal frame tied to chassis ground not DC ground. That is so ESD events mostly travel on the metal or cable shield and not into DC ground. The TPD needs to be tested in all ESD scenario upon a typical USB port, DC ground or chassis ground being on the metal jacket.

  • 0 in reply to Genatco
    TI__Genius 12410 points
    You are correct that this breakdown is too high for a 5.5V supply because this part was to support 15V Vbus supply.
    The TLP graph that you are referring to only goes up to 1.8A of current or about 1kV IEC61000-4-2 ESD. Also as it shows in the TLP graph the voltage has to reach 24V before it snaps back to the holding voltage and becomes conductive. This is congruent with the breakdown voltage of the diode which means if you have DC event under that level the diode will not be conductive and will not clamp the voltage.

    Also it is important to note there is a big difference between normal zener diodes that are rated for DC over voltage operation and these ESD and surge clamping diodes which are smaller and cannot handle DC OVP events only transients defined in the datasheet.

    Regards,
    Cameron
  • 0 in reply to Genatco
    TI__Genius 12410 points
    Thank you for pointing out that about the pin orientation. I will look at it and see how we can update the datasheet.
    I believe what you are describing is a hot plug scenario where you see ringing on the Vbus line when you plug the cable in correct?
    in that case you will want a part that breakdown on the Vbus line before 24V.

    Regards,
    Cameron
  • 0 in reply to Cameron Phillips
    Guru 54077 points

    Hi Cameron,

    I tested a new TPD VBUS zener via DMM diode check produces 0.860vdc drop. So the one in circuit shorted open when it also shorted out the MCU USB peripheral VBUS pin rail diode. My 7 port mini USB hub had no electrolytic capacitors on VBUS and tied all the ports to one DC supply, (e.g. the computers VBUS) or it could use an external wall plug 110AC switcher to produce 5VDC when computer was turned off. When the TPD shorted open, the hubs VBUS was +4.95v. So it must have been a difference in chassis ground and computers power supply DC ground being roughly 42 ohms measured via DMM that caused a voltage surge

    Cameron Phillips said:
    You are correct that this breakdown is too high for a 5.5V supply because this part was to support 15V Vbus supply. 

    The EK-TM4C1294XL launch pad is shipping with two TPD protecting OTG and ICDI USB ports. Everyone I measure VBUS pin to ground (no power) the ohms are very low 100k or less. A spanking brand new MCU in circuit VBUS measures (8.6 megohms), VBUS ESD rail didoes have taken a few hits. One TPD VBUS zener DMM diode check measure 0.62v with shorted VBUS rail diode on MCU pin but lifted off PCB pad, so it too took a hit.

    Is there even a +15 VBUS USB standard?