• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 349 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.

TPS25948: eFuse or Ideal_Diode selection

Vitaliy Gurevich
Prodigy 90 points
Part Number: TPS25948
Other Parts Discussed in Thread: TPS25947, LM7310

Tool/software:

Hi Kunal,

I want your expertise to select the best protection part for our device design.

I've attached a simple diagram to describe the configuration.

The device has two USB PD Source inputs, USB-C & Contacts, finally connected to the same Charger PMIC USB-C inputs, but not simultaneously!

1st is a regular USB-C PD receptacle connected directly to the charger PMIC and protected up to 28V_max with all set of OV/UV/InputCL and support of 5V OTG source.

OR

2nd is a "robust Contact" for the USB-C PD charger source connection in the docking or vehicular, routed to the same charger PMIC USB-C input with all above protections.

Therefore, I am looking for a Fuse or Ideal Diode on the 2nd "robust Contact" routed/shorted to the USB-C receptacle, VBUS input voltage.

This protection should be implemented on the flex, therefore the package should be small and have a simple footprint. 

What is a more suitable part - TPS259480AYWP or TPS259470ARPW or LM73100?

eFuse / Ideal Diode requirements:

1.“Always Enabled” Power path for the 2nd USB PD charger hot plugging connection to the PMIC and PON event

2. Reverse voltage protection - to prevent VBUS Voltage on the contacts when the 1st USB-C PD Source is attached

3. Vin_max_abs <28V USB-C PD

4. VBUS I_load =3-4A maximum

5. INTERNAL B-2-B FETs Rds_on < 20mOhm

6. OVP-adj

7 . UVP-adj

8 . Reverse Current protection (when short circuit on external contacts VBUS+&-)

9. Slew rate control while the 2nd Charger source connection (Auto-Retry)

Block_Diagramm.pdf

Thanks

Vitaliy

  • 0
    TI__Mastermind 38965 points

    Hi Vitaliy,

    I am supporting these devices now. Kunal is supporting high current eFuses.

    Before suggesting anything, can you share your timelines (device sampling requirement by month/year).

    Honestly we don't have a solution which satisfy all your requirements directly without some work arounds.

    Both device do not have always on so EN pin can be pulled up to other pull up power rail (if the end application have one) 

    In case of TPS25948, you get wscp package, typical ron is less than 20mOhms but reverse voltage protection is absent which can be implemented using a diode.

    In  case of TPS25947, package is QFN-10 so little higher package. The typical ron is slightly higher (28mOhms) but will provide reverse voltage protection so no need to add diode at ground. 

    Both devices will provide RCB, Ilim = 4A, OVP, inrush current control i.e. output voltage slew rate control during eFuse startup. 

    Adj UVP won't be present due to pulling EN to logic high but fixed UVP will still be present for both devices.

    TPS25947 might be a better solution here due to size constraints (extra diode not required for Reverse polarity protection). If RPP is not required, then TPS25948 will be better solution.

    Are docking connectors not reversible (similar to USB)?

    Best Regards,
    Arush

  • 0 in reply to Arush Gupta
    Prodigy 90 points

    Hi Arush,

    Thanks for your prompt response.
    1. The project timelines: 
    a. Development starting now
    b. 1st Proto July-August 2025
    c. Product launch is anticipated on July 31, 2026
    d. Quantities by year:
        2026 - 20,000 units
        2027 - 30,000 units
        2028 - 30,000 units
    e. The 3-year forecast: 80,000 units
    2. The docking Contact's connection is not reversible and limited by the mech cradle design, and the USB-C connector should support reversibility
    3. The EN concept on both TPS25947&948 is acceptable. I meant, when VBUS is applied to the sink, has to activate the eFuse power path "automatically" to meet the PD standard:
    • The device/sink will run the Rp/Rd rotation on the CC1&CC2 lines to get the source DFP to attach.
    • The DFP PS/source will connect Rp on CC1 Only contact of the device (not reversible)
    • The Rp/Rd voltage divider level will enable the VBUS switch on the DFP PS/source side, and 5V should be presented on the device/sink within 180msec.
    3. The TPS25947 and TPS25948 EN/UVP and OVLO pins can adjust the OV and UV thresholds by selecting the appropriate R1/R2/R3 resistors. 
     
    4.  You said "Both devices TPS25947 and TPS25948  will provide ReverseCurrentBlocking". 
    When the wall adapter is applied to the USB-C connector, I need to protect the external VBUS contact (no voltage is available for the VBUS external contact). 
    What is expected to be on Vin pin if VBUS > 5V is applied to Vout pin?
     
    5. I see that TPS25948 does not support Vin Reverse polarity protection and TPS25947 is Yes?
    Does it mean that if Vin reverse polarity -XXV is applied, the 948 will be damaged?
     
    6. I did not see your response regarding the LM73100 option?
    To be trusted, I can't see any significant differences between the ideal diode LM73100 and the eFuse TPS259470x. Could you please comment?
     
    7. I am summarizing the Fault conditions supported by eFuse and Ideal Diode.
    The real difference is a Reverse polarity protection. Do I miss something?
    Thanks
    Vitaliy
  • 0 in reply to Vitaliy Gurevich
    TI__Mastermind 38965 points

    Hi Vitaliy,

    Thank you for your patience. 

    Vitaliy Gurevich said:
    1. The project timelines: 
    a. Development starting now
    b. 1st Proto July-August 2025
    c. Product launch is anticipated on July 31, 2026
    d. Quantities by year:
        2026 - 20,000 units
        2027 - 30,000 units
        2028 - 30,000 units
    e. The 3-year forecast: 80,000 units

    Thank you for sharing this. We have an upcoming eFuse but the sampling timeline won't match the proto timeline. 

    Vitaliy Gurevich said:
    4.  You said "Both devices TPS25947 and TPS25948  will provide ReverseCurrentBlocking". 
    When the wall adapter is applied to the USB-C connector, I need to protect the external VBUS contact (no voltage is available for the VBUS external contact). 
    What is expected to be on Vin pin if VBUS > 5V is applied to Vout pin?

    Since the device will go into RCB, ideally we should expect 0V. Some leakage (uA order) can cause some voltage buildup through the resistor divider but that won't be very high.

    Vitaliy Gurevich said:
    5. I see that TPS25948 does not support Vin Reverse polarity protection and TPS25947 is Yes?
    Does it mean that if Vin reverse polarity -XXV is applied, the 948 will be damaged?

    Yes. If negative voltage is applied, in case of TPS25948, gnd to pin current may flow through the esd diodes and may damage to device. 

    Vitaliy Gurevich said:
    6. I did not see your response regarding the LM73100 option?
    To be trusted, I can't see any significant differences between the ideal diode LM73100 and the eFuse TPS259470x. Could you please comment?

    LM7310 is stripped down variant of TPS25947 but doesn't provide steady state current limiting hence I mistakenly ignored that in my first response.

    Vitaliy Gurevich said:
    7. I am summarizing the Fault conditions supported by eFuse and Ideal Diode.
    The real difference is a Reverse polarity protection. Do I miss something?

    Overall you have covered most of the things. I will just add some comments on this.

    1) After current limiting, the device will go into thermal shutdown depending on the power dissipation on the device. 

    2) here also, after current limiting, depending on the Vin, you may see device going into thermal shutdown very quickly. 

    Adjustable OVLO is different from 2*ILIM. We call this scalable current limiting as scalable fast trip (SFT). This is also present in TPS25947 device. This mechanism provide protection against the short circuit.

    Please let me know if there are any follow up questions.

    Best Regards,
    Arush

  • 0 in reply to Arush Gupta
    Prodigy 90 points

    Hi Arush,

    Thanks for your clarifications. I will check regarding RPP and let you know which p/n we will use.

    1. Regarding the quantities and project timelines, we will have 3-4 prototypes during the development stage from August 25 to May 26. We will order the prototype quantities accordingly. I'm not sure we will need the samples for evaluation. Maybe we will buy the EVB and a couple of samples.

    The quantities below are for the final production per 3 years.

    Quantities by year:
        2026 - 20,000 units
        2027 - 30,000 units
        2028 - 30,000 units
    The 3-year forecast: 80,000 units

    1. It seems like I will go to use one of the following versions:

    TPS259474ARPW (Most of Shutdown protections) + Adj OVLO / Breaker / Auto-Retry

    TPS259480AYWP (Most of Shutdown protections, No RPP, No OV Voltage clamp)+Adj(2×ILIM)/Auto-Retry

  • 0 in reply to Vitaliy Gurevich
    TI__Mastermind 38965 points

    Hi Vitaliy,

    Understood.

    I am closing this thread for now. You can reopen this thread if you have any follow up question or create a new thread and include this link in description. 

    Best Regards,
    Arush