• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 62 subscribers
  • Views 670 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.

Original question:

LM74930-Q1: Power supply reverse question

LM74930-Q1: Negative transient voltage and reverse current capability

Artur Bauder
Prodigy 10 points

Part Number: LM74930-Q1

 hi experts, 

If I deactivate reverse current blocking using the mode pin, will the Transistor Q2 stay on while there is a surge at the input?

In the first picture an empty DC-Link is assumed and in the second picture the DC-Link Cap is at 54,8V.

It is the negative surge voltage (-126,3 V clamped with TVS at VIN), which gives me a headache.

If Q2 stays on, providing reverse current capability, the DC-Link would be discharged, and this is fine for my application.

If Q2 switches off, the maximum C-A voltage (85V) would be violated.

Unfortunately, I can’t use any lower voltage TVS in negative direction at VIN, because I need the circuit to withstand permanent reverse voltage of -54,8 V.

best regards, 

Artur

  • 0
    TI__Guru** 107731 points

    Hi Artur

    Welcome to E2E!

    What is the application? why are you expecting negative voltages in the range of -126,3 V ? and how long this interval lasts ?

    If VS is +ve, then Q2 remains in ON state when reverse current blocking is disabled. However, your supply VIN gets shorted through Q1 and Q2 - right ?

    BR,

    Rakesh

  • 0 in reply to Rakesh
    Prodigy 10 points

    Hi Rakesh, 

    the application is an electromotoric brake actuator for railway. The circuit needs to survive +/-1kV 42 Ohm surge pulse with typical 8/20µs current waveform. That's why i have bidirectional TVS with 126,3 V maximum clamping voltage at the input so far. Of course, the additional Zener clamp with 60 V at VIN will also be used.

    What do you mean with +ve? Yes, the VIN supply would be shorted through Q1 and Q2. 

    The circuit will be supplied with nominal 48 V and max. +54,8 V, but needs to survive reverse voltage of -54,8 V permanently.

    Assume an empty DC-Link (left picture) and a -1kV surge pulse, the clamping voltage of -126,3 V would be too high, right?

    regards, 

    Artur

  • 0 in reply to Artur Bauder
    TI__Guru** 107731 points

    thanks for the details Artur.

    1kV 42 Ohm surge means around 24A of surge current. So, a 60V bi-directional TVS diode such as SMC30J58A/CA https://www.st.com/resource/en/datasheet/smc30j10a.pdf helps to clamps to very low voltage than 126.3V 

    Vcl = Vbr(max) + surge current * Rd(max) 

    =71.4 + 24*0.153 = 75V.

     

    Now the voltage across Q2 FET will be +54,8 V +75V = 130V. 

    You need to select Q2 with Vds rating higher than 130V

    To support higher voltage between A and C pins, we can use a depletion NFET like BSS126 for Q3 which disconnects C pin from VOUT when A becomes negative than C.

    Please get an EVM and try out with these changes and let me know.

    Thanks

    Rakesh

  • 0 in reply to Rakesh
    Prodigy 10 points

    hello Rakesh, 

    i didn't see a differential resistance in TVS datasheets before and always calcualted with max. clamping voltage, so thank you for the example!

    i recently also found an intersting TVS with very low clmaping voltage:

    The depletion FET is a great idea! i will try and test the circuit as soon as possible and will share my results with you.

    Thanks 

    Artur

  • 0 in reply to Artur Bauder
    TI__Guru** 107731 points

    yes Artur.

    This TVS diode looks better.

    Please evaluate and let us know.

    Best Regards,

    Rakesh