1
\$\begingroup\$

I am in the process of designing an input protection circuit for my smart load switch (TPS27SA08)

This circuit is intended to provide reverse polarity protection and overvoltage protection:

input protection

The input voltage of the circuit is 12V, 8A is the maximum current.

D1 is a bidirectional TVS diode with 17.1V maximum clamping voltage at 10A and 12V working voltage

Q1 is a P CHannel MOSFET with Vgs maximum voltage +- 20V and Vds of -55V.

The smart load switch voltage input goes from 8-36V

I have 2 questions:

  1. Do I need to add a fast blow fuse at input for an overcurrent situation or with the tvs is enough?
  2. Will the MOSFET survive an overcurrent and overvoltage scenario? Keeping in mind the maximum clamp voltage of the TVS and the maximum voltage of Vgs of the MOSFET.
\$\endgroup\$
1
\$\begingroup\$

To fully answer this you have to define what the electrical threats are that you are trying to protect your circuit against. For instance, you've used a bidirectional TVS at the input but, what perceived threat is the TVS there to deal with?

You have to decide this. Then you might need to add a fuse or a crowbar circuit or something else. I'd certainly use a protection zener on the gate-source and you don't need anything like a 2 watt resistor feeding the gate.

Do I need to add a fast blow fuse at input for an overcurrent situation or with the tvs is enough?

If it's an ESD threat then certainly you won't need a fuse because ESD events are well defined in energy levels and peak currents and most TVS devices are designed to work with ESD. However, if the "threat" is indirect lightning surges you may well have to fit a fuse should the TVS diode "operate" but fail short circuit. However, the 1N6384 is a beast and can handle a 600 watt "dump" for 10 ms i.e. it certainly looks good for some indirect lightning events. But, the devil is in the detail.

I reiterate - define the threats.

\$\endgroup\$
5
  • \$\begingroup\$ Thanks so much Andy!! Well my threads would be if I understand ok overvoltage (more than 20V, for example 100V ) and overcurrent (more than 10A for example 25A). \$\endgroup\$ – marcosbc Feb 5 at 13:00
  • \$\begingroup\$ No, you need to define the threats in terms of recognized real events that have a voltage source and an impedance. For instance the Human body model as per IEC61000-4-2 or an indirect lightning surge as per IEC61000-4-5 for example. Without knowing the exact nature of the threat you cannot determine if a particular TVS will survive or not. \$\endgroup\$ – Andy aka Feb 5 at 13:05
  • \$\begingroup\$ Then, for each you need to define the high voltage source withstand rating of your design such as this for indirect lightning. Now, if you can do that, you can properly design-in the right TVS diode or MOV or GDT and, you'll be able to work out what fuse rating is applicable. \$\endgroup\$ – Andy aka Feb 5 at 13:07
  • \$\begingroup\$ You see, if you say your threat is 100 volts, it will blow the TVS every time or, it will blow the fuse every time so, you need to think about this a bit more. The threat waveform is pretty much done and over in a few tens of microseconds hence, you have that benefit i.e. the protection device has to withstand the surge for only a limited period of time. But you may have "other threats" specific to your application. ESD and lightning are generic threats and are well-documented world-wide. \$\endgroup\$ – Andy aka Feb 5 at 13:08
  • \$\begingroup\$ excellent explanation! thanks so much andy! \$\endgroup\$ – marcosbc Feb 5 at 13:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.