Over voltage protection circuit for automotive use

[Muku]

Hello, everyone.

I am working on a reverse voltage protection circuit for use in my motorbike which uses 12V power system. As other users here suggested I should use a protection circuit to protect my device from voltage spike which occur in vehicle power bus due to the alternator. Basically when load is disconnected from alternator it dumps a huge amount of energy on the power lines and this can cause huge voltage spikes upto 120 -300 volts. In 12v systems it is clamped as 40v. I need to protect my electronics from this abrupt voltage spikes. I found this great circuit on ti website,
https://www.ti.com/lit/an/snva717/snva717.pdf?ts=1598604209548&ref_url=https%253A%252F%252Fwww.google.com%252F


I am using a switching regulator that has 14v input voltage and I need to keep the voltage around 14 and if >14 the circuit needs to turn off. I can achieve this by exchanging the 18V zener to 14v.
But I have few questions about this circuit,

First, the the documents that this circuit is taken from doesn't really say how high voltages this can protect from. But let assume we have load dump situation which is clamped to 40v so our input voltage is +40v



Then emitter of of the PNP transistor will see +40V and the base is at +18v because zener has started conducting that means base emitter voltage is -26V which more than EB base breakdown of the PNP.
This should destroy the PNP, right? I might be wrong, correct me if I am wrong.


 

[bdunham7]

The E-B breakdown voltage is for reverse voltage, here it will be conducting forward and dropping 0.6V.  Where you have marked "18V" should be 39.4V instead.  The other voltage will be dropped across the resistor.

[Doctorandus_P]

You are forgetting resistor "ROVS2".
When input is 40V, then the emitter of the MMBT3906 is at 39V6.
It's base is still 600mV lower, at 38V8.
The zener eats up 18V, so the current through ROVS2 is:
 (38.8 - 18)/ 27400 = 0.0007591240875912408

This opens the transistor, and shorts the Gate and Source of the P-FET.
Result is that everything after the P-FET is turned off during an overvoltage event.
The maximum voltage levels it can block are mostly determined by the P-FET, and the BAV99 diodes.

This is the normal way automotive overvoltage protection work.
Because the voltages and power levels are so high, you can not simply shunt excess energy to GND, as that would blow up most protection devices.

[Muku]

So, I need to implement the entire circuit for this to work? I was hoping I could just use the Over voltage disconnect part since the switcher I am using already has a soft start and current limiting feature and adding a PTC fuse would suffice. What would be a cheap way to implement a crude over voltage protection? Would is be possible to use just a TVS diode like woth ESD protection?

[Muku]

I drew a circuit based on the above circuit


Would this circuit be of any use? What would be the problem with such a circuit?

[MasterT]

What kind of electronics - how much current does it consume?  For 1-5 amps you can use common 3-pin voltage regulators, most  of them could withstand 40V at the input.

[Muku]

What kind of electronics - how much current does it consume?  For 1-5 amps you can use common 3-pin voltage regulators, most  of them could withstand 40V at the input.

Do you mean linear regulators? It is going to be a battery operated device with a small 350mAh battery so, linear regulators wouldn't really be a good choice.

[MasterT]

I see. Your simplified version of circuits seems O'K for me. What I would add, is a small series resistor & cap at the output, though capacitance inside of the load may be enough to mitigate slow MOSFET response, 100-s nanosec or so.
See also reply #5 https://www.eevblog.com/forum/beginners/zener-failed-short/new/?topicseen#new

[bdunham7]

So, I need to implement the entire circuit for this to work? I was hoping I could just use the Over voltage disconnect part since the switcher I am using already has a soft start and current limiting feature and adding a PTC fuse would suffice. What would be a cheap way to implement a crude over voltage protection? Would is be possible to use just a TVS diode like woth ESD protection?

I would retain the gate clamp in an circuit specifically designed to deal with transient overloads and inductance.  Other than that I don't see why it wouldn't work on its own as you have drawn.

You can't use plain clamping like a TVS or zener crowbar in this situation unless you can use a resistor or PTC before it to limit the maximum input current.  If your device draws less than 1A, this might be feasible, but beyond that you'd need heavier duty clamping devices.

[PeteH]

I like the linear regulator idea...  Set it for 14 or 15V so it's in dropout. Get a low drop out one... It'll ride through transients and clamp (regulate its output). If it gets too hot it'll timeout. Self protected.... Nominally at 12 or 14 it'll have low power loss since it's fully enhanced....

[Muku]

Thank you everyone for your replies. My device draws 2.5 amps, 3 amps just to be safe but that for small period of time when it is operating. Most of the time it is in sleep mode. It is a 433Mhz radio module which transmits GPS location from on board GPS. It operates in legal power range but due to Li-ion battery charger and GPS receiver the current consumption adds up.

If I may ask what is purpose of the the gate clamp, is it to maintain the gate to source voltage so the mosfet is safe? and would removing the inrush current control be of any problem. I have 22uF ceramic cap on the input side of my switcher which is MP2617H buck converter with li-on charger.

In the above circuit if mosfet drain to source voltage is less than -40v than it would be destroyed? So, the maximum voltage the circuit can protect against is the drain-source voltage of the mosfet?

 

[Muku]

I like the linear regulator idea...  Set it for 14 or 15V so it's in dropout. Get a low drop out one... It'll ride through transients and clamp (regulate its output). If it gets too hot it'll timeout. Self protected.... Nominally at 12 or 14 it'll have low power loss since it's fully enhanced....

It would be great if I could use a Linear Regulator but the I have to drop the voltage to 4.2v and then 3.3v from 14v for everything to operate.