LM5069: Does the MOSFET gets switched off during surge event?

Hi Robin,

Thanks for reaching out!

During surge event, the fast rising voltage at the input increases the current in the path much beyond the circuit breaker current threshold of LM5069. This causes controller to fast shutdown the MOSFET to ensure FET protection. Though not recommended, the turn-off can be avoided by disabling the overcurrent protection and by setting the OVP threshold above the clamping voltage of the input TVS diode. 

Can you share your system specifications and protection needs ? Does your system need overcurrent protection ? what is the load and how much voltage it can withstand.? What is the end product?

Best Regards, Rakesh

Hi Robin,

Can you share the requested information ?

Best Regards, Rakesh

Hello Rakesh

Sorry for my late reply, I was not in the office the last days.

The system runs at 24VDC +-5% and needs to be protected from:

- Surge

- Continous reverse polarity

- Continous overvoltage up to 60VDC

Overcurrent protection is not really needed. The load is a complete carrier board for a computer on module. So the voltage critical parts are the DC/DC converters which can whitstand 45VDC (abs max rating).

Thanks for your help.

Best regards

Hi Robin,

Thanks

Can you look at TPS2663  and let me know

Best Regards, Rakesh

Hello Rakesh

Current can be up to 10A so TPS2663 does not work.

So do you see any other solution for a continous overvoltage protection where a surge event does not cause a cut off of the supply? 

Best regards

Hi Robin,

TPS2663 is the best device to handle surge events from our portfolio. It can be used in parallel as per https://e2e.ti.com/support/power-management/f/196/p/899239/3325203#3325203 

Hi Rakesh

Following questions:

1. What is better about the TPS2663 compared to LM5069?

2. Does the TPS2663 not trigger at a surge event?

3. The max voltage of the TPS2663 is 60V, so I can not go higher than 36V with the TVS diode, right?

4. If I want to use multiple TPD2663 in "parallel" would there be any drawback/benefit if I do not connect the outputs together but split my load?

Best regards

Hi Rakesh

What would be the drawbacks of the solution in chapter 4 of the following document:

https://www.ti.com/lit/an/snva683/snva683.pdf?ts=1592998043921&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FLM5069

With this circuit a surge would not trigger the switching off of the FET.

Best regards

Hi Robbin,

Any device which does overcurrent protection will get turned OFF during surge event due to fast rising currents from IN to OUT. When the FET is turned OFF, the holdup cap at the output supports the load before the system recovers back (FET turn-on). With TPS2663, the recovery is fast compared to LM5069, which helps in optimizing the holdup cap needs.

You can use bigger TVS such as SMC package to offer better clamping. 

Splitting the load into branches with independent TPS2663 is good idea, if it is possible in your system.

Best Regards, Rakesh

Hi Rakesh

These fast rising currents you are talking about come from charching the capacitors to the higher voltage, right? Apart from that I don't see any reason for rising current when voltage gets higher because there is no ohmic load but DC/DC converters. 

In chapter 4 of the above linked document it says: "In some cases, the system needs to “clamp” the output voltage such that the peak OVLO voltage is maintained during an over-voltage surge".

So the voltage is clamped at the output. If it is clamped to a voltage not much higher than the nominal voltage isn't it possible to achieve that the circuit doesn't get turend off during a surge?

Thank you really much for your help. I really appreciate it.

Best regards