LM5118: Only 10mV output

Hello Calvin,

Thank you for using our device. I am sorry, but I have some problems understanding your explanation above.

You want to make sure that the converter is turning off when the battery voltage gets below 12.1V and the IGNITION IN signal is low.

Then you want to keep the device off until the IGNITION IN signal gets high.

But if the IGNITION IN signal is high, the device shall ignore if VIN falls below 12.1V.

Is this understanding correct?

What shall happen if the battery voltage is above 12.1V and the IGNITION IN signal is getting low again? Shall the converter stay on or follow the IGNITION IN signal and turn off?

Hey Brigitte,

I do a bad job of explaining this board to people, I've found this example helps: Imagine you wanted to listen to your radio in your car with the ignition key off, but you don't want your radio to kill your car's battery. So this board would automatically disconnect the radio when the battery voltage reached a certain point (~12.1V). At the same time, you want the radio come back on when you turn the key to 'ignition', and you need the radio to stay powered while your starting your car.

You want to make sure that the converter is turning off when the battery voltage gets below 12.1V and the IGNITION IN signal is low.

Yes. This is an automotive application if that helps.12.0-12.1V is a good threshold for turning off electronics that are kept alive when the ignition key (Ignition In) is off (low) to ensure the vehicle will start. 

Then you want to keep the device off until the IGNITION IN signal gets high. 

No. When Ignition In is brought 'low' (key off), CN1 Pins 6-8 should remain 'high' in order to power the boards load until the enable pin has been taken <0.5V (hence RDIV1 and RDIV2. The spec sheet says that the enable pin must be brought to >3.5V after a shutdown occurs. That's the reason for DIGN.). Ignition In also controls the gate of the ignition FET, which is the only circuit I want to mimic the "Ignition In" signal.

This device is doing two things:

1. Voltage regulation to keep electronics from restarting during engine starting due to voltage transients (where the ignition contact would be 'high'). The ignition fet should only be 'high' when Ignition In is high. 
   
2. Disconnecting it's load in order to keep the vehicles battery from being drained too much (ignition in would be low in this scenario). The spec sheet says that the enable pin can be used to turn off the LM5118. That's what I was attempting to achieve with the voltage divider (RDIV1 and RDIV2). CN1 Pins 6-8 should stay high even with Ignition In low, but CN1 pins 6-8 should be brought low (disconnected) if the vehicles battery voltage reaches about 12.1V (<0.5V on the enable pin of the 5118 thanks to RDIV1 and RDIV2).

My problem now is that I'm only getting somewhere between 10-50mV at the output (that range is from me changing RFBT & RFBB resistors to slightly different values in order to troubleshoot). On one of my test boards, I've pulled Q1, RIGN2, RIGN_PD, and LED1 off of the board to rule out any issues with that circuit, but had the same test results as when they were still on the board. Currently, if Ignition In is kept low, I get nothing at the output. Only when I bring Ignition In 'high' do I get the 10-50mV output that I mentioned. I've put the board under a thermal camera to try and identify anything that could possibly be shorted, overheating, etc, but nothing looks out of the ordinary. 

To add to troubleshooting, I've measured the UVLO pin and it only has 50mV. I believe that means it's in hiccup mode.

I've noticed on the eval board that there's a 0.1uF cap parallel to the UVLO divider. WeBench didn't recommend that, so I left that out. I would hope the input capacitors would knock any noise out of VIN.

Bump.

Could I get some feedback, please?