LMR16010: 10-second Enable delay

Hello

One issue that might come up is that the slow slew rate on the EN input may cause the comparator to chatter near the threshold.

The hysteresis current may help, but might not be enough.

The best way to evaluate this is to test on an EVM and look for any strange behavior on the switch node or Vout.

Also, as you mentioned, you will need a discharge diode between EN and VIN.  This should be a small Schottky.

Thanks

Any specification for the voltage range on EN that could cause chatter?  I could then figure out how long the chattering might last and see if it's a problem.

When the comparator is chattering would it just cause switching to start and stop?  So basically my output could turn on and off during that time, but once EN rises far enough from the threshold it would maintain normal operation?

In any case, would this all be solved by a series resistor between the capacitor and EN?  Suppose I use 100kΩ.  This would cause EN to be 100mV above the capacitor voltage, but that seems fine and I could adjust the capacitance to account for that if I really needed to.  Then when I hit the threshold and the EN current jumps to 4.6uA the EN pin would jump up by 460mV.  This should work, right?  Any recommendations on a minimum amount of hysteresis to make sure I don't run into any chatter?

I have tried the experiment with the series resistor and it seems to work - the charging time is not affected and I get a big jump on the EN voltage when I reach the threshold.  I'd still like your thoughts on this, and a recommended minimum for the hysteresis.

One thing I did notice while monitoring the EN pin is that, as the EN voltage passes through about 800mV, I see a little noise on the EN voltage, as if the current being drawn by the pin is slightly fluctuating.  It's small, only about 200mV.  Best I can tell it starts at 770mV and stops at 840mV.  This is happening far from the 1.2V threshold and doesn't seem to have any negative effect, but it seems noteworthy so wanted to make sure it's not a sign of any potential problems.

Hello

If you are getting good results with your testing, that is encouraging.

You will need to test over your full input voltage, load current, and temperature range; if possible.

Any chattering would show up as the device trying to turn on and off during the slow ramp up.

There is another internal threshold on that pin that starts-up the internal VCC regulator.  It is approximatively

one MOSFET threshold voltage.  I assume that is what you are seeing in the 800mV range.

I will assign your post to the engineer in charge of E2E for August.

Thanks

What am I looking for during this testing?  The presence of chattering?  Or just overall functionality?  

If using a resistor to add hysteresis is there any reason to believe this will work any differently than a normal resistor ladder, or introduce any problems?  

My application itself does not require any hysteresis - the voltage at EN will always be increasing as it passes the threshold, so whatever hysteresis I need will only be to keep the comparator from chattering.  The datasheet does not make any mention of a minimum or recommended hysteresis.  So this is still an open question.  The UVLO hysteresis is 285mV, perhaps that would be a good value to use?  So, a 79kΩ resistor?

Do you have any concerns about the noise at ~800mV?  Or does the VCC regulator explain this, and I can safely ignore the noise?

Is there a min and max for the EN pin current or hysteresis current?  Only the typical values are listed.  The 1uA current sets my startup delay and ideally that would be within about +/-20%

Thank you.

The risk of a problem at startup - would this just be the part "chattering", and it would maybe stop switching momentarily as EN rises through the threshold?  Or could it be a more serious problem like the part failing to start entirely?

If it's just a momentary stop in switching my application should be ok, I can just treat that like a slightly longer startup routine.  But if the part might not startup at all, that's obviously a more serious concern.

I don't see why it won't start up. I think we were more concerned about possible chatter. Depending on the load, this may not be desirable behavior in some applications. 

Got it.

With the resistor in place between the capacitor and EN I get the expected hysteresis step in voltage on EN, the same as I would get with a normal resistor ladder.  Given that is there any reason my circuit would chatter more than a typical resistor ladder circuit?  I'd think mine could possibly even chatter less, since the EN voltage continues increasing even after it hits the threshold.