LM339: Question about voltage spike on output pin when input reference voltage raising up

Hi Expert,

Uploaded the missing picture as shown below.

Thanks

April

Hi April,

during power-up (rise of 3V3_SB from 0V to 3.3V) for a brief period the supply voltage of LM339 is below the minimum supply voltage necessary for normal operation. During this time interval the output voltage of LM339 cannot be predicted.

Kai 

Hello April,

As Kai mentioned, many things are happening at power-up.

A scope photo showing both of the inputs, the output and the power supply will help clarify the issue. As well as zooming in on the "glitch" to see the shape and timing.

The supply ramp-rate and the other input (COMP_VIN_2p5) behavior is needed to find the root cause.

If this is just a sequencing issue, try removing the three 0.1uF capacitors on the COMP_VREF line, as these caps do add a bit of a lag to the reference voltage when coming up.

We do have some comparators that have an internal power-on reset, which can help with these start-up glitches, such as the TLV7021 family, TLV9022-Q1 & LM393LV. These require that the supply voltage cross minimum supply and about 20-30us delay from then to prevent glitches. Unfortunately, we do not have any in Quad right now, just dual and singles.