I presented this issue to TI support and they bailed out on me and suggested I use this forum to see if anyone could offer a suggestion or solution, I am about to try the TLV3401 comparator to see if I can use it in this application.
Customer is using the SN74AUP1G14, wants to trigger a timed external reset on a microprocessor for a low power application. His circuitry is a capacitor on the input to the gate that will charge through a resistor until the threshold of the inverter is reached, when the schmidt/inverter changes state it will provide a reset signal to the microprocesser, the microprocessor will then reset and run for a short period and will go to sleep, just before entering the sleep condition a port will be used to reset the capacitor, starting a new timing cycle, at least that is the theory. The circuit was breadboarded and appeared to work fine until I monitored the actual current drain. As the capacitor ramps up or down the current being drawn by the 74AUP1G14 also ramps up and just before it changes state it is drawing around 200 microamperes. This of course totally defeats the idea of using this as a low power external reset control. I used a seperate power supply to provide the charge current to the capacitor so only the 74AUP14 current was monitored, the chip is bypassed right at the VCC pin and I have verfied that there is no oscillation taking place with a TEK TDS-5104 oscilloscope set to the full 4 GHz bandwidth. I also tried a SN74AUP1G00 and found the same issue, on slow input transistions the current on it went up to around 260 microamperes just before it switched. You can easily replicate this by using a 10 uF capacitor from the gate input to ground and then putting a 1 MegOhm resistor from the "Y" output to the gate input forming a relaxation oscillator, monitor the current and it will be very high, not at all what was expected.
