We have a design with four ptd08a006w providing four different supplies. The four ptd08a006w share one inhibit line that is pulled-up via a 10kΩ resistor to 3.3V (generated by the controlling UCD). These are the only five connections to this line. (four modules, one pull-up).
One of our systems has developed a fault after nearly a month in operation in which the inhibit line goes to less than 1.2V, causing all four modules to shut-down. Under this condition I measure from 1.5kΩ to 700Ω to ground (variable depending on condition, see below, but independent of measurement polarity ). On the running system, the inhibit line remains at around 3.33V (depending on the UCD supply, of course; which in this system is around 3.38V).
The fault is intermittent, in the sense that when I probe the INH line with a multimeter that can push up to 1mA into it, the fault will go away (for a day or more) and the impedance to ground would raise to 14kΩ or so. This is a reliable way to get the modules to start again. Measure with the multimeter, wait for the impedance to raise above 10kΩ, then power the system up. A lower-current multimeter would not change the fault condition. We have thoroughly verified the connections, and ensured that there is no solder or flux residue present on them.
I cannot find any documentation regarding the circuitry behind the INH line, and I see no obvious reason for an intermittent fault that gets cleared by the simple fact of measuring it. We have no obvious way to isolate which one of the modules is at fault without removing each module individually. I could simply try to use a smaller value pull-up (e.g., 1kΩ) and hope that it works, but I would like to know what is the root cause of this problem.
