Admittedly, few PCB's have this, but given a proper ground and power plane, all IC's operating from Vcc ise the same Vcc.

The MAX1232, if that happens to be your supervisor, trips at somewhere between 4.5 and 4.74 volts, typically 4.62 volts. The DS1230Y BBRAM, OTOH, uses between 4.25 volts and 4.5 volts, typically 4.37 volts as its trip-point for write-lock. The DS1230AB has the same trip point specification as the MAX1232.

However, the DS89C4x0, just as one example, detects brownout or power-failure at between 4.2 and 4.6 volts, typically 4.375 volts.

If you opt to "roll-yer-own" BBRAM subsystem, you have an even more complex problem to solve, simply in selecting a voltage to trigger the write-lock. Enforcing it is another thorny issue.

Since all these components operate "between" two trip voltages and "typically" at a specific one predicting what will happen is not so simple.

My own very preliminary and purely anecdotal observations suggest that there's plenty of room for foul-ups between the various trip voltages, regardless of what they are, and those observations, however weak they were, were made with a proper ground and power plane, and very low power-gnd noise, to ensure that all the components had the same Vcc.

BTW, the Dallas BBRAMs simply write-inhibit the part, so, if the firmware that does the writes is smart enough to verify that its writing has been properly recorded by reading it, it will behave appropriately when it has not. Eventually, the MCU's power-fail interrupt will take over.

RE