I've searched the board, and I think I've found the info I need, but if someone could reality-check my idea, I'd be very appreciative.
I have a Garmin 60CSx. It seems that Garmins talk RS232, but don't spit out the full 12V, more like 2.5 or so. I also have a 74LS00 chip.
Could I could use the 7400 to invert, and bring to a solid 5V, the signal from the GPS, without going and getting a MAX232 or other level-shifter? Also, if the Garmin really does send out 12V, this cheap chip will bear the brunt and protect the Arduino, won't it?
I suspect that voltage is more than 2.5V. A DVM isn't a very good tool for measuring voltages on digital data lines, as the DVM needs some time for a stable measurement. If the RS232 signal was quiet you could get a good measurement, but the nature of a GPS (in NMEA mode, anyway) is to spit out whatever sentences it is configured to transmit once per second, so the TX line is pretty active.
I don't think the LS series parts like more than 5V, so true RS232C voltages would probably kill it (eventually, if not immediately). I don't know what the failure mode of the 74LS00 is; I would be very reluctant to use it to "protect" something else, as the failure mode could be "pass the too-high voltage to the output pin", thereby killing the downstream part.
In addition, the 74LS series don't like static, and if you actually use the circuit very much it may die an early death due to ESD.
I agree, best to use a MAX232 or something similar. The Garmin might be sending +5 to -5V, which won't do the 74LS00 any good. If you have a 'scope available, look at the serial line with that!