Gentlemen,=20 Thank you for your replies. My comments are as follows... Thu, 30 Mar 2017, David C Brown : > SPDT dip switches like this: > http://www.futureelectronics.com/en/technologies/electromechanical/switch= es/dip-switch/Pages/5846308-206-121ST.aspx?IM=3D0 > take up twice the space of a SPST dip switch. =20 > Or, using jumpers... That thing is huge, David! Consider well I have an 8-position DIP switch i= n one design, and in another design I have that same DIP with 8 switches an= d another DIP with 6 switches. Smaller is often better when designing for i= n-car use. High-value SMD resistors are tiny and would take less PCB space= than a much larger "cross-over" switch. - - - - - - - - - - - - - - - - - - - - - - - Thu, 30 Mar 2017, : > If you are really after low power then you don't want any pull-ups. Conceptually, I concur, Alan, but actual implementation is another matter. = Code becomes more complex and what if something happens and the PIC gets s= tuck and the pin is left to an Output? Also, I don't think I am following = your train of thought completely. (Sorry.) What are the long term implica= tions of that 5us "short circuit" in terms of product life and reliability?= What if I am connecting to a DIP switch which is either open-circuit (flo= ating) or GND? How can you READ a floating state? (You can't.) By the way, I need to poll DIP switches regularly, not just when the Igniti= on is ON. I don't know when someone will flip a switch. - - - - - - - - - - - - - - - - - - - - - - - Thu, 30 Mar 2017, stephen.forrest@agilent.com: > Make that pin an output and set it to '1' when > you want to read the switches. Set it to '0'=20 > otherwise - zero current now, regardless of=20 > the switch setting. You have maximum of 20 mA=20 > drive (depending on PIC) at 3.0 V so about=20 > 1.2 K ohm minimum per switch. Now you also=20 > need to take account of the time for the=20 > inputs to stabilize after turning on the=20 > pull-ups... Stephen, you are suggesting almost the same thing as Alan in that no extern= al pull-ups should be used at all, and software wizardry should solve the p= roblem. But more specifically, what you seem to be saying is that the PIC = "INPUTS" (which in my case are tied to DIP switches, the opposite side bein= g to Ground, or inputs tied to NPN or PNP circuits -- see my PNG schematic = below) be set to OUTPUTS (at voltage level LO), and then when I want to REA= D I would first need to ENABLE the PIC's INTERNAL WEAK PULL-UPS (which are = not available for all I/O pins), wait for things to stabilize, then READ, t= hen switch back to the Output (LO). That would add complexity to the code,= but I agree it seems reasonable to eliminate external pull-ups. =20 If I am not understand what you or Alan meant, my apologizes. I simply nee= d a bit more explanation from you. However, if this really is the best approach, why do most designs that I se= e still use external pull-ups? What are the caveats to the "no external pu= ll-ups at all, keep the PIC pin an Output most of the time" design? - - - - - - - - - - - - - - - - - - - - - - - Some of you commented on the Stack Exchange discussion I linked but there w= ere no specific or detailed comments on the schematic I provided for everyo= ne to view here (which is a circuit I've actually built and used in a car): https://cl.ly/jn0P/Image%202017-03-30%20at%2010.29.42%20AM.png (NOTE: Those two inputs are inverted. When a +12v signal appears at the +1= 2V input, the PIC input pin sees GND. And when a GND appears at the GND in= put, the PIC sees HI.) Specifically, how many of you have designed for automotive environments? A= nd in your experience when using a PIC, how are you handling the +12 and Gr= ound inputs? Is your analog circuit similar or the same as mine (see above= PNG)? If not, why not? (I'm simply curious.) Thank you, James Wages --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .