Robert Nansel wrote: (I sent a couple of messages regarding the origins of the 5V level to the list a few weeks ago, but I don't have a definitive historical answer as to such.) On the technical side though, the supply level is going to effect power consumption, hence operating temperature range, drive and loading factors, propagation speed due to stored charge (RC factors), switching thresholds/limits due to varying current thru an R into a transistor base, etc. At the electronic level, there are a lot of analog issues in digital design. Then of course there are junction breakdown voltages, some literature I have here suggests there isn't really a lot of spare room outside the specified operating voltages. Then there are edge-triggerred flip-flops which rely on the stored charge in some of the transistors for the edge-triggerring and which I suspect will start doing funny things outside the specs, again due to RC factors. (As mentioned in an earlier message) I don't know whether the 5V was an initial, somewhat-arbitrary design parameter from which other design factors followed, or if the opposite: a design factor which followed from other initial parameters. One guess is it was a minimum chosen to keep junction breakdown requirements and power dissipation low, while providing enough headroom for switching currents and adequate noise margins. Either way, once it has been set in the design, the above-mentioned device specs and characterisations are going to be determined and specified relative to the preferred supply level. Note that manuf. specs typically provide both absolute maximums (over which the device may be damaged), as well as operating max/min within which the other specs are guaranteed. The specs typically don't say that it won't actually work outside the operating range Fairchild databooks generally have quite a treatise about the functioning of the internals of the IC families at the beginning of the book which can make for interesting reading, TI books less so, in Nat Semi books it seems to be non-existent. A year or two ago there was a discussion on the list where we were testing a variety of 74x175's (albeit configured in an odd manner) for some PDP-8 front panel and were observing changing and idiosyncratic behaviour over a very narrow operating voltage range. I hope you're not looking to mimic TTL - you'll have trouble sourcing multi-emitter transistors in discrete form. (I know - it's outside your design constraint anyways.) More seriously, looking to TTL or standard DTL ICs for design would kind of chew up the transistor count very quickly. Is there a reason you're not looking to period/original discrete-component logic designs?