That is basically it, educational, since till you make something that works, you really dont know what is going on in the system. Which is why I am prosuing a 65C816 computer build as time becomes avalible.

Dimitri

http://www.youtube.com/watch?v=GcDshWmhF4A&feature=fvw
http://www.youtube.com/watch?v=SYi9sJkS19Q

Yes they did, but retained the P&B name.


The R10 line was (still is) very versatile. At one time, the series included time delay relays, remanent magnetic latching relays and units with built-in LED indicators. Most of that stuff has disappeared, but the basic R10 remains the same. At one time, the standard R10s could be had with up to 8 form-C contacts. The most I ever used in any design was 6 form-C. BTW, Omron makes a 4 form-C PCB mount relay in their G6A-434P series. I use it in the automatic block signal system track circuit detectors I build for large-scale model railroad use.

These days, if I need a time delay function, I build it from a standard relay and a comparator time delay circuit (very stable, even with unregulated power sources). I still use my relay flip-flop circuit on occasion, most notably in air compressor control units where two compressors are involved. The flop alternates between which compressor runs first so as to equalize wear on the machinery. I tend to avoid solid state controls in such applications, due to the high level of electrical noise that is present.

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!

These videos, which I'd seen before, inspired the idea. However, imagine a 16-by-16 matrix of such levers comprising a 32 byte RAM module and electrically actuated.

The little see-saws are interesting.

Good to know about the relays, was not sure if they were really the same R10 relays or if Tyco was just using the same model designation.

Been bouncing back and forth all day on how to make switches with floating gates, and what kind of set up would be required. A couple of transistors and a capacitor to keep a charge on a transistor.

Something similar, however when it wants to re-write the data in a field, it will have to first write all 0's then it will be able to rewrite the data.



Saves 2 transistors, but adds a capacitor, and a write delay as the processor can't write data in one cycle to allow the capacitor to drain, however with a small enough capacitor, depending on how much energy drain per read there is, I may be able to clear the data set quickly enough to do it all in one cycle.

Mind you if the memresistor stuff came to market in a package similar to a transistor, that would be neat to use as memory units. Just because its leading edge right now and its doubtful anyone but silicon fab houses will actually get to "play" with it on the component level.

Dimitri