🚀 go-pugleaf

> I know pullups are required for CMOS inputs if the inputs can go into
> high-impedence.  Does the same hold true for TTL?  What about HCT?
> 

Gee, Jeremy, third question today! :-)

I think you are a bit confused - INPUTS do not "go into high impedance,"
outputs do - specifically, such things as "tristatable" TTL outputs.
(CMOS inputs ARE high impedance, and you sometimes see resistors on CMOS
inputs connected to TTL outputs, but that's because CMOS levels for "1" and
"0" are "farther apart" than the typical TTL output, so pull-ups and pull-downs
are needed to make sure the CMOS device sees the right level.)

Pull-ups or pull-downs are needed on any tri-state output when you want to
make certain that the inputs on that line will definitely see a valid state
when all drivers (outputs) are tristated - if such resistors aren't present,
the line will "float" at an intermediate level.  A TTL input will *usually*
read a "floating" line as a "high" (logical 1), but this ISN'T guaranteed.
(On the other hand, ECL inputs ARE guaranteed to be pulled LOW internally,
so be careful about making assumptions about undriven lines!)

Putting pull-ups and pull-downs on inputs is always a good idea if there is
any time when the line can be put into a state such that these inputs aren't
being actively driven.  Choose the size of the resistor according to how
many inputs are on that line (there's no need for a resistor at each input!),
and what the drive requirements are for inputs in that technology.  Pull-DOWNS
are sometimes also used in the case of an input that needs to be tied low,
but which may also need to be overridden (driven to the high state) for test
purposes.


Bob Myers  KC0EW   Hewlett-Packard Co.      |Opinions expressed here are not
                   User Interface Tech. Div.|those of my employer or any other
myers@fc.hp.com    Fort Collins, Colorado   |sentient life-form on this planet.