Re: OT - TTL Circuit: Pullup vs Supply

Oh, what the hell, I might as well put in my two cents worth.

It's probably mostly to limit breakdown current in case of a power
supply surge. In most TTL, inputs have a clamp diode to ground and the
input is actually a transistor emitter (especially in gates---something
like a 7430 has ONE input transistor with 8 emitters going to the 8
input pin). If the emitter swings around 6 volts positive to the base
you'll avalanche the emitter/base junction which will degrade the
transistor (at the very least) or destroy it if enough current flows.

Now when everything gets to humping you'll have inputs going below
ground due to board trace inductance. This is limited somewhat by the
input clamp diode, but will still make the other inputs appear more
positive than they are internally pulling the base reference lower.

Add to this that ground and power aren't the same across the board. You
could have an IC input being fed by another IC with a lower ground
voltage (causing the input to go negative) while itself having a higher
"+5" than the chip driving it. Next thing you know you start blowing
inputs.

So, to be safe, they use pull-up resistors on unused pins OR connect
them to other inputs OR (if you have an unused gate somewhere) to an
output pin that will stay at a "1". An example of the this is to use an
extra inverter, ground its input, and use the output to provide a "1"
for a whole mess of inputs. This "1" won't be anywhere near +5 volts on
old-style TTL, but will still be a perfectly good "1".

(Well, THAT should confuse everyone---hopefully it will be helpful.)

Jess Askey wrote:
> can anyone explain what any pro's or con's there might be on tying
> static input pins on a TTL device 'high' through a pull-up resistor vs
> straight to the +5V supply? Im working on laying out a ASIC replacement
> PCB and I had the board designed with pull-up resistors but tracing them
> is becoming a PITA so I thought about just tying them to the +5V rail as
> an alternative.
> ---------------------------------------------------------------------------
> ** Unsubscribe, subscribe, or view the archives at
> http://www.vectorlist.org
> ** Please direct other questions, comments, or problems to
> chris@westnet.com
>

---------------------------------------------------------------------------
** Unsubscribe, subscribe, or view the archives at http://www.vectorlist.org
** Please direct other questions, comments, or problems to chris@westnet.com
Received on Thu Sep 7 04:45:06 2006