Maker Pro
Maker Pro

Impedance of digital inputs

P

Patrick

Jan 1, 1970
0
Would it be safe to say that ALL digital inputs (pins on
microprocessors, microcontrollers, gates, etc.) are always high input
impedance? i.e. When doing circuit analysis is it okay to model a
digital input as a very large (100K or higher) resistance to ground?
Thanks in advance...
 
K

Kevin Aylward

Jan 1, 1970
0
Patrick said:
Would it be safe to say that ALL digital inputs (pins on
microprocessors, microcontrollers, gates, etc.) are always high input
impedance? i.e. When doing circuit analysis is it okay to model a
digital input as a very large (100K or higher) resistance to ground?
Thanks in advance...

Essentially yes, if they are *CMOS*. They look like, say, 1p-10p
capacitance, with a very high dc impedance. For the most part they can
be considered open circuits for DC. A more accurate model would be to
include diodes from the inputs to the supply rails to model leakage.

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.
 
P

Patrick

Jan 1, 1970
0
Sounds good. What about digital inputs of other logic families, e.g.
TTL? If I remember correctly TTL inputs draw a few micro-amps of
current with a logic HIGH (5V) input.
 
J

John Fields

Jan 1, 1970
0
Would it be safe to say that ALL digital inputs (pins on
microprocessors, microcontrollers, gates, etc.) are always high input
impedance?
 
J

John Fields

Jan 1, 1970
0
err....do people still use TTL?

---
Do it matter?

His question was:

"Would it be safe to say that ALL digital inputs (pins on
microprocessors, microcontrollers, gates, etc.) are always high
input impedance?"

and the answer is clearly "no".
 
P

Pooh Bear

Jan 1, 1970
0
Patrick said:
Sounds good. What about digital inputs of other logic families, e.g.
TTL? If I remember correctly TTL inputs draw a few micro-amps of
current with a logic HIGH (5V) input.

They draw a hell of a lot more when LOW.


Graham
 
P

Pooh Bear

Jan 1, 1970
0
Rich said:
Actually, they're _sourcing_ current when low. ;-) ;-)

Cheers!
Rich

Yeah, yeah, I know ! Didn't want to spoil the flow of it though.

Graham
 
K

Kevin Aylward

Jan 1, 1970
0
Well, actually, no they are not. There are sinking electrons *into* the
input!


Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.
 
J

John Fields

Jan 1, 1970
0
Well, actually, no they are not. There are sinking electrons *into* the
input!

They're sourcing _conventional_ current, Kevin.
 
R

Rich, Under the Affluence

Jan 1, 1970
0
They're sourcing _conventional_ current, Kevin.

The way kewl thing about this is, it doesn't matter! To switch from
conventional to electron flow, just change all the pluses to minuses,
and vice versa. ;-)

The thing that really weirds me out [but The Philosophizer loves it]
is that, in all of the equations that relate electricity and
magnetism, you can swap those two, and the answers still come out
right!

I'm working on a 3-D diagram of the two 7-dimensional hypertoruses
that the Universe is derived from...

Cheers!
Rich
 
J

John Fields

Jan 1, 1970
0
They can't. Conventional current is imaginary...

---
So's the square root of minus one, but that doesn't stop us from
using it.

Don't be an ass, Kevin. We all know that and we all use
conventional current every day.
 
P

PeteS

Jan 1, 1970
0
On the OP question, I just had an interesting experience with a Xilinx
part. In *unprogrammed* mode, the bus hold circuitry is active in
series with an equivalent resistor of about 25k to the I/O pin.

I caught that just as I was connecting things up to it (because I
wanted to leave it unprogrammed for initial testing of other parts of
the board) and had to whip up a simple piece of verilog to force inputs
to *be* inputs and outputs to be at the suitable level. It didn't do
anything else, so I simply had a bunch of assignments.

The old rule of thumb that FPGA IO pins are inputs when unprogrammed
doesn't always hold ;)

Cheers

PeteS
 
J

Jasen Betts

Jan 1, 1970
0
The way kewl thing about this is, it doesn't matter! To switch from
conventional to electron flow, just change all the pluses to minuses,
and vice versa. ;-)
The thing that really weirds me out [but The Philosophizer loves it]
is that, in all of the equations that relate electricity and
magnetism, you can swap those two, and the answers still come out
right!

One exception is the Hall effect.

that one depends on the sign of the charge on the charge carriers and has
been measured to have the opposite sign in P-type semiconductors than it
does in regular metals and N-type semiconductors.
I'm working on a 3-D diagram of the two 7-dimensional hypertoruses
that the Universe is derived from...

huh?

Bye.
Jasen
 
Top