Parallel port with open collector ouputs....

[oparr@hotmail.com]

How prevalent is open collector outputs (with no internal pull-up
resistors) on the data pins?

 

[Sjouke Burry]

oparr@hotmail.com wrote:

How prevalent is open collector outputs (with no internal pull-up
resistors) on the data pins?
Have not seen it anywhere on the port connector pins..

But I have seen schematics with the chip being open collector,
and a set of pull up resistors attached.

So in a custom made port you could have open collector pins.

A ttl device attached would not mind, but a cmos one would fail.

 

[oparr@hotmail.com]

Have not seen it anywhere on the port connector pins..

"Normally the Printer Card will have internal pull-up resistors, but
as you would expect, not all will. Some may just have open collector
outputs, while others may even have normal totem pole outputs. In
order to make your device work correctly on as many Printer Ports as
possible, you can use an external resistor as well."

Read the above in a PP Interfacing doc. I want to apply some "risk
management" and skimp on external pull-up resistors.

On Dec 12, 2:17 pm, Sjouke Burry <burrynulnulf...@ppllaanneett.nnll>
wrote:
>

 

[www.interfacebus.com]

On Dec 12, 1:18 pm, "op...@hotmail.com" <op...@hotmail.com> wrote:

How prevalent is open collector outputs (with no internal pull-up
resistors) on the data pins?

The Centronics Printer Port and the IEEE-1284 interface that replaced
it both used open collector ICs at the source [driver]. The Pull-up
resistor is expected to be at the other end of the 'cable'. So if left
unconnected you have an open collector that's floating [not pulled
up].

http://www.interfacebus.com/Design_Connector_1284.html

 

[Jasen Betts]

On 2008-12-12, oparr@hotmail.com <oparr@hotmail.com> wrote:

How prevalent is open collector outputs (with no internal pull-up
resistors) on the data pins?

dunno. I thought most had 10K or so pull up.

 

[oparr@hotmail.com]

The Centronics Printer Port and the IEEE-1284 interface that
replaced it both used open collector ICs at the source

Yes, I read that too. Long story short.....If you were designing a
device that had to interface with the PP, would you go to the extra
expense of including pull-up resistors on each data line, bearing in
mind that only very few OC variants exist these days? And if you do,
what pull-up value would you choose, bearing in mind that it would be
in parallel with the value chosen in most PPs out there today and it
would be on its own with OC variants? 10K? 4.7K?

The best I've seen in terms of advice is not to go below 4.7K but if
4.7K **or lower** is already being used in the device....Hmmmm. You
see what I'm getting at....Leaving well enough alone with no extra
pull-ups added and not catering to the few OC variants may be the
better waty to go.

BTW, just about every PP link can be found on this page (look for PP
section);

http://www.epanorama.net/links/pc_interface.html#parallel

On Dec 12, 11:57 pm, "www.interfacebus.com"
<Interfacebus.Engin...@gmail.com> wrote:
>

 

[oparr@hotmail.com]

I thought most had 10K or so pull up

Most do hence the dilemma. Should one go to the extra expense (and
possible risk) catering to the few that don't.


On Dec 13, 6:23 am, Jasen Betts <ja...@xnet.co.nz> wrote:
>

 

[www.interfacebus.com]

On Dec 13, 9:43 am, "op...@hotmail.com" <op...@hotmail.com> wrote:

The Centronics Printer Port and the IEEE-1284 interface that
replaced it both used open collector ICs at the source

Yes, I read that too. Long story short.....If you were designing a
device that had to interface with the PP, would you go to the extra
expense of including pull-up resistors on each data line, bearing in
mind that only very few OC variants exist these days? And if you do,
what pull-up value would you choose, bearing in mind that it would be
in parallel with the value chosen in most PPs out there today and it
would be on its own with OC variants? 10K? 4.7K?

The best I've seen in terms of advice is not to go below 4.7K but if
4.7K **or lower** is already being used in the device....Hmmmm. You
see what I'm getting at....Leaving well enough alone with no extra
pull-ups added and not catering to the few OC variants may be the
better waty to go.

BTW, just about every PP link can be found on this page (look for PP
section);

http://www.epanorama.net/links/pc_interface.html#parallel

On Dec 12, 11:57 pm, "www.interfacebus.com"



Interfacebus.Engin...@gmail.com> wrote:- Hide quoted text -

- Show quoted text -

It's more than just a common value being used. This is the equation to
calculate pull-up values.
http://www.interfacebus.com/IC_Output_Input_Pullup_Resistor_Values.html

 

[oparr@hotmail.com]

It's more than just a common value being used. This is the equation
to calculate pull-up values.

You don't know what is used in the PP circuit. You will be using
external pull-up resistors in most cases where they are not even
required.

On Dec 13, 3:23 pm, "www.interfacebus.com"
<Interfacebus.Engin...@gmail.com> wrote:
>

 

[Jasen Betts]

On 2008-12-13, oparr@hotmail.com <oparr@hotmail.com> wrote:

I thought most had 10K or so pull up

Most do hence the dilemma. Should one go to the extra expense (and
possible risk) catering to the few that don't.

risk? use a part with internal pullup

 

[www.interfacebus.com]

On Dec 13, 6:31 pm, "op...@hotmail.com" <op...@hotmail.com> wrote:

It's more than just a common value being used. This is the equation
to calculate pull-up values.

You don't know what is used in the PP circuit. You will be using
external pull-up resistors in most cases where they are not even
required.

On Dec 13, 3:23 pm, "www.interfacebus.com"



Interfacebus.Engin...@gmail.com> wrote:- Hide quoted text -

- Show quoted text -

True, that was a poorly designed standard [or no standard]. So I would
stay with the higher values of pull-up resistors. But the equation to
calculate the value fits regardless.
http://www.interfacebus.com/Design_Centronics_Connector_PinOuts.html