From: Spiro Trikaliotis (trik-news_at_gmx.de)
Date: 2003-10-13 20:48:25
Hello Panayiotis,
since you asked me how to subscribe to cbm-hackers, I'm forwarding this to
the list with your full name.
BTW: You can subscribe to cbm-hackers at cbm-hackers-request@cling.gu.se and
the word "subscribe" in the Subject. AFAIk, there is even a archive, but I
don't know exactly where it is. From my experience, I'm sure the people
there will do everything they can do to help you. :-)
There were even more answers, but I have not had the time to forward them.
I'm sure anyone will give me the archive address so you can look for it
yourself.
On Fri, Oct 10, 2003 at 04:16:52PM +0300, Panayiotis Michalopoulos wrote:
> Please thank for me all the people that responded and kindly forward to them
> my remarks.
> Unfortunately, I'm afraid they haven't completely understood the problem.
>
>
> > Marko Mäkelä <marko.makela@hut.fi> wrote on Wed, 8 Oct 2003 22:54:41 +0300:
> >
> > > Hmm, aren't the voltages and currents defined in IEEE 1284? Contact me
> > > privately if you have trouble locating that standard; I should be able
> > > to access it via http://ieeexplore.ieee.org from the university network.
>
> I already have the standard. It says that for Level I the sourcing and
> sinking currents are 0.32mA (too low to use a resistor in series) and 14mA
> (good enough) respectively. For Level II both currents are 14mA. The point
> is now, how can I tell which level (if any) my port is build for? There is
> nothing in the modo's manual and the manufacturer's customer service never
> replied. In addition to that, I have not been able to find any modo
> manufacturer providing such info.
> > Groepaz <groepaz@gmx.net> wrote on Wed, 8 Oct 2003 23:20:01 +0200:
> >
> > > i would advice your friend to use some simple ttl line drivers (theres
> > > some chip with 8 drivers, dunno which... any other 74xxx gates will
> > > do aswell if you know what you are doing :)). the line drivers will not
> > > only protect your lpt> port, it will also result in a better signal
> > > quality, especially if what you connect
> > > needs (for todays measurements) a lot of current.
>
> Of course I am using drivers. The problem is that I need to use
> bidirectional data transfer. For this reason, half my drivers have their
> outputs connected to the parallel port (for peripheral to PC data
> transfers). The dangerous case is when both the peripheral and the PC are
> set as outputs at the same time. If the PC and the peripheral have opposite
> logic values on a certain pin, one totem pole output will act as a short
> circuit on the other. This will damage either or both of them.
>
> > > apart from that using 8bit epp style i/o with lpt is a bad idea IMHO,
> > > unless you dont care much about if your program will run stable anywhere,
> > > > anytime :)
>
> Why not? As far as I can tell there is no stability problem.
> > Lee Davison <Lee.Davison@merlincommunications.com> wrote on Wed, 8 Oct
> > 2003 23:26:47:
> >
> > > There is, simply connect a resistor in series with each bi-directional
> > > line
> > > you wish to protect. A value of 180 to 220 ohms should work.
>
> A 20 Ohm resistor would allow a short circuit current of approx 18mA. I
> think that this would be too high. I have already tried with 6K8 resistors
> that allow a short circuit current of less than 1mA, but the logic-0 voltage
> becomes higher than the TTL standard of 0V8.
>
> > > There is more than one "maximum current" limit. The lowest one is
> > > usually the maximum that can be sourced or sunk and the output still
> > > be within the defined voltage limits for the logic levels for the
> > > device. This is usually only a few mA at best.
>
> I can deal with small voltage changes as long as the 2V-0V8 TTL standard is
> maintained.
> > Gianmario Scotti <Gianmario.Scotti@nokia.com> wrote on Thu, 9 Oct 2003
> > 05:47:59:
> >
> > > Well, the solution is simple: drive the port lines with a source that
> > > can do
> > > three-state.
> > > Have it go hihg-impedance while turned on or reset, and let it go out of
> > > high-impedance only when the parallel port is working as
> > > input. For example, 74HC541 or 74HC540, depending on whether you need
> > > inverting or non-inverting outs.
> > > So I'd probably try to monitor STROBE/, put drivers in high-impedance
> > > when active low. Check out the pins at +http://www.ctips.com/spp.html
> > > I'm sure people can suggest other pins to monitor, but the principle is
> > > the same: when the host (the PC) says "i'm output", put the drivers in hi-Z.
> > > I leave the details of the implementation to you.
>
> I have already done that of course. However, there are 2 problems:
> 1. When I connect my peripheral to the PC and its software is not
> running yet, there is always a chance both the PC and the peripheral to be
> set as outputs, thus damaging one of them.
>
> 2. If the peripheral is connected and some other program sends data to
> the parallel port (i.e. a wrong print command), then damage can also occur.
Spiro
Message was sent through the cbm-hackers mailing list
Archive generated by hypermail pre-2.1.8.