Re: Serial Port voltage drops with Linux

From: Jens Benecke (jens@pinguin.conetix.de)
Date: Tue Jan 18 2000 - 04:35:47 EST


On Tue, Jan 18, 2000 at 01:09:58AM +0000, Thorsten Kranzkowski wrote:
> On Mon, Jan 17, 2000 at 03:07:35PM +0100, Jens Benecke wrote:
> > The situation is this: When we connect the hardware to the serial port,
> > the signal amplitude breaks down to lower than +3V (which is miniumum
> > for the PC to detect a "0"). With DOS, the signal amplitude varies from
> > -10..+10 V (approx) for 1/0 signals. With Linux, it's -10..<+3V, and
> > that is not enough.
> Umm - Linux doesn't adjust Power supply voltage levels and it doesn't
> swap new serial line driver IC's in place either :-))

I almost thought so. :)
 
> You have made __double__ shure you didn't connect two output lines
> together (eg. TxD from computer to TxD from your device)? And you also
> checked your GND connection?

That's the problem. Our hardware (I didn't design it..) uses the RTS, DTR
and CTS pins for communication and not the [TR]xD, because we didn't plan
to put a RS232 protocol chip on our hardware (too little data). I regret
this descision now - but it cannot be changed (deadline = Feb 1.)...

Details (documentation draft) in German are available at
http://terra.et7.tu-harburg.de/~prakt/bla/ .
 
> > Linux? In theory (yeah, right) it's hardware that should behave the
> > same under any OS, at least in respect to voltage levels and power
> Yes, they _do_ behave equally :)

That's nice to hear. Then I'd like to know what makes the difference. You
are invited to look at it yourself here, we'd even give you a free lunch
:)
 
> > BTW, we tried three different serial port hardware boards (ISA) so the
> > hardware should not be at fault.
> ...and have booted DOS on the very same computer with the very same
> ISA-boards and your device?

Exactly. Strangely, the port status LEDs (we put a serial test circuit on
the port for checking - basically 25 red/green LEDs) were in "Linux mode"
on bootup, and changed only to "QBasic mode" when the QB app was executed
first time. Then, reboot to Linux, they changed back to "Linux mode" when
any application tried to access any serial port - e.g. gpm, setserial, or
our application. And stayed there.
 
> How did you build the power supply of your device? An external
> AC-powersupply ('Netzteil') or from some unused 'to be set to 12V' status

IIRC, we tried both, but I'll CC this to Guido (who actually did all the
testing and programming) and might be able to explain more details.

> lines? If the latter, are they set to 'high level' when you connect your
> device?

I'll double-check this.
 
> > Thanks a lot!
> Keine Ursache! Thorsten

This kind of support is what I like with Linux. :) The alternative would
have been MS Virual C++ or something, so we'd all be happy to get this
running under Linux.
 

-- 
_ciao, Jens_______________________________ http://www.pinguin.conetix.de
·
"[Microsoft] ... guarantees 99.8% NT uptime for certain hard-/software.
 That's exactly the 3 minutes daily that my NT server needs to reboot."
							-- ZDnet editorial

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Sun Jan 23 2000 - 21:00:17 EST