Searching \ for '[OT] PIC and modem problems' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/devices.htm?key=pic
Search entire site for: 'PIC and modem problems'.

Exact match. Not showing close matches.
PICList Thread
'[OT] PIC and modem problems'
2000\05\18@042737 by Steinar Olsen

flavicon
face
Hi,

I use a PIC to communicate with a standard Hayes compatible modem on 9600
baud RS-232.
The main issue is to call up the system with another modem, get the
remote modem to pick up the call automaticly, and communicate via RS-232 to
the uProc.

The system works fine, but there is one problem.
During the time the two modems are communicating after picking up the call,
to set speeds and other initialising before starting the main data
comunication,
it will hang up immidiatly if it gets a input on the RS-232 port.

This is a problem on the uProc side
beacuse the system outputs regulary data on the RS-232, and if this
output comes before the modem is finished seting up (Online indicator lights)
the modem will hang up.

Yes, there is a number of "no good" sollution to this, like
stop the dataoutput when reading "RING" at the RS-232, or use the
"Call indicator" and "online output" to disconnect RS-232 during
init.. But there must be another way to prevent this from happening.
Is there something I overlook in the AT command set?
Why is this happening, is there a meaning behind this function?
(And no, my datapakages does not contain ATH in the start. ;-) Just numbers
and I have tried with different outputs as well)

Help??
Anybody??
:-)













Steinar Olsen
Electronics Engineer
________________________________________________________________
Norsk Elektro Optikk A/S          Work:   spam_OUTSteinarTakeThisOuTspamneo.no
                                 Home:   .....StolsenKILLspamspam@spam@c2i.net
Solheimsveien 62A                 Web:    http://www.neo.no
P.O. Box 384                      Direct: +47 67 91 11 58
N-1471 SKAARER                    Phone:  +47 67 97 47 00
NORWAY                            Fax:    +47 67 97 49 00
================================================================

2000\05\18@050555 by Michael Rigby-Jones

flavicon
face
{Quote hidden}

Why is this a "no good" solution?

{Quote hidden}

So if I understand correctly, you are sending data to the modem all the
time, even if it isn't "on-line".  IMHO this isn't a good idea.  If your
data ever included valid AT commands (whether you put them there on not) you
could seriously screw up the settings on this modem.  You really should be
checking at least the Carrier Detect line and maybe the CTS line.

Mike












{Quote hidden}

2000\05\18@050803 by Alan B Pearce

face picon face
>This is a problem on the uProc side
>beacuse the system outputs regulary data on the RS-232, and if this
>output comes before the modem is finished seting up (Online indicator lights)
>the modem will hang up.

This appears to be a common problem with Hayes compatible modems. I had occasion
to get involved with trying to connect a modem to autoanswer when connected to a
machine which was trying to poll terminals continuously. The polling from the
host machine stopped the modem from answering the dial up call. The solution was
to make a little adapter box to go between the modem and controller, consisting
f a 1489 RS232 receiver, a 1488 RS232 transmitter, and a 78L05 to supply the
1489. The host controller had +/-12V available, and between the two chips, there
were enough gates available to gate the transmit line off until the carrier
detect from the modem went true. It all got built into one of those gadget
adapter boxes with db25 plugs on each side.

2000\05\18@065637 by Steinar Olsen

flavicon
face
This should not be a problem unless some of
the data containes the excact string "AT" followed by data, and
none of my data consist of such a number combination.

So the question remains.. Why does it hang up during incoming call
when it's RS-232 input is nowhere near an AT command?
Is that the problem that it is not an AT command, in that case, why
hang up, instead of ignoring the input as usual?

>>
>So if I understand correctly, you are sending data to the modem all the
>time, even if it isn't "on-line".  IMHO this isn't a good idea.  If your
>data ever included valid AT commands (whether you put them there on not) you
>could seriously screw up the settings on this modem.  You really should be
>checking at least the Carrier Detect line and maybe the CTS line.
>
>Mike


Steinar Olsen
Electronics Engineer
________________________________________________________________
Norsk Elektro Optikk A/S          Work:   @spam@SteinarKILLspamspamneo.no
                                 Home:   KILLspamStolsenKILLspamspamc2i.net
Solheimsveien 62A                 Web:    http://www.neo.no
P.O. Box 384                      Direct: +47 67 91 11 58
N-1471 SKAARER                    Phone:  +47 67 97 47 00
NORWAY                            Fax:    +47 67 97 49 00
================================================================

2000\05\18@070258 by Alan B Pearce

face picon face
>So the question remains.. Why does it hang up during incoming call
>when it's RS-232 input is nowhere near an AT command?

The problem is not that there is no "AT" command. The problem is that there is
activity on the RS232 line, so the internal micro tries to parse this activity
looking for an "AT" command instead of doing the call connect function when the
phone rings.

2000\05\18@070515 by Steinar Olsen

flavicon
face
Yupp, that was the sollution I went for as well.
But I had hoped to get rid of that "in the between" HW..

I'm still courious of WHY the modem does'nt ignore the
data input during incomming calls (as long as it is no AT command) instead of
hanging up.


>This appears to be a common problem with Hayes compatible modems. I had
occasion
>to get involved with trying to connect a modem to autoanswer when
connected to a
>machine which was trying to poll terminals continuously. The polling from the
>host machine stopped the modem from answering the dial up call. The
solution was
>to make a little adapter box to go between the modem and controller,
consisting
>f a 1489 RS232 receiver, a 1488 RS232 transmitter, and a 78L05 to supply the
>1489. The host controller had +/-12V available, and between the two chips,
there
>were enough gates available to gate the transmit line off until the carrier
>detect from the modem went true. It all got built into one of those gadget
>adapter boxes with db25 plugs on each side.
>
>

Steinar Olsen
Electronics Engineer
________________________________________________________________
Norsk Elektro Optikk A/S          Work:   RemoveMESteinarTakeThisOuTspamneo.no
                                 Home:   spamBeGoneStolsenspamBeGonespamc2i.net
Solheimsveien 62A                 Web:    http://www.neo.no
P.O. Box 384                      Direct: +47 67 91 11 58
N-1471 SKAARER                    Phone:  +47 67 97 47 00
NORWAY                            Fax:    +47 67 97 49 00
================================================================

2000\05\18@074121 by Steinar Olsen

flavicon
face
So that is what acctually happens...
Could'nt they have made it so it could manage
both tings at a same time.. *grrrr* :-)

Well well.. Many thnx for all help.
Steinar


>The problem is not that there is no "AT" command. The problem is that
there is
>activity on the RS232 line, so the internal micro tries to parse this
activity
>looking for an "AT" command instead of doing the call connect function
when the
>phone rings.
>
>

Steinar Olsen
Electronics Engineer
________________________________________________________________
Norsk Elektro Optikk A/S          Work:   TakeThisOuTSteinarEraseMEspamspam_OUTneo.no
                                 Home:   RemoveMEStolsenspamTakeThisOuTc2i.net
Solheimsveien 62A                 Web:    http://www.neo.no
P.O. Box 384                      Direct: +47 67 91 11 58
N-1471 SKAARER                    Phone:  +47 67 97 47 00
NORWAY                            Fax:    +47 67 97 49 00
================================================================

2000\05\18@095440 by Dale Botkin

flavicon
face
On Thu, 18 May 2000, Steinar Olsen wrote:

> So the question remains.. Why does it hang up during incoming call
> when it's RS-232 input is nowhere near an AT command?
> Is that the problem that it is not an AT command, in that case, why
> hang up, instead of ignoring the input as usual?

I remember from the modem command specs that I have looked at that the
modem is supposed to disconnect immediately if it sees serial data any
time after the dial command and before CD goes active.  It's an abort
feature, and I don't think there is a way around it.  I'd just have the
micro look at carrier detect from the modem, I think it's your only
reliable solution.  You could, though, probably figure out a way to have
the RS232 interface do it for you - use the CD signal to gate data to the
modem.  Whoops -- that might make it difficult to send the AT command.
There's probably a way to do it and I just haven't had enough coffee to
see it.

Dale
---
The most exciting phrase to hear in science, the one that heralds new
discoveries, is not "Eureka!" (I found it!) but "That's funny ..."
               -- Isaac Asimov

2000\05\18@102455 by Harold Hallikainen

picon face
       I have also found this to be the case. If you send data during
negotiation, the modem hangs up. I've had systems set up to echo data
back to the user at the other end of the modem link. Modem status
messages during connect would hang up the modem. So, in general, don't
send data unless DCD is true OR you are sending commands (like ATDT...).

Harold

On Thu, 18 May 2000 10:27:18 +0200 Steinar Olsen <steinarEraseMEspam.....NEO.NO> writes:
{Quote hidden}

________________________________________________________________
YOU'RE PAYING TOO MUCH FOR THE INTERNET!
Juno now offers FREE Internet Access!
Try it today - there's no risk!  For your FREE software, visit:
dl.http://www.juno.com/get/tagj.

2000\05\18@153442 by Harold Hallikainen

picon face
On Thu, 18 May 2000 13:40:35 +0200 Steinar Olsen <RemoveMEsteinarspam_OUTspamKILLspamNEO.NO> writes:
> So that is what acctually happens...
> Could'nt they have made it so it could manage
> both tings at a same time.. *grrrr* :-)
>
>

       This reminds me of early parallel port dot matrix printers that had one
line of buffer. They could either receive data or print data. This got
fun when you added their serial port (again, with no buffer). You had to
watch the CTS line so you wouldn't send data when it was printing (or
leave a long pause after every CR).

Harold

________________________________________________________________
YOU'RE PAYING TOO MUCH FOR THE INTERNET!
Juno now offers FREE Internet Access!
Try it today - there's no risk!  For your FREE software, visit:
dl.http://www.juno.com/get/tagj.

2000\05\18@174829 by Plunkett, Dennis

flavicon
face
19/5/2000


This is correct while operating V22bis and some other command sets, it is
not part of the Hays command response, as this only defines the "AT" set,
and not the physical data link.

The modem will monitor and watch the data lines while attempting to train
up, if it sees the ESC ESC ESC sequence it will stop!

What is the message that the modem sends back during this phase? As this
will give you a hit, do you see the RING message come in?
I know that this sounds dumb, but do the modems connect when the PIC is not
connected? Repeat the test with a PC instead of the PIC and see what occurs.

As for the control lines it is possible to dissable these in the modem, as
it only gives CD to indicate that there is valid carrier. The RTS and CTS
lines in the newer modems are a direct control of the state of the input
buffers


Dennis



> {Original Message removed}

2000\05\18@220853 by W. Sierke

picon face
It has always been my experience that a modem will abort making a
connection, from the time the ATD command is issued to the modem, until the
'CONNECT' message is issued by the modem (CD signal raised), if any
character is issued to the modem. I've never tested it to the extent of any
other type of transmission (BREAK, mis-framed characters, invalid start
bits, etc). Neither have I investigated whether in fact any commands are
provided to disable this behaviour.

This appears to have been the default behaviour on all modems I have ever
encountered (and have either deliberately or accidentally tested for this
behaviour!).

The 'escape' sequence, I believe, is intended to break into an established
connection. That it also happens to terminate the establishment phase is
almost certainly coincidental.


Wayne

{Original Message removed}

2000\05\19@005733 by Dale Botkin

flavicon
face
On Fri, 19 May 2000, W. Sierke wrote:

> It has always been my experience that a modem will abort making a
> connection, from the time the ATD command is issued to the modem, until the
> 'CONNECT' message is issued by the modem (CD signal raised), if any
> character is issued to the modem. I've never tested it to the extent of any
> other type of transmission (BREAK, mis-framed characters, invalid start
> bits, etc). Neither have I investigated whether in fact any commands are
> provided to disable this behaviour.
8< snip...

It's correct and documented in most of the older modems - you know, old
enough to have had documentation to begin with.  You are also correct that
the escape sequence is used to gain control of an established connection.
If your modem has decent docs, they will confirm this; if not, a quick
check of the docs for modems that HAD complete docs will do.  The
implementations are almost all very close.  In fact, I believe this even
extends across the Hayes AT and USR command sets style modems.

Dale

More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...