Multiplying a PC's serial output

[Patrick Keenan]

Hello,
I suppose this sort of qualifies as electronics design...

We're looking at splitting the signal from a laptop's serial port in a way
that is probably going to need some electronic support. The signal is
actually ASCII output from a court reporter's laptop. Data travels one
way only, out from the reporter.

There's a passive system that's basically just a plastic splitter box , but
it limits to five outputs and is intended for temporary use around a table.

These folks, however, want a permanent install in a larger room, so
there's maybe 10 - 12 drops and perhaps a few hundred feet of cable. We
suspect that's a bit much to push from the laptop and want to be sure that
the data never fades or corrupts on the line and that nothing ever gets
damaged. There's lots of excess cat5 cable installed under the room, so if
that can be used installation should be simple..

Does anyone know of readily available devices for such a task?

Thanks!
-pk

 

[CptDondo]

Hello,
I suppose this sort of qualifies as electronics design...

We're looking at splitting the signal from a laptop's serial port in a way
that is probably going to need some electronic support. The signal is
actually ASCII output from a court reporter's laptop. Data travels one
way only, out from the reporter.

There's a passive system that's basically just a plastic splitter box , but
it limits to five outputs and is intended for temporary use around a table.

These folks, however, want a permanent install in a larger room, so
there's maybe 10 - 12 drops and perhaps a few hundred feet of cable. We
suspect that's a bit much to push from the laptop and want to be sure that
the data never fades or corrupts on the line and that nothing ever gets
damaged. There's lots of excess cat5 cable installed under the room, so if
that can be used installation should be simple..

Does anyone know of readily available devices for such a task?

Not sure, but you really want to opto-isolate all of the PCs. I've
blown up many computers by plugging in serial cables. (I do embedded
development thus I tend to plug a lot of strange things together.)

I've seen as much as 70 volts difference between grounds on different
equipment - enough to burn chips off the motherboard.

Serial ports are my favorite way to destroy equipment. Heck, I shipped
one board back today for a post-mortem because the serial port quit
working; I suspect due to some sort of ground problem....

Look at offerings from Black Box. I'd also think about something like
the TS7300 from embeddedarm.com; I think they have an FPGA bitstream
with something like 20 or 30 serial ports for that hardware.

--Yan

 

[Meat Plow]

Hello,
I suppose this sort of qualifies as electronics design...

We're looking at splitting the signal from a laptop's serial port in a way
that is probably going to need some electronic support. The signal is
actually ASCII output from a court reporter's laptop. Data travels one
way only, out from the reporter.

There's a passive system that's basically just a plastic splitter box , but
it limits to five outputs and is intended for temporary use around a table.

These folks, however, want a permanent install in a larger room, so
there's maybe 10 - 12 drops and perhaps a few hundred feet of cable. We
suspect that's a bit much to push from the laptop and want to be sure that
the data never fades or corrupts on the line and that nothing ever gets
damaged. There's lots of excess cat5 cable installed under the room, so if
that can be used installation should be simple..

Does anyone know of readily available devices for such a task?

Thanks!
-pk

http://www.barcode-manufacturer.com/serial-ethernet/

 

[Patrick Keenan]

http://www.barcode-manufacturer.com/serial-ethernet/

Thanks! I've seen this kind of thing. but am not sure that it actually
will meet the limitations of this particular application, in particular the
software running at the target PC. Certainly does bear investigation,
though and I will get a couple as it looks like it addresses part of how to
extend this application (which can otherwise be well beyond obscenely
expensive).

thanks again.
-pk

 

[Patrick Keenan]

Not sure, but you really want to opto-isolate all of the PCs. I've blown
up many computers by plugging in serial cables. (I do embedded
development thus I tend to plug a lot of strange things together.)

I've seen as much as 70 volts difference between grounds on different
equipment - enough to burn chips off the motherboard.

Serial ports are my favorite way to destroy equipment. Heck, I shipped
one board back today for a post-mortem because the serial port quit
working; I suspect due to some sort of ground problem....

Look at offerings from Black Box. I'd also think about something like the
TS7300 from embeddedarm.com; I think they have an FPGA bitstream with
something like 20 or 30 serial ports for that hardware.

--Yan

Frankly that's one of my main concerns and I've been looking at BlackBox and
other spots, and the various serial line drivers and isolators. Just not
sure how they actually go together. To use these as an example,

http://www.bb-elec.com/product_family.asp?FamilyId=1&Trail=2&TrailType=Top

in particular the 9SPOP2,

there'd be one at each serial port and basically a bus.. kind of a rake
shape, not a Y - of wire between?

Thanks again for your comments.
-pk

 

[Andy]

<http://www.rs485.com/prs232hub.html>
The RS232HUB is a 4-port RS232 repeater hub(non-isolated) with one
Master RS232 port and 3-RS232 Slave ports. Data received by the Master
port is transfered to the 3 Slave ports, and data from a Slave port(s)
is transfered to the master port only. Care must be taken to insure
that only one device responds at a time to avoid "data collisions" in
some applications. Jumpers are provided to enable or disable the
Transmit and/or Receive function on each port. The operating
temperature range is -40C to +85C.

 

[Roger Hamlett]

Hello,
I suppose this sort of qualifies as electronics design...

We're looking at splitting the signal from a laptop's serial port in a way
that is probably going to need some electronic support. The signal is
actually ASCII output from a court reporter's laptop. Data travels one
way only, out from the reporter.

There's a passive system that's basically just a plastic splitter box , but
it limits to five outputs and is intended for temporary use around a table.

These folks, however, want a permanent install in a larger room, so
there's maybe 10 - 12 drops and perhaps a few hundred feet of cable. We
suspect that's a bit much to push from the laptop and want to be sure that
the data never fades or corrupts on the line and that nothing ever gets
damaged. There's lots of excess cat5 cable installed under the room, so if
that can be used installation should be simple..

Does anyone know of readily available devices for such a task?

Thanks!
-pk

One answer, is to get RS232 to RS485 converter units. These are
available from industrial suppliers, in either Din rail mount versions,
or small standalone boxes. RS485, gives you support for the larger
number of drops, and some are available with opto-isolation as well.
This gives you a good 'safe' solution.
Go to www.moxa.com, and look at their TCC-801 (avoids need for power
supplies on the listening points).
Alternatively, as you may start finding that the other computers are
less and less likely to have RS232 serial ports, consider a USB to RS485
adapter for the listening locations.

Best Wishes

 

[CptDondo]

Frankly that's one of my main concerns and I've been looking at BlackBox and
other spots, and the various serial line drivers and isolators. Just not
sure how they actually go together. To use these as an example,

http://www.bb-elec.com/product_family.asp?FamilyId=1&Trail=2&TrailType=Top

in particular the 9SPOP2,

there'd be one at each serial port and basically a bus.. kind of a rake
shape, not a Y - of wire between?

A vendor just sent me this:

8018630000 Weidmuller minicoupler 5VDC input, 5-48VDC output

They're about $4 ea, din-rail mountable, and if I understand things
correctly, you could just run a block of these on your TX and GND, and
then connect the other laptops to the opto-isolated outputs. You'd have
to think through the power supplies (you'd need at least 2, one for each
side of the opto isolation), and possibly one for each laptop. But it's
a lot cheaper than individual opto-isolators, and with individual power
supplies you'd have no problems with signal.

--Yan

 

[Rich Grise]

I've seen as much as 70 volts difference between grounds on different
equipment - enough to burn chips off the motherboard.

If that's true, then you have seriously dangerous electrical wiring,
not to mention WAY out of code.

You need to get your wiring repaired before you kill somebody.

Thanks,
Rich

 

[CptDondo]

If that's true, then you have seriously dangerous electrical wiring,
not to mention WAY out of code.

You need to get your wiring repaired before you kill somebody.

Run a serial cable for 2,000' along a 480 VAC line and you'll get 70 V
induction.....

Meets code but you'd better have serious isolation.

A log of what we do is not earthed because of groundstrikes from
lighting and because the equipment is mobile and very large. So it's
possible to get floating ground potential between equipment.

We test for that in the lab but sometimes I screw up and blow something
off the wall.

 

[whit3rd]

We're looking at splitting the signal from a laptop's serial port in a way
that is probably going to need some electronic support. The signal is
actually ASCII output from a court reporter's laptop. Data travels one
way only, out from the reporter.

There's a passive system that's basically just a plastic splitter box , but
it limits to five outputs \

There are lots of print server boxes that put a port (or three) onto
Ethernet. If one is used next to your court reporter, the data
would be available (with some software diddling) to any user on
the local Ethernet branch. Just use Ethernet broadcast features.

If you have lots of CAT5 cable, it's probably trivial to put a
suitable
Ethernet port at each station, and connect 'em to a small
hub.

If the reporter's box is programmable enough, it could be done through
any available instant-messaging software, too; I don't know how
important the 'ASCII output on serial port' constraint is.

 

[Patrick Keenan]

http://www.barcode-manufacturer.com/serial-ethernet/

Thanks to all who have responded to this question.

In case it's useful to anyone later, this is what turns out to be not just
the most appropriate but likely the best solution:

TCP-Com, from TAL Tech.
http://www.taltech.com/products/tcpcom.html?gclid=CNGTyeX0_Y4CFQYjWAodDHvU2g

"TCP-Com is a software based RS232 to TCP/IP converter. TCP-Com allows any
of the RS232 serial ports on your PC to interface directly to a TCP/IP
network. For example, you can use TCP-Com to turn a PC into a “Serial Device
Server” so that you can connect any RS232 serial device directly to a TCP/IP
network and communicate with that device from any other workstation in the
same network or across the Internet.

TCP-Com can also create “Virtual” RS232 serial ports that are actually
connections to a TCP/IP port. This allows you to use existing Windows based
serial communications software to send and receive data across a TCP/IP
network."

Using the free-30-day-demo version, we were able to connect a reporter's
system to a receiving system over a local wireless network connection.
None of the custom cables or the send/receive adapter pairs, or
USB-to-serial adapters, were needed.

It took us about 15 minutes to start up the systems, install the software,
figure out the settings, adjust the reporting software and make the whole
thing work. It'd be hard for it to be easier or less cluttered.

So we'll be recommending this, with the client simply repatching the
existing network cabling to the desks that need this feed. At most,
because access limits have to be imposed (particularly for in-camera
sessions), another router will have to be installed to isolate those desks.

Thanks again,
Patrick Keenan

 

[Patrick Keenan]

Hello,
I suppose this sort of qualifies as electronics design...

We're looking at splitting the signal from a laptop's serial port in a way
that is probably going to need some electronic support. The signal is
actually ASCII output from a court reporter's laptop. Data travels one
way only, out from the reporter.

There's a passive system that's basically just a plastic splitter box ,
but it limits to five outputs and is intended for temporary use around a
table.

These folks, however, want a permanent install in a larger room, so
there's maybe 10 - 12 drops and perhaps a few hundred feet of cable. We
suspect that's a bit much to push from the laptop and want to be sure
that the data never fades or corrupts on the line and that nothing ever
gets damaged. There's lots of excess cat5 cable installed under the
room, so if that can be used installation should be simple..

Does anyone know of readily available devices for such a task?

Thanks!
-pk

Thanks to all who have responded to this question.

In case it's useful to anyone later, this is what turns out to be not just
the most appropriate but likely the best solution in this specific context.

There may be other vendors of this type of redirector, but this is the one I
found and tested:

TCP-Com, from TAL Tech.
http://www.taltech.com/products/tcpcom.html?gclid=CNGTyeX0_Y4CFQYjWAodDHvU2g

"TCP-Com is a software based RS232 to TCP/IP converter. TCP-Com allows any
of the RS232 serial ports on your PC to interface directly to a TCP/IP
network. For example, you can use TCP-Com to turn a PC into a "Serial Device
Server" so that you can connect any RS232 serial device directly to a TCP/IP
network and communicate with that device from any other workstation in the
same network or across the Internet.

TCP-Com can also create "Virtual" RS232 serial ports that are actually
connections to a TCP/IP port. This allows you to use existing Windows based
serial communications software to send and receive data across a TCP/IP
network."

Using the free-30-day-demo version, we were able to connect a reporter's
system to a receiving system over a local wireless network connection.
None of the custom cables or the send/receive adapter pairs, or
USB-to-serial adapters, were needed.

It took us about 15 minutes to start up the systems, install the software,
figure out the settings, adjust the reporting software and make the whole
thing work. It'd be hard for it to be easier or less cluttered.

So we'll be recommending this, with the client simply repatching the
existing network cabling to the desks that need this feed; all of the desks
have multiple ethernet jacks, some of which are simply not patched to
anything at the moment. At most, because access limits have to be imposed
(particularly for in-camera
sessions), another router will have to be installed to isolate those desks.

Thanks again,
Patrick Keenan