I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what would be
the most suitable cable I have 'in stock' Only need the send and return -
no 5v. Would balanced mic cable do - or two co-ax? I do have some decent
video co-ax but two would be a bit big to fit the D connector shells.

--
*It doesn't take a genius to spot a goat in a flock of sheep *

Dave Plowman London SW
To e-mail, change noise into sound.

Dave Plowman (News) wrote:
I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what would be
the most suitable cable I have 'in stock' Only need the send and return -
no 5v. Would balanced mic cable do - or two co-ax? I do have some decent
video co-ax but two would be a bit big to fit the D connector shells.

Mic cable - I've used star quad for countless serial cables, both RS232
and RS422.

In article ,
"Dave Plowman (News)" writes:
I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what would be
the most suitable cable I have 'in stock' Only need the send and return -
no 5v. Would balanced mic cable do - or two co-ax? I do have some decent
video co-ax but two would be a bit big to fit the D connector shells.

4m is not long for an RS232 style serial cable run, and almost
anything will do, although it gets more critical at higher bit rates.

Conventionally, untwisted screened cable is used, although there
are variants which use balanced drivers for the signals, which
require a twisted pair per signal.

Not sure what you mean by "return". Do you mean send and receive, or
just unidirectional send and the ground return wire?

--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]

Dave Plowman (News) wrote:
I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what would be
the most suitable cable I have 'in stock' Only need the send and return -
no 5v. Would balanced mic cable do - or two co-ax? I do have some decent
video co-ax but two would be a bit big to fit the D connector shells.

At that length you really need signal return and *ground* at least.

You wont get a huge speed over 4 meters.
I would use cat 5 for fixed wiring, or maybe even lightweight mains..


Bear in mind you will have to use either no flow control or Xon/Xoff if
you only use three wires.

5, to get DTR/DSR, is better.

In article ,
Andrew Gabriel wrote:
In article ,
"Dave Plowman (News)" writes:
I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what
would be the most suitable cable I have 'in stock' Only need the send
and return - no 5v. Would balanced mic cable do - or two co-ax? I do
have some decent video co-ax but two would be a bit big to fit the D
connector shells.

4m is not long for an RS232 style serial cable run, and almost
anything will do, although it gets more critical at higher bit rates.

The device I need to connect to says 115200 baud.

Conventionally, untwisted screened cable is used, although there
are variants which use balanced drivers for the signals, which
require a twisted pair per signal.

This is definitely unbalanced.

Not sure what you mean by "return". Do you mean send and receive, or
just unidirectional send and the ground return wire?

No - TX and RX and obviously ground. Those are the only connections on the
device DB9.

--
*Give me ambiguity or give me something else.

Dave Plowman London SW
To e-mail, change noise into sound.

On Sat, 11 Jul 2009 10:47:16 +0100, Dave Plowman (News) wrote:

I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what would
be the most suitable cable I have 'in stock' Only need the send and
return - no 5v.

RS232 serial is +/- 12v. USB is 5v... But you refer to D type
connectors so I assume you mean RS232. B-)

A bit of twin screened mib cable will work. Serial cables are
normally only x wires with an overall screen. No twisted pairs or
anything the signals are not balanced and quite big.

--
Cheers
Dave.

Dave Plowman (News) wrote:
In article ,
Andrew Gabriel wrote:
In article ,
"Dave Plowman (News)" writes:
I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what
would be the most suitable cable I have 'in stock' Only need the send
and return - no 5v. Would balanced mic cable do - or two co-ax? I do
have some decent video co-ax but two would be a bit big to fit the D
connector shells.

4m is not long for an RS232 style serial cable run, and almost
anything will do, although it gets more critical at higher bit rates.

The device I need to connect to says 115200 baud.

Conventionally, untwisted screened cable is used, although there
are variants which use balanced drivers for the signals, which
require a twisted pair per signal.

This is definitely unbalanced.

Not sure what you mean by "return". Do you mean send and receive, or
just unidirectional send and the ground return wire?

No - TX and RX and obviously ground. Those are the only connections on the
device DB9.

I think that is the same speed as the old Mac Apple Talk system. I used
to run that all over the house in 3 pair internal telephone wire with
other way connected to BT and 0/12v dc. I ran it unbalanced between 0v
and one space wire. Never observed any problems with many 10s of metres.

Bob

In article ,
"Dave Plowman (News)" writes:
In article ,
Andrew Gabriel wrote:
In article ,
"Dave Plowman (News)" writes:
I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what
would be the most suitable cable I have 'in stock' Only need the send
and return - no 5v. Would balanced mic cable do - or two co-ax? I do
have some decent video co-ax but two would be a bit big to fit the D
connector shells.

4m is not long for an RS232 style serial cable run, and almost
anything will do, although it gets more critical at higher bit rates.

The device I need to connect to says 115200 baud.

OK, that's quite high, but shouldn't be a problem. I've run 921600 baud
at 2m with correct cable without any data corruption, and it may have
gone further (didn't test). All modern UARTs do 8 or 16 times over-sampling
anyway, which means they cope with much worse signals than the original
RS232 specs allowed for. If you are running an error detecting/correcting
protocol over the link, you can get away with some corruption. If not,
then you need to consider the effect of any corruption.

You'll need a 3-core (or more) screened cable, and I would avoid twisted.

I used to buy quite a lot of made-up serial leads from CPC and they were
cheaper than you could buy the connectors alone, but their range seems to
have dropped to just two products now, and not long enough for you.

Conventionally, untwisted screened cable is used, although there
are variants which use balanced drivers for the signals, which
require a twisted pair per signal.

This is definitely unbalanced.

Not sure what you mean by "return". Do you mean send and receive, or
just unidirectional send and the ground return wire?

No - TX and RX and obviously ground. Those are the only connections on the
device DB9.

OK. Although DB9 combines signal ground and screen ground, I would
not use the screen as the signal ground in the cable, hence 3-core
plus screen, and connect the screen at just one end.

Presumably the device uses xon/xoff flow control, or guarantees to
be able to receive at full rate and doesn't send large blocks of data.

--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]

In message , "Dave Plowman (News)"
writes
I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what would be
the most suitable cable I have 'in stock' Only need the send and return -
no 5v. Would balanced mic cable do - or two co-ax? I do have some decent
video co-ax but two would be a bit big to fit the D connector shells.

Almost whatever you have to hand, as long as it's shielded. 4 metres is
trivial if we're talking about RS-232.

IIRC the max length for RS-232 by the specs is about 15 metres.

Obviously, the speed is fairly important and lower speeds will work
better over longer cables but I've used 115k at ten metres and 19.2 over
some silly distances (around 35 metres for a terminal) without
noticeable problems. Just pay attention to the shield and ground.

--
Clint Sharp

Dave Plowman (News) wrote:
I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what would be
the most suitable cable I have 'in stock' Only need the send and return -
no 5v. Would balanced mic cable do - or two co-ax? I do have some decent
video co-ax but two would be a bit big to fit the D connector shells.

Pretty much anything will work at that distance - even at 115K.

If you are wiring 9 way connectors then just connect:

9w 9w

2 3
3 2
5 5

For 25 way:

9w 9w

2 3
3 2
7 7

and for mixed sizes:

9w 25w

2 2 * note no twist required on mixed connectors
3 3
5 7



--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

On 11 July, 11:30, The Natural Philosopher
wrote:

Bear in mind you will have to use either no flow control or Xon/Xoff if
you only use three wires.

5, to get DTR/DSR, is better.

No, you use 5 wires with CTS/RTS as well. These control handshaking
at the character-by-character level, to avoid over-running buffers.

DSR/DTR are only needed if you can afford 7 wire, or if you really
need to deal with the "device ready" issue for batch operations.

Although you could potentially use XON/XOFF to replace either or both
of these, they're an unreliably way to replace CTS/RTS. If a character
buffer is full, the risk is that the XOFF is lost too. Besides which,
many modern devices don't do XON/XOFF.

Running in 3 wire splurge mode (no handshaking) works for slow send
speeds and fast receivers. Typically these days it works so long as
your tiny processor is sending data to your PC and the commands your
PC sends to the tiny processor are short. If you're doing large
downloads from the PC (big G code programs, display bitmaps, anything
over 1k long), then you can still get handshaking problems.

Electrically, RS232 is fussiest and 422 et al have wider margins. Even
so it's an electrically robust protocol and rarely a problem. If you
don't have multi-core, then do it with twisted pair phone cable but
put one signal and a ground down each pair. The data rates are low
enough that there's little coupling between the legs of the pair
anyway.

In article o.uk,
Dave Liquorice wrote:
I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what
would be the most suitable cable I have 'in stock' Only need the send
and return - no 5v.

RS232 serial is +/- 12v. USB is 5v... But you refer to D type
connectors so I assume you mean RS232. B-)

It's for the serial port on a PC. According to the blurb with the device
I'm connecting to it pins 1&9 have 5 volts DC on them to power things like
Bluetooth.

--
*Acupuncture is a jab well done*

Dave Plowman London SW
To e-mail, change noise into sound.

In article ,
John Rumm wrote:
I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what
would be the most suitable cable I have 'in stock' Only need the send
and return - no 5v. Would balanced mic cable do - or two co-ax? I do
have some decent video co-ax but two would be a bit big to fit the D
connector shells.

Pretty much anything will work at that distance - even at 115K.

If you are wiring 9 way connectors then just connect:

9w 9w

2 3
3 2
5 5

This setup uses a 'pass through' - not 'null modem'.

--
*If PROGRESS is for advancement, what does that make CONGRESS mean?

Dave Plowman London SW
To e-mail, change noise into sound.

On Sat, 11 Jul 2009 15:27:05 +0100, Dave Plowman (News) wrote:

It's for the serial port on a PC. According to the blurb with the device
I'm connecting to it pins 1&9 have 5 volts DC on them to power things
like Bluetooth.

RS232 DB9 pin 1 is Carrier Detect and pin 9 Ring Indicator so this
port is non-standard. I assume this non-standard port is on the bit
of kit you want to connect to the PC and it just has the abilty to
power a Bluetooth device if required.

--
Cheers
Dave.

In article o.uk,
"Dave Liquorice" writes:
On Sat, 11 Jul 2009 15:27:05 +0100, Dave Plowman (News) wrote:

It's for the serial port on a PC. According to the blurb with the device
I'm connecting to it pins 1&9 have 5 volts DC on them to power things
like Bluetooth.

RS232 DB9 pin 1 is Carrier Detect and pin 9 Ring Indicator so this
port is non-standard. I assume this non-standard port is on the bit
of kit you want to connect to the PC and it just has the abilty to
power a Bluetooth device if required.

There are a number of serial port boards which can be jumpered to
provide 5V or 12V outputs for powering the attached device. I have
one in my collection upstairs somewhere. (At one time, I did quite
a bit of work on the Solaris serial port driver, which necessitated
getting a wide selection of different boards to test it against.)
The other way to do this is to force DTR and RTS to known states,
and then use them to power a low-power device, and I've done this
a few times. At least Dave's device seems to be behaving as a DCE
and the power is provided on what are normally DCE outputs, so it
should be benign if plugged into a DTE with a 1-to-1 cable.

--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]

Andy Dingley wrote:
On 11 July, 11:30, The Natural Philosopher
wrote:

Bear in mind you will have to use either no flow control or Xon/Xoff if
you only use three wires.

5, to get DTR/DSR, is better.

No, you use 5 wires with CTS/RTS as well. These control handshaking
at the character-by-character level, to avoid over-running buffers.

Muy bad. Been too long. You are of course completely correct.

Clint Sharp wrote:
In message , "Dave Plowman (News)"
writes
I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what
would be
the most suitable cable I have 'in stock' Only need the send and return -
no 5v. Would balanced mic cable do - or two co-ax? I do have some decent
video co-ax but two would be a bit big to fit the D connector shells.

Almost whatever you have to hand, as long as it's shielded. 4 metres is
trivial if we're talking about RS-232.

IIRC the max length for RS-232 by the specs is about 15 metres.

Obviously, the speed is fairly important and lower speeds will work
better over longer cables but I've used 115k at ten metres and 19.2 over
some silly distances (around 35 metres for a terminal) without
noticeable problems. Just pay attention to the shield and ground.

Used to run 19.2k around cat 5 to glass terminals - serious distances,
but couldn't get higher, so I assumed that the OP wouldn't;'t be able to
go much above 100k, which is what everyone else seems to say.

On 11 Jul 2009 18:32:30 GMT, Andrew Gabriel wrote:

The other way to do this is to force DTR and RTS to known states,
and then use them to power a low-power device, and I've done this
a few times.

Yep so have I and got confused when said device doesn't work 'cause
the associated software wasn't controling the lines as expected...

--
Cheers
Dave.

On Sat, 11 Jul 2009 13:25:25 +0100, Clint Sharp
wrote:

IIRC the max length for RS-232 by the specs is about 15 metres.

The maximum specification cable length is 50 feet, or the cable length
equal to a capacitance of 2500 pF. It is the puffage rather than the
length which is critical so using a cable with low capacitance allows
you to span longer distances without going beyond the limitations of
the standard. Using CAT-5 cable with a typical capacitance of 17
pF/ft, the maximum allowed cable length is about 50m.

Somewhere in the garage I've some line drivers that will allow you to
go to about a km if you want.

The Natural Philosopher wrote:
Clint Sharp wrote:
In message , "Dave Plowman (News)"
writes
I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what
would be
the most suitable cable I have 'in stock' Only need the send and
return -
no 5v. Would balanced mic cable do - or two co-ax? I do have some decent
video co-ax but two would be a bit big to fit the D connector shells.

Almost whatever you have to hand, as long as it's shielded. 4 metres
is trivial if we're talking about RS-232.

IIRC the max length for RS-232 by the specs is about 15 metres.

Obviously, the speed is fairly important and lower speeds will work
better over longer cables but I've used 115k at ten metres and 19.2
over some silly distances (around 35 metres for a terminal) without
noticeable problems. Just pay attention to the shield and ground.

Used to run 19.2k around cat 5 to glass terminals - serious distances,
but couldn't get higher, so I assumed that the OP wouldn't;'t be able to
go much above 100k, which is what everyone else seems to say.

Many UARTs won't run more than 115K, so that is a fair bet ;-)

--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

In article , Dave Plowman (News)
writes

The device I need to connect to says 115200 baud.

No - TX and RX and obviously ground. Those are the only connections on the
device DB9.

High baud rate and no hardware handshaking, though 4m is not long for
RS232. I'd suggest a good quality screened cable. The mic cable
suggested by another poster should do the trick. Connect the screen to
the plug body at the PC end.

--
(\__/)
(='.'=) Bunny says Windows 7 is Vi$ta reloaded.
(")_(") http://imgs.xkcd.com/comics/windows_7.png

In article , Clint Sharp
writes

Obviously, the speed is fairly important and lower speeds will work
better over longer cables but I've used 115k at ten metres and 19.2 over
some silly distances (around 35 metres for a terminal) without
noticeable problems.

I've seen several installations using 9600 baud at a couple hundred
metres, though at those lengths quality shielded cable is needed and the
line drivers are more prone to failure due to induced noise. This was
RS232, not one of the balanced variants.

--
(\__/)
(='.'=) Bunny says Windows 7 is Vi$ta reloaded.
(")_(") http://imgs.xkcd.com/comics/windows_7.png

In article ,
Mike Tomlinson wrote:
No - TX and RX and obviously ground. Those are the only connections on the
device DB9.

High baud rate and no hardware handshaking, though 4m is not long for
RS232. I'd suggest a good quality screened cable. The mic cable
suggested by another poster should do the trick. Connect the screen to
the plug body at the PC end.

That's a good idea even if it's also connected to pin 5?

--
*Your kid may be an honours student, but you're still an idiot.

Dave Plowman London SW
To e-mail, change noise into sound.

In article , Dave Plowman (News)
writes

That's a good idea even if it's also connected to pin 5?

Yes. pin 5 is the ground return for the serial data. Connecting the
screen of the cable to the body of the D plug will ensure the cable is
screened. You mention it being in a workshop, with the electrically
noisy environment that implies.

--
(\__/)
(='.'=) Bunny says Windows 7 is Vi$ta reloaded.
(")_(") http://imgs.xkcd.com/comics/windows_7.png

Dave Plowman (News) wrote:
In article ,
Mike Tomlinson wrote:
No - TX and RX and obviously ground. Those are the only connections on the
device DB9.

High baud rate and no hardware handshaking, though 4m is not long for
RS232. I'd suggest a good quality screened cable. The mic cable
suggested by another poster should do the trick. Connect the screen to
the plug body at the PC end.

That's a good idea even if it's also connected to pin 5?

Its better to keep signal ground and screen separate... so generally I
use a core in a screened cable for signal ground, and save the screen
for its intended purpose.


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

Peter Parry wrote:
On Sat, 11 Jul 2009 13:25:25 +0100, Clint Sharp
wrote:

IIRC the max length for RS-232 by the specs is about 15 metres.

The maximum specification cable length is 50 feet, or the cable length
equal to a capacitance of 2500 pF. It is the puffage rather than the
length which is critical so using a cable with low capacitance allows
you to span longer distances without going beyond the limitations of
the standard. Using CAT-5 cable with a typical capacitance of 17
pF/ft, the maximum allowed cable length is about 50m.

Somewhere in the garage I've some line drivers that will allow you to
go to about a km if you want.

Might be worth noting that most PCs etc have ports with drivers based on
the RS232C spec (lower voltage signalling) and not the full RS232 one.
So results will vary with different kit depending on its spec.



--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

John Rumm wrote:
Dave Plowman (News) wrote:
In article ,
Mike Tomlinson wrote:
No - TX and RX and obviously ground. Those are the only connections
on the
device DB9.

High baud rate and no hardware handshaking, though 4m is not long for
RS232. I'd suggest a good quality screened cable. The mic cable
suggested by another poster should do the trick. Connect the screen to
the plug body at the PC end.

That's a good idea even if it's also connected to pin 5?

Its better to keep signal ground and screen separate... so generally I
use a core in a screened cable for signal ground, and save the screen
for its intended purpose.

I take it that one end of the cable has a flying braided lead to earth
the screen then?

Dave

In article ,
John Rumm wrote:
Dave Plowman (News) wrote:
In article ,
Mike Tomlinson wrote:
No - TX and RX and obviously ground. Those are the only connections
on the device DB9.

High baud rate and no hardware handshaking, though 4m is not long for
RS232. I'd suggest a good quality screened cable. The mic cable
suggested by another poster should do the trick. Connect the screen
to the plug body at the PC end.

That's a good idea even if it's also connected to pin 5?

Its better to keep signal ground and screen separate... so generally I
use a core in a screened cable for signal ground, and save the screen
for its intended purpose.

I don't actually have any 3 core screened.

As it happens mic cable works just fine. There was a fault in the wiring
instructions for the serial to RG45 lead they supplied with the D socket
loose - TX and RX reversed. Which to be fair was in the errata at the end
of the instructions. ;-)

--
*Isn't it a bit unnerving that doctors call what they do "practice?"

Dave Plowman London SW
To e-mail, change noise into sound.

On Jul 12, 12:20*am, John Rumm wrote:
The Natural Philosopher wrote:
Clint Sharp wrote:
In message , "Dave Plowman (News)"
writes
I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what
would be
the most suitable cable I have 'in stock' Only need the send and
return -
no 5v. Would balanced mic cable do - or two co-ax? I do have some decent
video co-ax but two would be a bit big to fit the D connector shells.

Almost whatever you have to hand, as long as it's shielded. 4 metres
is trivial if we're talking about RS-232.

IIRC the max length for RS-232 by the specs is about 15 metres.

Obviously, the speed is fairly important and lower speeds will work
better over longer cables but I've used 115k at ten metres and 19.2
over some silly distances (around 35 metres for a terminal) without
noticeable problems. Just pay attention to the shield and ground.

Used to run 19.2k around cat 5 to glass terminals - serious distances,
but couldn't get higher, so I assumed that the OP wouldn't;'t be able to
go much above 100k, which is what everyone else seems to say.

Many UARTs won't run more than 115K, so that is a fair bet ;-)

But many more do.

Gazillions of UARTS in microcontrollers will run way faster than that.
I've done 1mbit/s in a PIC, between two devices on a PCB.

The crystal selection becomes critical at higher speeds if you want to
use "standard" baud rates.

MBQ

"Man at B&Q" wrote in message
...


Many UARTs won't run more than 115K, so that is a fair bet ;-)

But many more do.

Gazillions of UARTS in microcontrollers will run way faster than that.
I've done 1mbit/s in a PIC, between two devices on a PCB.

We used to run the archive tape drives on System X at 880k over 150m of
cable.
We invented a new tape protocol just for the job as we hated the stupid
ascii interface Tandberg had chucked together.
That was in the '80s

Dave wrote:

Its better to keep signal ground and screen separate... so generally I
use a core in a screened cable for signal ground, and save the screen
for its intended purpose.

I take it that one end of the cable has a flying braided lead to earth
the screen then?

On the 25 way connector pin 1 is reserved for screen, and pin 7 for
signal ground. The screen is also in theory paralleled with the shell of
the D connector.

--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

Dave Plowman (News) wrote:
In article ,
John Rumm wrote:
Dave Plowman (News) wrote:
In article ,
Mike Tomlinson wrote:
No - TX and RX and obviously ground. Those are the only connections
on the device DB9.
High baud rate and no hardware handshaking, though 4m is not long for
RS232. I'd suggest a good quality screened cable. The mic cable
suggested by another poster should do the trick. Connect the screen
to the plug body at the PC end.
That's a good idea even if it's also connected to pin 5?

Its better to keep signal ground and screen separate... so generally I
use a core in a screened cable for signal ground, and save the screen
for its intended purpose.

I don't actually have any 3 core screened.

As it happens mic cable works just fine. There was a fault in the wiring
instructions for the serial to RG45 lead they supplied with the D socket
loose - TX and RX reversed. Which to be fair was in the errata at the end
of the instructions. ;-)

Often happens - not helped by the swap over of TX and RX between
connector sizes.


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

Man at B&Q wrote:
On Jul 12, 12:20 am, John Rumm wrote:
The Natural Philosopher wrote:
Clint Sharp wrote:
In message , "Dave Plowman (News)"
writes
I need a longish serial cable to get from the workshop computer to the
bench. Say 4 mtr. I already have the D connectors and wonder what
would be
the most suitable cable I have 'in stock' Only need the send and
return -
no 5v. Would balanced mic cable do - or two co-ax? I do have some decent
video co-ax but two would be a bit big to fit the D connector shells.
Almost whatever you have to hand, as long as it's shielded. 4 metres
is trivial if we're talking about RS-232.
IIRC the max length for RS-232 by the specs is about 15 metres.
Obviously, the speed is fairly important and lower speeds will work
better over longer cables but I've used 115k at ten metres and 19.2
over some silly distances (around 35 metres for a terminal) without
noticeable problems. Just pay attention to the shield and ground.
Used to run 19.2k around cat 5 to glass terminals - serious distances,
but couldn't get higher, so I assumed that the OP wouldn't;'t be able to
go much above 100k, which is what everyone else seems to say.
Many UARTs won't run more than 115K, so that is a fair bet ;-)

But many more do.

Gazillions of UARTS in microcontrollers will run way faster than that.
I've done 1mbit/s in a PIC, between two devices on a PCB.

Indeed - but most are 8250(A) or 16550 style devices in PCs and similar.
With the selected xtal these usually do about 50 - 115K bps.

Given a 68HC302 or similar with the embedded "communications element"
you can do all manner of exotica with the serial lines!

The crystal selection becomes critical at higher speeds if you want to
use "standard" baud rates.

Yup, there are not many choices if you want all the usual ones.

--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

John Rumm wrote:
Dave wrote:

Its better to keep signal ground and screen separate... so generally
I use a core in a screened cable for signal ground, and save the
screen for its intended purpose.

I take it that one end of the cable has a flying braided lead to earth
the screen then?

On the 25 way connector pin 1 is reserved for screen, and pin 7 for
signal ground. The screen is also in theory paralleled with the shell of
the D connector.

Thanks John, I don't have my spec handy at the moment and it has been a
long time since I played with these conectors.

Dave

In article ,
John Rumm writes:
Man at B&Q wrote:

Gazillions of UARTS in microcontrollers will run way faster than that.
I've done 1mbit/s in a PIC, between two devices on a PCB.

Indeed - but most are 8250(A) or 16550 style devices in PCs and similar.
With the selected xtal these usually do about 50 - 115K bps.

Don't think 8250 has been used in anything since the advent of 386's.
Oldest I could find when sorting through Interactive UNIX's stock
of really old test cards at work was a 16450 on an 8-bit ISA-bus card.

Southbridge chips contain embedded 16550A equivalents at least, and
some turn out to be 16650 if you probe for them. Plug-in cards are
16650 through to 16950 (increasing buffer sizes, increasing numbers
of features for high speed working with less driver intervention, but
unfortunately increasingly less compatible between manufacturers).
Very many cards nowadays use 8x the clock frequency a 16550 uses,
with a divide by 8 which can be enabled/disabled to get to nearly a
megabit, or to give the conventional 16550 speeds. (Sometimes the
divide by 8 enabled/disabled applies to all ports on a multi-port
card, which is a bummer as you can't set the baud rates independantly
on each port to all of the commonly used values.)

Given a 68HC302 or similar with the embedded "communications element"
you can do all manner of exotica with the serial lines!

The crystal selection becomes critical at higher speeds if you want to
use "standard" baud rates.

As I said earlier in the thread, modern UARTs are much more tollerant
of mis-shaped signals and sample the data stream at 8 or 16 times the
baud rate. This means they'll work with cable lengths well in excess
of those in the original RS242 and V.24 specs.

--
Andrew Gabriel
[email address is not usable -- followup in the newsgroup]

Dave wrote:
John Rumm wrote:
Dave wrote:

Its better to keep signal ground and screen separate... so generally
I use a core in a screened cable for signal ground, and save the
screen for its intended purpose.

I take it that one end of the cable has a flying braided lead to
earth the screen then?

On the 25 way connector pin 1 is reserved for screen, and pin 7 for
signal ground. The screen is also in theory paralleled with the shell
of the D connector.

Thanks John, I don't have my spec handy at the moment and it has been a
long time since I played with these conectors.

You know what they say about RS232 - the 232 bit describes the number of
valid ways you can connect three wires to a 25 pin plug ;-)

--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

Andrew Gabriel wrote:
In article ,
John Rumm writes:
Man at B&Q wrote:
Gazillions of UARTS in microcontrollers will run way faster than that.
I've done 1mbit/s in a PIC, between two devices on a PCB.
Indeed - but most are 8250(A) or 16550 style devices in PCs and similar.
With the selected xtal these usually do about 50 - 115K bps.

Don't think 8250 has been used in anything since the advent of 386's.
Oldest I could find when sorting through Interactive UNIX's stock
of really old test cards at work was a 16450 on an 8-bit ISA-bus card.

Yup they went out with the ark... I have a schematic from a PC-AT tech
ref manual that does show one though (from the days where a 6MHz 80286
was the best you could get ;-)

The 16450 is not much better since it does not have a big enough fifo.
So couple it with a OS with louse real time performance (like windows)
and its useless.

Southbridge chips contain embedded 16550A equivalents at least, and
some turn out to be 16650 if you probe for them. Plug-in cards are
16650 through to 16950 (increasing buffer sizes, increasing numbers
of features for high speed working with less driver intervention, but
unfortunately increasingly less compatible between manufacturers).
Very many cards nowadays use 8x the clock frequency a 16550 uses,
with a divide by 8 which can be enabled/disabled to get to nearly a
megabit, or to give the conventional 16550 speeds. (Sometimes the
divide by 8 enabled/disabled applies to all ports on a multi-port
card, which is a bummer as you can't set the baud rates independantly
on each port to all of the commonly used values.)

I remember phoning round the multi serial card manufacturers a few years
ago, trying to find one that would support 75 bps for use with older
radio modems. It completely threw them off their sales pitch having
someone not interested in how fast it could go, but how slow ;-)


Given a 68HC302 or similar with the embedded "communications element"
you can do all manner of exotica with the serial lines!

The crystal selection becomes critical at higher speeds if you want to
use "standard" baud rates.

As I said earlier in the thread, modern UARTs are much more tollerant
of mis-shaped signals and sample the data stream at 8 or 16 times the
baud rate. This means they'll work with cable lengths well in excess
of those in the original RS242 and V.24 specs.

Agreed. In fact even much of the older RS232 stuff would routinely go
way beyond spec.

--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

John Rumm wrote:

I remember phoning round the multi serial card manufacturers a few
years ago, trying to find one that would support 75 bps for use with
older radio modems. It completely threw them off their sales pitch
having someone not interested in how fast it could go, but how slow ;-)


That's nowt - try looking for 45.45bps support :-)

Apparently the limitation is the number of clock divider stages in the
UARTs. When they increase the clock speed to support higher data rates,
they don't have enough dividers to get below 300bps.


--
Ian White