RS232 cable length

[SD2100]

Hi, What is the maximum distance for reliable RS232 communications, I've read that up to 50 metres using special cable at reduced baud rates can be acheived. I'm currently testing this at a baud rate of 2400 using Cat5 cable of approx 35 metres in length and has proved to be very reliable.

 

[Dippy]

I'm quoting, so don't blame me!

".. the maximum cable length is 50 feet, or the cable length equal to a capacitance of 2500 pF. This means that using a cable with low capacitance allows you to span longer distances without going beyond the limitations of the standard. If for example UTP CAT-5 cable is used with a typical capacitance of 17 pF/ft, the maximum allowed cable length is 147 feet.

Texas Instruments has done some practical experiments years ago at different baud rates. Keep in mind, that the RS232 standard was originally developed for 20 kbps. By halving the maximum communication speed, the allowed cable length increases a factor ten!


RS232 cable length according to Texas Instruments
Baud rate Maximum cable length (ft)
19200 50
9600 500
4800 1000
2400 3000"

 

[hippy]

The standard may also apply to genuine RS232 +15V/-15V signalling so distance may be reduced using the 0V/5V signalling scheme a PICAXE uses. Adjusting the 10K/22K values on a PICAXE receiver may help.

If distance does become a problem, you can always add MAX232's or other line drivers; have used DMX ( 250kb ) over 1km of unshielded mains cable using a balanced line driver with no problems.

 

[SD2100]

Thanks for that, I'm using an old PC at one end (house) and an 18X with MAX232 at the other (shed). I might be putting the same thing in another shed which is 50m away so this distance is looking to be ok as well.

 

[manuka]

Picaxe to Picaxe serial is <b>extremely </b> robust &amp; as a party trick we regularly manage to send 2400bps data thru' damp string, chains of students holding hands, 100s of m of bell wire &amp; even SWER (single wire earth return) etc. I've yet to find the maximum length supported in fact!

Although such conductors are colourful, their installation over even 50m may be an on going hassle. Considered 433MHz wireless links ?

 

[ylp88]

National Semiconductor used &quot;50-100 feet&quot; in one of their online seminars. They didn't quote any specific speed but did mention &quot;20Kb (120Kb, 250Kb &amp; 1M with some drivers)&quot; as the maximum speed, on the same slide.

<b><i>ylp88 </b> </i>

 

[SD2100]

I bought the 433Mhz TX/RX units a month ago and did think about using them and still might if the serial connection starts to play up, but at the moment the serial is performing better than expected.

 

[xstamp]

I once needed to transmit serial data at 1200 bps from a garage to the pit wall at a motor racing track, without laying a cable and found IR to be a simple solution for transmission over several tens of metres. A 555 astable drove a high power led at 38KHz and an IR detector/amp chip (eg. HIM602V) was used at the receiving end of the link. Both the transmitter and receiver were connected directly to PC serial ports and housed in tubes with IR lenses (sadly, no longer available from Maplin Electronics) to produce a narrow beam and also reduce the effects of sunlight. You would of course be limited to line-of-site using this technique.

 

[SD2100]

I might have to setup some mirrors to get around the lemon tree

Isn't IR in sunlight a little bit hit &amp; miss ???, I've had IR things false triggering etc in the sun !!!.

Edited by - Phil75 on 03/05/2006 13:27:56

 

[xstamp]

You certainly have to take care regarding sunlight, like reducing the viewing angle of the receiver and using a modulated beam, but it can be a very robust means of data communications. Especially if you have lots of electrical interference from radio transmitters and engine ignition etc (long RS-232 runs can be very sensitive to noise). We also tried laser beams but found the very narrow beam width made alignment difficult.

 

[SD2100]

I played a bit with lasers over a distance but rain and fog cause havoc. Also at distance the beam tends to spread a bit

Edited by - Phil75 on 03/05/2006 15:54:21

 

[Dippy]

A wide beam is better than a narrow one as it isn't affected by water droplets so much.
(Sorry for stating the obvious.)

A tiny thing to consider re: 433/434 MHz, if the transmitter is of healthy spec. 433/434 is fairly 'close' to the bottom end of TV and will almost certainly disturb DTT (UK spec) if nearby. As the bloke next door reminded me when I was transmitting at 25mW.

So, if your data is fairly frequent and your neighbour is fairly close then he/she will be ever so grateful if you stick to cable.

It may not be a problem of course and you may loathe your neighbour.... or, if you're lucky, 'next door' is 10 miles away!

 

[SD2100]

If their dog can bark then so can my transmitter. hehe

 

[SD2100]

What would be the best use of the cable to reduce the likely hood of interference.

The Cat 5 cable I'm using has 4 twisted pairs, one pair I'm using for TX/RX, and one wire of another pair is used for signal ground, the spare wire of this pair is floating at the moment, probably not good as will pickup interference.

Option 1:
one pair for TX/RX
another pair with both wires parallelled for ground.

Option 2:
one pair TX and ground
another pair RX and ground

Unused pairs will be tied to ground also.

Edited by - Phil75 on 04/05/2006 07:51:21

 

[andrewpro]

I would go the one ground per signal line route. So RX and grnd, TX and grnd. This will keep any interference/crosstalk hopefully to a minimum.

--Andy P

 

[xstamp]

With all those wires available you could consider using RS-485 drivers/receiver chips (try Maxim/Dallas for samples). This differential standard will drive around 2km of twisted core cable, at high speed, with very good noise immunity.