A low voltage/RS232 type question

Joined
6 Sep 2008
Messages
2,334
Reaction score
258
Country
United Kingdom
I have a ventilation system in the loft, that came with a wired controller; a display and couple of buttons about the size of a light switch. It also had bundled with it approximately 10 metres of telephone cable terminated in a 4P4C connector at one end (intended to plug into the ventilation unit's BMS socket) and bare wire the other, for connecting to the controller thing

The cable was no way long enough to make it down stairs, and the house is more or less built now so routing it would make quite a mess. I did however have a spare cat 6 network cable within 10 metres, routing around the eaves, so I took the phone cable to the cat 6, crimped each of the 4 phone wires to a pair in the cat 6 (I figured why not use 2 cores in the cat6, lower the resistance..) and then connected the controller up at the other end. Powering it all up showed a fault code on the controller

I dug out a modem wire and shaped the 6P4C connector to fit the BMS socket, then wired that up - controller works fine on this 1 metre cable. I also noted that if I disconnect only wire #2 from the controller, leaving the other 3 connected, then I get the same fault code with the short wire
Thinking I had a broken wire, I tested the adapted (the joined telephone + cat6 ) cable for continuity by joining telephone pairs 1,2 and 3,4 together then multimetering the other end, noting a resistance of 2.5 ish ohms for each loop. I then joined 1 and 3 and 2 and 4 and checked them again; all connected as they should, and the multimeter went OL when I had an assistant disconnect the wires from each other.. so right now I don't think there's a break in the wire, or a short between cores. NOTE; I couldn't test right from the 4P4C plug because my multimeter probe cant reach the contacts. I've hence made the (perhaps unreasonable) assumption that because it was new out of a packet an hour ago, it can't have a problem. Continuity has been tested from the point where it splices into the cat 6, down 3 floors to the controller end.

So my question is, is the resistance likely to be a problem? The controller has 0-5v written on it, and I recall reading that the signalling on the wires is something low voltage and serial, maybe the 4 cores are -v, +ve, tx and rx; I know that the controller powers off if cores 1 or 4 are disconnected, and shows a fault code if 2 is disconnected, nothing happens if 3 is disconnected, but I didn't check if the controller will operate the vent unit with #3 disconnected. The controller, even though it shows a fault code, seems to be capable of operating the vent unit. Note that the vent unit has its own equivalent LCD display and it never mentions any fault codes

Should I still be looking at a break or fault in the wire (as noted I haven't been able to test the whole wire), or could some amount (and what would be a high amount?) of resistance cause a problem akin to a disconnected wire?

Thanks!
 
Last edited:
Sponsored Links
Paralleled cores for data is asking for problems. Use just four cores. Serial can be run over network cable, although there is an impedance mis match. I have seen it run some distance.

The ventilation may not use serial, it may use its own data stream of some sort, which may not like the long distance.
 
NOTE; I couldn't test right from the 4P4C plug because my multimeter probe cant reach the contacts. I've hence made the (perhaps unreasonable) assumption that because it was new out of a packet an hour ago,

Use a sewing needle as a probe to reach the contacts of the 4P4C plug. Or buy a 4P4C socket to use as a test rig

dangerous assumption. Connect the controller to the 10 metre cable and see if the system works using the supplied plug and cable.
 
As above- check the thing out with the supplied cable first. I've had problems with serial data streams over long distance- thermometers which transmitted temp readings continually with an odd signalling arrangement (wasn't quite BCD but was weird). Anyway, to get reliable data at more than 30m I had to stick opamps in the lines to boost the signal (these things were tx only, your jobbie would be more complex).
 
Sponsored Links
The controller has 0-5v written on it,

RS 232 is +/- 12 volt signal with O volt ( minimum +/- 9 volts ) so unlikely to be RS232

It might be RS422 or RS485 which are differential 5 volt signaling and that should work provided the pairs are not crossed.

Or it could be something else.

Can you supply maker and type and maybe an installation / technical manual could be found.
 
Use a sewing needle as a probe to reach the contacts of the 4P4C plug. Or buy a 4P4C socket to use as a test rig

dangerous assumption. Connect the controller to the 10 metre cable and see if the system works using the supplied plug and cable.

He has tested it directly with the supplied cable. It works.
 
I read the original post as implying that the crimping was done before any testing


I took this to assume he had connected it directly.....

I know that the controller powers off if cores 1 or 4 are disconnected, and shows a fault code if 2 is disconnected, nothing happens if 3 is disconnected, but I didn't check if the controller will operate the vent unit with #3 disconnected. The controller, even though it shows a fault code, seems to be capable of operating the vent unit. Note that the vent unit has its own equivalent LCD display and it never mentions any fault codes
 
RS 232 is +/- 12 volt signal with O volt ( minimum +/- 9 volts ) so unlikely to be RS232

It might be RS422 or RS485 which are differential 5 volt signaling and that should work provided the pairs are not crossed.

Or it could be something else.

Can you supply maker and type and maybe an installation / technical manual could be found.
RS232 has almost no standards, I have worked on kit described as RS232 which only switched between 0 and +3.3V, it was easier to interface it with a 4 wire system (422 or 485) than RS232 working at +/-12.
Officially RS232 has quite a short limit on cable length (circa 10m I think) and sometimes it's just not possible to make it work enen that far without some sort of amplification or buffering. PS2 mouse or keyboard were infamous for this.
 
RS232 has almost no standards,
It does have a standard, quite a strict one. A lot of equipment described as being RS232 is not RS232.

RS232 is fully specified for its original intended purpose of providing data communication between a main frame computer and its peripheral units.

PS2 mouse or keyboard were infamous for this.
Mice and keyboards do not use RS232, they use a single ended serial protocol using a 0 to 5 volt range of signals.

This is a Texas Instruments document which provides the full specification

http://www.ti.com/general/docs/lit/getliterature.tsp?baseLiteratureNumber=slla037&fileType=pdf
 
It does have a standard, quite a strict one. A lot of equipment described as being RS232 is not RS232.

RS232 is fully specified for its original intended purpose of providing data communication between a main frame computer and its peripheral units.


Mice and keyboards do not use RS232, they use a single ended serial protocol using a 0 to 5 volt range of signals.

This is a Texas Instruments document which provides the full specification

http://www.ti.com/general/docs/lit/getliterature.tsp?baseLiteratureNumber=slla037&fileType=pdf
We used to work from a Sony spec as we had some sort of tie with them, I assume it matches the TI standards, which as you correctly state is written in quite a strict format, from memory (and I'm more than happy to be corrected if I remember any of this wrong from some 20 years ago) the standard is +/- 5 to 15V with no signal to exceed +25 or -20V .

Unfortunately manufacturers had different ideas and would come up with all sorts of ways to reduce their costs. Many of the systems we worked on used something close to the miniminal +/-5V. The 'no mans' area of +/-3V effectively became a nonsense as the 4 to 7 Kohm load impedance in the spec was frequently nearer to 1 or 2K, pulling the 5V down towards (or even below) 3V.

Some manufacturers used 0 to 5V for RS232 directly off TTL and we always seemed to find a way to make them all work together.

The PC's I used to build usually included a card with 4 RS232 ports employing MAX232 chips running at +/- 7V. We fitted plates with either 4 PS2 (6 pin mini DIN) OR 2 DB9 sockets and before you question it further we habitually used the mini DIN for RS232 without modification as they took up only one valuable space. Depending on how the PC was to be incorporated into the system and how short of slots we were, we sometimes used plates with 2 DB15's, one for VGA and the other for the 4 RS232's.

I can absolutely promise that mouse and keyboard PS2's frequently used to have +/- voltages present.
Furthermore keyboard/mouse could be randomly connected to RS232 or PS2 ports and any adapters contained no components.

The only other bit involved would be the software but I feel that is unlikely to be able to change voltages of the dedicated i/p - o/p chips.

I haven't really played with any of this for a long time and suddenly realise how rusty I have become, sadly my original note book disappeared the day before I was made redundant from that job.
 
employing MAX232 chips running at +/- 7V.

The MAX232 Meets or Exceeds TIA/EIA-232-F and ITU Recommendation V.28

My memory is fuzzy but recalls that the +/- 15 volt was eventually reduced when improved receivers ( i.e. integrated circuits ) became available to replace the circuits built from discrete transistors.

When designing interfaces into paging systems I came across many variations ( being polite ) of the RS232 standard. A Warden Call / alarm system linked its various components with ""RS232"" but their ""RS232"" did not work with our RS232 interface module. They used voltages 0 and +24. A single transistor and pull up resistor controlled the 24 volt to transmit and a Schmitt trigger referenced at 12 volts formed their receiver. Their atttude was ""RS232 works for us, it should work for you""

The good old days have gone, designing for minimum component count is a lost art
 
Last edited:
Do you have an oscilloscope?

If not, open up one end or the other and post a closeup photo of the circuit board.
 
dangerous assumption. Connect the controller to the 10 metre cable and see if the system works using the supplied plug and cable.

+1, found a break in one of the cores of the supplied cable. A new plug on it and we're away!

endecot; was it me you were requesting a photo from?
 

DIYnot Local

Staff member

If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.

Select the supplier or trade you require, enter your location to begin your search.


Are you a trade or supplier? You can create your listing free at DIYnot Local

 
Sponsored Links
Back
Top