Would you run Cat 6 armoured alongside SWA in garden?

[Vortigern]

It is just convenient to cable tie the CAT6 cable onto the SWA. Does anyone think I would get significant signal noise? I have seen wads of cat5 normal lashed over suspended ceilings and fluors and never heard of anyone complaining about signal loss and that is over hundrends of units collectively. Not really interested in the separation thing as it is at home for my garage cctv.

 

[baldelectrician]

Use an armoured Cat 5/6 or even duct grade C6

 

[Vortigern]

Yeh it will be armoured for outdoors use. Bear in mind I would not do this on a job, it's only because it's my own house.

 

[Intoelectrics]

It is just convenient to cable tie the CAT6 cable onto the SWA. Does anyone think I would get significant signal noise? I have seen wads of cat5 normal lashed over suspended ceilings and fluors and never heard of anyone complaining about signal loss and that is over hundrends of units collectively. Not really interested in the separation thing as it is at home for my garage cctv.

I've seen it done this way many times on garages and outbuildings and never known an issue with signal noise. I don't know enough about it though to say either way. Would it make a difference if the swa was a 3 phase supply not just single phase?

 

[plugsandsparks]

Have loads of SWA close to Cat6 at duct entry points and the like. Unless you are pushing the max distance, i doubt you will see any signal loss.
You can always check by running Ping off your laptop or whatever, no packets loss = happy days, lol

NSFW...... ha

 

[Vortigern]

SWA is 4mm 4 core left over from another job around 20m run. Just a 16a sockets and 6a lights (LED) just run some power tools occasionally so not much juice flowing through the old copper.

 

[Lucien Nunes]

never heard of anyone complaining about signal loss

The term 'signal loss' is confusing here. All data cables experience loss (meaning electrical attenuation) and over long distances it is very significant. Over 100m of Cat6 at 100Mhz, only 1% of the transmitted signal power reaches the far end, the remaining 99% is dissipated in the cable. But this is still in-spec and will not itself cause packet loss (meaning unreadable data) because although the signal is weak, it can still be clean. However for the same level of interfering noise, the signal-to-noise ratio will be 100 times worse with that level of attenuation present, compared to a short hop; so there is more likelihood of that interference corrupting the data and causing packet loss.

50Hz mains itself doesn't interfere much with balanced twisted-pair data. The data cable twist is effective at low frequencies and it's miles out of band as far as the receiver is concerned. What is more likely to interfere is mains-borne RF that is produced by current-using equipment. How severe that is and over what frequency spectrum will depend on the equipment. There is a degree of shielding from the armour but at higher frequencies the rejection by the cable construction is reduced. So you have a lot of variables; the data signal attenuation, the interference level and its spectrum, the interference rejection by the cable and its spectrum. But one thing is very predictable and that is the inverse square relationship between the distance between the cables and the degree of coupling. So 10cm apart compared to strapped together could be 100 times better at preventing interference. That is why it is a general strategy to use that spacing as a safeguard, making the other, unknown parameters less critical.

Just a 16a sockets and 6a lights (LED) just run some power tools occasionally so not much juice flowing through the old copper.

The 50Hz current itself is not a major consideration, but the interference from brush motors in power tools might be. They are notorious for radiating hash in all directions.

You can always check by running Ping off your laptop or whatever, no packets loss = happy days, lol

I'm not sure that's a good test. It could be just on the edge, and subject to failure with a slight increase in mains-borne noise or long-term drift in characteristics.

 

[plugsandsparks]

TBH Lucien, would be amazed he experiences any issues, TCP will deal with most of the issues you describe by re-transmission and UDP like video/audio will show up as noise in the video / audio quality. His run say is 30M so well within S/N ratio. Let say he is running 1 Gb/s link speed on cat6 - he can always drop to 100Mb/s and still not see any difference in performance

Come on... you know it will work

 

[Lucien Nunes]

Indeed, and I didn't actually answer the question. Yes I agree, 20m will work fine. I was just setting the scene a bit as to what is likely to make a difference and what not.

 

[pc1966]

Generally speaking it takes a lot of adjacent noise to disrupt even unscreened twisted pair cables, and with both the CAT-6 cable and the power cable having separately earthed shields I would not expect any problems.

However, as Lucian points out you do have a length limit for reliable networking and 100m is about it. Unless I needed copper cables for a good reason, say powering a PoE camera at the end, I would run fibre as then no realistic length limit and these days the fibre SFP and switches are not that expensive.

The main objection to running both in the duct is the usual rules/guidance for separation of power and ELV circuits, but again with earthed armour (or shields) there is no realistic risk of a power cross.

 

[Vortigern]

I suppose the good news is, if I am running power tools I wont really care what the internet is doing. I think you have comprehensively answered any unasked questions @Lucien Nunes even the ones I asked as a bonus. But in other words, as I suspect a person like me and my requirements would not appreciate any significant interference. Q.E.D. To be more precised the whole run is circa 30m with 18m piggy backing on the swa. The cable (cat) is screened and pure copper not cca etc.

 

[plugsandsparks]

Or stick an Wireless Access Point in the bedroom window. Funnily enough i am about to do the same thing for a cabin in the garden, similar distance. Currently in two minds whether to use an Outdoor access point and use Wifi (not sure about the TV having good enough wifi capability) or run a cat6 in with the power (SWA) and then put an access point in the cabin. Either way i want my laptop to work off wifi so an access point is going to be required to get good speed at 5Ghz

 

[Eric12]

As long as its earthed from a different source from the power (with a rod for example if power is fed via a tn system) then this prevents any interference in the cat cable. That's what I read somewhere at least !

 

[Lucien Nunes]

Don't trust everything you read! A clean local earth isolates only one of many possible sources of noise (that carried by the earthing system of the power installation) and wouldn't affect interference between cables strapped together. Its advantages at frequencies of interest here are likely to be negligible. However, with a TN power system present it introduces an unbonded extraneous conductor which could lead to shock risk, and when interconnected with the system earth could result in fault currents diverting via data shields and causing equipment damage if the Ra is low enough.

 

[sparkymark77]

I did do a job with a cabin at the end of the garden where they wanted wifi too.
The run was around 100 metres and both power and cat 5 swa’s were moled under the garden in the same hole for about 80 metres. I’m not saying it’s good practice but there was no way the client was paying for two lots of moling!
When I checked at the end all seemed to work fine with decent speeds and I’ve not had any complaints and that was 3 years ago!