IET
Decrease font size
Increase font size
Topic Title: Separation of Band I and Band II circuits
Topic Summary:
Created On: 21 April 2009 01:45 PM
Status: Read Only
Linear : Threading : Single : Branch
1 2 Next Last unread
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 21 April 2009 01:45 PM
User is offline View Users Profile Print this message



Platypus

Posts: 465
Joined: 20 January 2005

Hi,

I'd be grateful for some guidance here please.

I'm using wire basket tray for cable containment for the main trunk runs in my house, because it's cheap, fairly easy to work with and affords basic containment with good cooling (I hate slinging cables across joists loose - and it's a good way to minimise risk of cable damage when I work on some of the roof timbers next year).

Now, I've got a couple of roof voids that are narrow where I'm aiming to take a 200x35mm basket and a 100x60mm basket down.

200 basket contains mains, nicely spaced out and 100mm basket will contain random mixed Band I (network cables, SELV heating control cables, aerial coax etc).

Practical constraints are showing that's I'll be lucky to get more than 10-20mm of seperation between the trays. Trays will be bonded to MET.

I can't pin down a presciptive regulation(s) indicating of this is acceptable. We'll have to assume that not all Band I cables have insulation to the higest voltage in the Band II system.

Section 528 of the 17th state that a seperate trunking system is adequate. So on that basis, would it be fair to consider basket tray as a "conduit, trunking or ducting system"?

Reg 414.4.1 part ii states that SELV and PELV circuits could have protection by means of double insulation as a sufficient means of protection from LV circuits, so it also appears that as the LV circuits will be in 6242Y T+E, then we may consider that as sufficient in itself.

Over-worrying probably, given the number of public sector sites I've worked in where Band I and II are slung in a big heap under a false floor, but I'd like my installation to be by the book, and to be clear on why it is.

Cheers

Tim
 21 April 2009 06:15 PM
User is offline View Users Profile Print this message



micjamesq

Posts: 777
Joined: 23 January 2009

A lot of the information for this sort of thing, where data cabling is used, comes under the title of EMC and prevention of mutual detrimental influences.

BS 7671 includes basic regulations for EMC WRT separating system voltages and the old favorite, 515 "Prevention of mutual detrimental influences". As you have noted in your original post section 528 - proximity to other cables, equipment provides basic guidance in this situation and as mentioned above 515 provides a little bit more.

The EMC directive uses HD 60384 which in turn BS 7671:2008 is based. The EMC directive and other stuff can be found here:

Electromagnetic Compatibility (EMC) Directive 2004/108/EC

If you do not want to read through all the bumph I will give you a little summary here:

First of all I can say, which may help you a great deal, is: for horizontal cabling where the final circuit length is less than 35m no separation is required in the case of screened cabling (both power and data) provided that 528.1 is adhered to. Personally I would not recommend this, which bring us onto the following:

Installation Type: - :Minimum Separation Distance: - :Without Divider: - :Aluminum Divider: - :Steel Divider

Unscreened power cable and unscreened data cable: 200mm : 100mm : 50mm
Unscreened power cable and screened data cable: 50mm : 20mm: 5mm
Screened power cable and unscreened data cable: 20mm : 10mm: 2mm
Screened power cable and screened data cable: 0mm : 0mm : 0mm

Note: Screened data cable must comply with EN 50288

Hope you can understand the above effort

Ideally, where practicable, it would be beneficial to try and run the different cables in separate trays, but I think the above table should give you a little bit of wiggle room.

Hope this helps

Regards

-------------------------
E & OE
 21 April 2009 10:08 PM
User is offline View Users Profile Print this message



Platypus

Posts: 465
Joined: 20 January 2005

Thank you for a very informed reply.

The cables will be in separate baskets - it's just I can't get those very far apart in places. But there's no reason I couldn't bolt a steel divider between them, especially as I've just found an internet company that sell metal stock and sheet cut to order at reasonable prices. In fact, just hanging a shield off the side of the comms tray on the back of some fastener bolts would be very simple.

My network cables *might* become STP - undecided yet - that's far off.

So if I'm understanding this - most of the distance and shielding is to do with EMC, not safety?

I can live with breaching EMC guidelines a bit if needs must (in practise I've never seen much of an issue with Cat5e data cabling.

I'll go and look at EN50288 if I can find a text somewhere - I will have some shielded RS485 comms for random computery widgets.

Again, many thanks

Tim
 22 April 2009 12:46 AM
User is offline View Users Profile Print this message



spinlondon

Posts: 5494
Joined: 10 December 2004

Any current-carrying conductor, radiates an electromagnetic field.
A conductor that is not energized will pick up energy from any existing electromagnetic field around it. This can cause induced voltage which can cause damage to equipment or other parts of the installation.
It can also cause noise on data cables which may mask the signal carried by the cable.
Using earthed screened cables helps reduce higher frequency electromagnetic field interference by acting as a sort of Faraday cage. Isolating the inner cables from external electromagnetic fields. Remember only earth one end of the shield.
However this will not help against the lower frequencies transmitted by transformers and their associated wiring.
Using Twisted pair cables such as Cat5 etc. can help with the lower frequencies.
The electromagnetic field is cancelled by each twist.
Using Ferrite chokes around power cables at the equipment end of the cable, especially those plugged into sensitive equipment such as Amplifiers, Computers etc. can also help.
You can get a pretty good idea of the extent of the magnetic field around a conductor by holding a Volt Stick close to the energized conductor. The greater the voltage range of the Volt Stick, the more sensitive it is.

Edited: 22 April 2009 at 12:48 AM by spinlondon
 22 April 2009 01:19 AM
User is offline View Users Profile Print this message



Platypus

Posts: 465
Joined: 20 January 2005

Hi,

Thanks. I know about those effects. So what we're saying is the fairly onerous BS7671 regs regarding separation distances are more geared towards EM coupling, and not safety per se?

It's the regs I'm trying to get a handle on.

On one hand they say the separation, or insulation requirements are to prevent dangerous voltages appearing in comms and other ELV cables - fair enough.

OTOH, they also seem to be addressing "detrimental influences".

Long and short - is I'm happy enough to do what I can, but bend the rules on detrimental influences. It's a house, not a medical facility and we need to be practical. But I won't bend the rules regarding safety related matters.

Unless someone starts banging massively high rates of change of current though the LV wiring, nothing I have is going to be much worried even if it's lying on top of the LV cabling, especially as most of my signal circuits are either balanced or "heavy-signal" (relay drives) and distances are short (parallel path max length typically 8m, then the cable groups diverge).

It would be nice if the regs had "MUST" and "SHOULD" clauses, rather like computer protocol RFCs

Thanks anyway

Tim
 22 April 2009 01:30 AM
User is offline View Users Profile Print this message



spinlondon

Posts: 5494
Joined: 10 December 2004

Sorry, wasn't trying to teach you how to suck eggs.
The Chapter relating to Electromagnetic effects is very new. They are probablly still working it out.
In the past I've seen heaps of SWA power cables just dumped on top of screened control cables we'ed installed on tray, with no ill effects.
When I've run Cat5, I've allways kept it seperated from power. In dado trunking the physical distance between the Data and power is only a couple of inches at most. I have been told that it is alright to run Data alongside power for up to a meter. Never had to do that, so I don't know for sure.

Edited: 22 April 2009 at 01:32 AM by spinlondon
 22 April 2009 10:39 AM
User is offline View Users Profile Print this message



Platypus

Posts: 465
Joined: 20 January 2005

Originally posted by: spinlondon

Sorry, wasn't trying to teach you how to suck eggs.



Quite alright sir - you don't know me from Adam

I appreciate the long reply anyway.



The Chapter relating to Electromagnetic effects is very new. They are probablly still working it out.

In the past I've seen heaps of SWA power cables just dumped on top of screened control cables we'ed installed on tray, with no ill effects.

When I've run Cat5, I've allways kept it seperated from power. In dado trunking the physical distance between the Data and power is only a couple of inches at most. I have been told that it is alright to run Data alongside power for up to a meter. Never had to do that, so I don't know for sure.


Come to think of it, when I was doing sysadmin work for Imperial College, we had miles of dado trunking with power and Cat5e. Most of the comms compartment was stuffed full (2 drops min to every desk, sometimes 4). So the Cat5e (unshielded) and phones where practically touching the mains for 20m runs, save for the PVC divider wall. All of our network stuff was TDR tested in house for speeds to 1 gigabit/sec (and the odd bit of copper to 10gig) and no problems ever showed up, other than the odd faulty termination.

It was our computer room that had a ton of spaghetti under the false floor and that worked fine, including RS232 - over the years it was gradually replaced and the new stuff ended up on trays, more neatly.

One wonders if "they" aren't perhaps worrying too much.

Oh well. Might just put it down as a departure on the EIC then. Common sense needs to play a part.

Cheers

Tim

Edited: 22 April 2009 at 10:40 AM by Platypus
 22 April 2009 12:17 PM
User is offline View Users Profile Print this message



kalookicedogs

Posts: 181
Joined: 23 October 2008

Are we getting separation confused with segregation?

-------------------------
My ex wife was an amazing house keeper. Problem is I never got the damn thing back.
 22 April 2009 04:54 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 17795
Joined: 13 August 2003

Remember only earth one end of the shield.

The latest version of BS EN 50310 ("Application of equipotential bonding and earthing in buildings with Information Technology Equipment") seems to have reversed the ideas on that one and now asks for cable screens to be bonded at both ends!

- Andy.
 22 April 2009 10:53 PM
User is offline View Users Profile Print this message



spinlondon

Posts: 5494
Joined: 10 December 2004

I think bonding at both ends causes some earth loop feed back.
 22 April 2009 11:37 PM
User is offline View Users Profile Print this message



nickperkins

Posts: 162
Joined: 25 July 2008

Please note text from a previous post

BS 6701:2004:
4.4.4.3.4 For electricity supply cabling installed inside buildings and operating at nominal voltages between 50 V a.c. and 600 V a.c. (120 V d.c. to 900 V d.c. to earth), the telecommunications cabling and electricity supply cabling shall be separated by either:
a) a distance of not less than 50 mm; or
b) a divider meeting the requirements of BS 7671;
unless one or more of the following conditions is met:

NOTE 1 If one or more of these conditions is met then separation is not deemed to be necessary.

1) the electricity supply cables are enclosed in a separate conduit or trunking which, if metallic, is earthed in accordance with BS 7671;
2) the electricity supply cables are of a mineral-insulated type;
3) the electricity supply cables are of an earthed armoured construction;
4) the electricity supply cables are of a flexible double-insulated type (e.g. "kettle leads"
supplying 240 V mains power to telecommunications equipment in cabinets).

NOTE 2 Standard electrical 240 V "twin and earth type" cabling is not flexible double-insulated.
However there are lots of other relevant standards on this and the ECA recommend 150mm between twin & earth and unscreened signal cable.


Nick
 23 April 2009 10:54 AM
User is offline View Users Profile Print this message



kalookicedogs

Posts: 181
Joined: 23 October 2008

Originally posted by: spinlondon

I think bonding at both ends causes some earth loop feed back.



As the frequencies used for signals, and the frequencies present in the electromagnetic ambient continue to increase, the requirement for having few very small holes or gaps in the overall shielding of a product (including its enclosure, connectors, glands and cables) gets tougher.

Many years ago it used to be normal practice in some industries to only bond cable shields at one end. This was because the poor circuit designs typically used in those industries forced any cable shield noise currents to flow in their circuits. But a cable shield that is only connected at one end provides no useful shielding for signals and noise at wavelengths shorter than six times the length of the cable (for instance: a cable 5 metres long would not provide useful shielding at greater than 10MHz if its shield was only RF bonded at one end).

Equipotential bonding of mechanical protection within cables should not be confused with earthing of screens or shield for EMC purposes.

-------------------------
My ex wife was an amazing house keeper. Problem is I never got the damn thing back.
 23 April 2009 11:30 AM
User is offline View Users Profile Print this message



spinlondon

Posts: 5494
Joined: 10 December 2004

Unfortunately the majority of instruments used for BMS wouldn't allow for the bonding of the screen.
 23 April 2009 12:17 PM
User is offline View Users Profile Print this message



kalookicedogs

Posts: 181
Joined: 23 October 2008

Originally posted by: spinlondon

Unfortunately the majority of instruments used for BMS wouldn't allow for the bonding of the screen.


Probably no need. As I said in my previous post the higher the frequency (shorter wavelength) then the greater the need for screening with minimal gaps and holes. A 4-20mA signal back from say an actuated valve or pressure transducer would be relatively low frequency. Screen termination would generally be offered within the control panel only.

-------------------------
My ex wife was an amazing house keeper. Problem is I never got the damn thing back.
 23 April 2009 12:22 PM
User is offline View Users Profile Print this message



gkenyon

Posts: 5354
Joined: 06 May 2002

Originally posted by: spinlondon

Unfortunately the majority of instruments used for BMS wouldn't allow for the bonding of the screen.
This is a big problem - what "screen" are you talking about?

The "Overall screen" needs to be bonded at both ends, unless there's a very good technical reason not to (e.g. cable goes outside the building), and it's best to use a 360-degree bond on to a metal gland-plate wherever possible.

The foil screens (with or without drain) over individual pairs is another thing entirely.


"Earth Loops" - another issue again, and there's more than one type of earth loop, each with their own "remedy".

If your installation conforms to BS EN50310, then in general bonding over-screens, over-shields, armour, etc., at both ends (or indeed multiple points along a very long run) does not normally cause a problem.

-------------------------
EUR ING Graham Kenyon CEng MIET TechIOSH
G Kenyon Technology Ltd

Web-Site: www.gkenyontech.com
 23 April 2009 12:54 PM
User is offline View Users Profile Print this message



spinlondon

Posts: 5494
Joined: 10 December 2004

When I did the BMS install at southend college, I think there were 10 AHUs 9 on the roofs.
Every control cable was a mesh screened Belden type. not certain whether the screen was steel, aluminium or tinned copper. The termination of the screen was acomplished by cutting back the insulation, pulling out the cores through the side of the screen, sleeving the compacted screen, attaching a bootlace ferrule and terminating.
I don't recall any of the instruments or actuators haveing an earth terminal, where the screen could be terminated.
When putting away cables where pairs are foil screened on other jobs, on some we were instructed trim back the foil, sleeve the drain and terminate where there was an earth terninal. If no earth terminal was present, we were to trim back the drain as well.

Edited: 23 April 2009 at 01:00 PM by spinlondon
 23 April 2009 06:32 PM
User is offline View Users Profile Print this message



gkenyon

Posts: 5354
Joined: 06 May 2002

In some conditions, what you were told to do can create aerials for transmitting, or receiving.

This type of overall meshed screen is best terminated at both ends, and preferably on the clang-plate, rather than bringing the screen into the enclosure, but if the casings of the equipment are non-metallic, then that's another issue.



But to be honest, I see all sorts of specifications that tell the installation contractor to do the OPPOSITE of what's actually considered "Best Practice".

One example was a requirement to ensure all additional earth-bonding provided to containment and cubicles for IT and comms equipment, provided for EMC purpoeses, was "coiled to ensure neatness, and provide strain relief for doop openings and where earth cables leave containment".

Of course, this just makes an air-cored inductor, so preventing the earth connection from conducting high freqency noise to earth. i.e. EMC earthing won't work.

I've also seen this trick done on surge suppressors/arrestors - similar result: ineffective suppression, and fried kit!


Don't believe everything you read in a specification, and beware that under various legislation you have obligations to ensure your work "complies", certainly for safety and EMC, regardless of what's written in the spec.

-------------------------
EUR ING Graham Kenyon CEng MIET TechIOSH
G Kenyon Technology Ltd

Web-Site: www.gkenyontech.com
 23 April 2009 09:33 PM
User is offline View Users Profile Print this message



Platypus

Posts: 465
Joined: 20 January 2005

Just to be boring - does T+E class a double insulated? I would have said yes based on common sense (well, you lean on it if it were running up a wall and that (in the absence of earthed metallic shielding) pretty much indicates it's Class II (OK it's not an appliance, but you get my drift).

But I keep seeing vague suggestions it's not considered so...

I read the relevant BS7671 regs again and again whilst waiting for EDF to come and put my cutout fuse back today and am still getting confused.

Ignoring EMC issues and looking purely at saftey:

If it's OK to run singles in one compartment of dado trunking, with a bit of PVC wall between then and a load of comms cables of all types, why isn't it OK to stick a load of comms cables next to (or on top of for that matter) a load of T+E. You'd lean on T+E without fearing for your life - are comms cables faint hearted?

The reason I say, for the sake of argument "ignore EMC" is, for non critical applications, degraded or random errors in comms cables won't end the world - but mains appearing in them might.

But it is OK to stick comms cables next to LV flex supplying a cabinet of computer equipment...

It just seems a bit inconsistent - or perhaps overstated in places.

?...

Cheers

Tim

Edited: 23 April 2009 at 09:34 PM by Platypus
 24 April 2009 08:54 AM
User is offline View Users Profile Print this message



kalookicedogs

Posts: 181
Joined: 23 October 2008

Originally posted by: Platypus

Just to be boring - does T+E class a double insulation?



Tim



No. T and E coverings are not classed as double insulation. The covering on the copper is the insulation. The overall covering is classed as the sheath and whislt offering insulationg properties, it should not be classed as insulation.

-------------------------
My ex wife was an amazing house keeper. Problem is I never got the damn thing back.
 24 April 2009 09:57 AM
User is offline View Users Profile Print this message



Platypus

Posts: 465
Joined: 20 January 2005

Originally posted by: kalookicedogs

Originally posted by: Platypus



Just to be boring - does T+E class a double insulation?

Tim


No. T and E coverings are not classed as double insulation. The covering on the copper is the insulation. The overall covering is classed as the sheath and whislt offering insulationg properties, it should not be classed as insulation.


Thanks - that was how I originally understood it, 10+years ago. Wondered if it had changed (eg meter tails are referred to as double insulated, whereas 20 years ago, they woul dhave also been called "sheathed".

wibble...
Statistics

New here?

  • To participate in discussions, please log in and introduce yourself.

See Also:



FuseTalk Standard Edition v3.2 - © 1999-2022 FuseTalk Inc. All rights reserved.

 
..