IET logo
 
IET
Decrease font size
Increase font size
Topic Title: Surge Protection
Topic Summary:
Created On: 07 November 2017 09:27 PM
Status: Read Only
Linear : Threading : Single : Branch
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
 07 November 2017 09:27 PM
User is offline View Users Profile Print this message



Andy65

Posts: 10
Joined: 30 August 2013

Good evening ladies and gents.

I come seeking advice on SPD at a commercial property.

A brief description of the installation first which I'll try to keep it short.
The site comprises multiple buildings which have been added to and altered over the past 50 years. There are around 10 sub-mains feeding different areas from a central point. The central point is separate from all other buildings. The main building has LPS.

This site is used for product testing, has office spacewith socket, lighting and air conditioning, there is an IT testing floor and a couple of small server rooms.

I am looking for practical advice on what protection is required and where.

I gather that we need type 1 at the incomer and type 2 at the end of the submains (which feed busbar risers and other DBs).

I've know what BS7671 says and have spoken to manufacturers and technical help lines (who pretty much just read the regs to me but offered no practical advice)

The wholesaler reported a lead time of over a month to get a device so I guess that they are either in such high demand or more likely that nobody actually fits them so there are no products on the shelf!

I'd like some confirmation of what is actually required, preferred product manufacturers and how it relates to the LPS and Surge Protection BRitish Standard.

I keep it at that for now, feel free to ask Q's.
 08 November 2017 11:18 AM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 15990
Joined: 13 August 2003

What sort of supply do you have (any LV overhead)? and what's the AQ rating of your site (i.e. do you have more than 25 thunderstorm days/year there)? If the LV supply is entirely underground or you're on the UK mainland there's probably not automatic demand for SPDs. (Hence the wholesaler not having SPDs on the shelf - there's not the demand.)

You can opt to do a risk assessment instead which may indicate that SPDs would be required (which could take into account things like the likely costs of damage vs the costs of the SPDs or on safety systems) - but you'll need a fair bit of site specific information to go down that route.

- Andy.
 08 November 2017 12:41 PM
User is offline View Users Profile Print this message



gkenyon

Posts: 4954
Joined: 06 May 2002

Difficult question.

First, is the existing LPS to BS EN 62305-series, or BS 6651.

If it's BS 6651, the best thing to recommend is re-assessment of the lightning protection strategy against BS EN 62305, because you can't guarantee any SPDs to current requirements will achieve their aim.

Next, with the "multiple buildings" scenario - what are the site LPS zones, and how is other stuff that might transfer surges between buildings, or import them to the site, (like data and telephone network copper cabling) addressed?

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

Web-Site: www.gkenyontech.com
 08 November 2017 05:01 PM
User is offline View Users Profile Print this message



westom

Posts: 9
Joined: 24 August 2013

Protection is always about where hundreds of thousands of joules harmlessly dissipate. That means a transient current must not be anywhere inside a building.

That destructive current is hunting for earth ground. Appliances are damaged when in a path incoming (maybe from AC mains), through appliance, to earth.

An SPD is only as effective as the item that absorbs hundreds of thousands of joules. An SPD does not do protection. An effective SPD making a low impedance connection to earth ground. SPDs are only connecting devices. Actual protection is defined by that connection to and quality of single point earth ground.

Multiple buildings means each must have its own protection system. Every wire between each building must connect low impedance (ie less than 3 meter) to earth when leaving one building and again to the single point ground for that other building. That means every wire inside every cable.

If not done, then a lightning strike to one building acts like a lightning rod connected to electronics in that other building.

TV coax cable is hardwired direct to earth with a hardwire. No protector required. Telephone cannot connect direct to earth. So an SPD must make that connection to earth. An SPD only does what that hardwire does better.

A professionals Tech Tip demonstrates what is necessary even if wires are underground. Entitled "Coordinated Facility Protection":
https://www.erico.com/catalog/literature/E875B-WWEN.pdf

Protection is always about where hundreds of thousands of joules harmlessly dissipate. An SPD is only as effective as its earth ground.
 08 November 2017 06:29 PM
User is offline View Users Profile Print this message



Andy65

Posts: 10
Joined: 30 August 2013

Hi Andy and Graham

Thanks for your responses. I knew that this wouldn't be straight forward.

I have just taken over management of the contractor we use to maintain and inspect the LPS. I think that they are quite a large company and I feel that they get in and out quickly, complete the work on their sheet and that's it. I also question the knowlege of their operatives.

The LPS is 40 or more years old and don't think it complies with current standards. The site has been added and added to over the years, it seems by people with no idea.

Historically i think it's been a case of calling this company and leaving them to it hoping all would be fine. I am now looking deeper and have asked for a survey to be carried out with regard to adding to the system with SPDs.

I have obtained a copy of Bs62305 and I am aware that there is quite an in depth risk assessment to be carried out before determining the need for a LPS. I have asked the contractor to carry out this work and look at the need for surge devices.

We have two onsite 11kV transformers, one at 800kVa the other at 500kVa. Both now have with a TN earring arrangement. Sub mains between buildings are SWA and are generally buried although a couple are clipped direct or on cable tray. The LPS is not currently bonded to the Main Protective Bonding Conductors.

There is all types of cabling and pipework between the buildings.
 08 November 2017 06:35 PM
User is offline View Users Profile Print this message



Andy65

Posts: 10
Joined: 30 August 2013

Thanks Westo

What bugs me is that much if this hasn't been mentioned by the company we use for maintenance. I makes me question their desire for the job if not their expertise.

The other thing that springs to mind is WTF is the requirement for surge protection in the BS 7671 if it is such a complex and bespoke subject? It's almost a passing thought in the regs yet fitting SPD is not clear cut.
 08 November 2017 09:00 PM
User is offline View Users Profile Print this message



westom

Posts: 9
Joined: 24 August 2013

Originally posted by: Andy65
What bugs me is that much if this hasn't been mentioned by the company we use for maintenance. I makes me question their desire for the job if not their expertise.


First determine if a standard / recommendation is only about protecting human life or about electrical equipment. Most standards only address human safety issues. Then many, who do not know the difference, will muddy the waters citing speculation and subjective reasoning (no numbers) as if fact.

Second, some fundamental EE principles apply that sparky is never taught. Those principles are irrelevant to 50 Hz electricity that he learned.

For example, connect a 200 watt transmitter to a long wire antenna. Touch one part of that antenna to feel no voltage. Touch another part to be shocked by over 100 volts. Why two completely different voltages on the same wire? What sparky learns (and what so many layman assume) is that 100 volts should be everywhere on that wire. But fundamental electrical concepts including impedance and electromagnetic theory result in a completely different conclusions.

Surge protection involves radio frequency electricity. So ham radio operators knew (should understand today) these concepts that are completely foreign to sparky.

Third, application notes from Polyphaser were legendary for explaining these concept in layman language. Unfortunately IT have no idea what is always necessary for technical people. They keep changing the URL addresses. So we can never cite a relevant document - that URL changes maybe annually.

One current app note (hurry before these fools change the address again) is: ... apparently this site cannot accept URLs - you must reconstruct it:
http colon slash slash triple w dot pdicorp.com slash SiteMedia/SiteResources/WhitePapersandTechnicalNotes/1485-019.pdf?ext=.pdf

Protectors are simple, dumb science. Despite what the naive assume, no protector does protection. Protector is only a connecting device to what does protection. A protector must be sufficiently size to first not threaten human life. Then must be larger to not fail during any surge for decades. Then it can connect thousands of amps to another item that harmlessly dissipates that energy - single point earth ground.

Single point earth ground is a concept explained by something called a Faraday cage. That ground must address equipotential and conductivity. Each structure (including buildings, muntions igloos, radio towers, airport tarmac, etc must have / provide its own single point earth ground. Wall receptacle safety ground obviously is not earth ground (despite what sparky assumes). That is the art of protection.

Some case studies demonstrate how it is done. Orange County FL suffered repeat surge damage. They finally brought in someone who knew these basic electrical concepts. Only thing fixed was the earthng - no protectors used or needed: ... again construct it:
http colon slash slash triple w dot psihq.com slash AllCopper.htm

A Nebraska radio station suffered repeated surge damage. Station engineers were so ill informed as to actually remove earth grounds. This solution even addressed the 'primary' surge protection layer out at a utility pole: ... again reconstruct it:
http colon slash slash triple w dot copper.org slash applications/electrical/pq/casestudy/nebraska.html


Four, this stuff has been well published and understood for over 100 years. Both GE and Westinghouse did much research on the Empire State Building in its first years (1930s). That structure would suffer about 23 direct strikes annually to electronics. And still, a majority cannot discuss and do not understand what is required. Too many only recite hearsay, advertising, speculation, and other urban myths. All quickly identified by what is not provided and what must always exist in every informed recommendation - numbers.

I appreciate your frustration. Hopefully this explains why so much confusion and so little informed information exists. And helps to separate wild speculation from informed concepts.

Edited: 08 November 2017 at 09:14 PM by westom
 08 November 2017 10:51 PM
User is offline View Users Profile Print this message



AJJewsbury

Posts: 15990
Joined: 13 August 2003

single point earth ground.

But it isn't necessarily a single point earthing (in UK practice at least) is it? PME obviously supplies many parallel connections to true earth and even a local TN-S system from a private transformer would have the LV earthing electrode plus (normally) the lightning electrode in parallel, and both in parallel with bonding to any extraneous-conductive-parts. From the installation's point of view it's less worried about the absolute voltage above true earth, more about the voltage difference between conductors and its own earth reference (i.e. the main earth terminal).

The forum is quite happy with URLs - e.g. http://www.pdicorp.com/SiteMed...icalNotes/1485-019.pdf (although as a new poster you might find that all your links aren't displayed by default - but established users can re-post them for you easily enough and you'll get the privilege in due course - nothing personal, just one of the defences against spammers).

The other thing that springs to mind is WTF is the requirement for surge protection in the BS 7671 if it is such a complex and bespoke subject? It's almost a passing thought in the regs yet fitting SPD is not clear cut.

That's not too far out - there's little to no tradition of using SPDs in the UK and the current regulations don't really change that - you have the option of basing the decision to use SPDs on the number of thunderstorm days - which as far as I know is never a met anywhere in the UK. So you can forget about SPDs and still claim full compliance with BS 7671. That might not, of course, meet your/customer requirements, or where BS 7671 is applied outside of the UK (there are quite a few parts of the world that do) so using SPDs is (now) covered in BS 7671. (There are suggestions that this might change somewhat with the next edition of BS 7671, but that's all conjecture for the moment).

- Andy.
 09 November 2017 06:20 AM
User is offline View Users Profile Print this message



Andy65

Posts: 10
Joined: 30 August 2013

Thanks Andy, I am in th U K. I'm not in the slightest bit worried about what the requirements for other countries are as BS7671 is enough of a riddle for me.
As I say we have well over 30 DBs dotted around the building, one, which was fitted in a roof plantroom 2-3 years ago has randomly got type 2 SPD fitted. No explanation as to why (no, before you ask it's not mentioned on the EIC! silly me what EIC) This was an approved contractor who did work here for 10 years, EICRs and missed that the main panel in the intake room had no earth connection to the transformer (there were other contractors who did large jobs and missed it too). The PIRs showed either TNS or TNCS for earthing system type and the infamous N/A for the particulars of the installation including ZE. It had been like this for 40 odd years! It took 5 months for UKPN to come and TNS us!

Thanks all for you replies, I shall await the outcome of the LPS company's assessment. I shall be looking for another contractor next year to look after our system.
 09 November 2017 12:59 PM
User is offline View Users Profile Print this message



westom

Posts: 9
Joined: 24 August 2013

Originally posted by: AJJewsbury
But it isn't necessarily a single point earthing (in UK practice at least) is it?

A critical expression for protection which is irrelevant to sparky is impedance. Electricity is never same at both ends of a wire. That difference is so little (for protecting humans) that sparky is never taught it. But that difference is major for protecting hardware (appliances).

'Type' is another term only relevant to human safety. It says little in terms of appliance protection.

TNS and TNC have an earth ground at the transformer. That means a structure has no single point earth ground (no hardware protection). TNCS can provide that earth ground. Can but may not due to a critical numbers - low impedance (ie less than 3 meters).

Every wire in every incoming cable must make a low impedance (ie hardwire has no sharp bends or splices) connection to the one common ground for the entire building. That should be obvious from a first provided figure that shows how all interconnected buildings must be earthed.

Single point earth ground can be a group of earthing electrodes. Or it can be a copper wire encircling the building. Ufer ground has long been a preferred ground because it uses an electrical conductor - concrete footings - to surround the entire building with a single point ground.

One example of a contractor without grasp is that Type 2 protector located in a roof plantroom. It has no low impedance connection to earth. And its earthing cannot be a a single point electrode. So it does virtually no protection. Or may even make damage easier. It might conduct metallic mode currents. But those are not destructive currents. Concern is for longitudinal mode currents. Additional relevant factors that are completely ignored in standards that only address human protection; do not address hardware protection.

Potentially destructive surges occur maybe once every seven years. And less often in the UK. However that number can vary significantly even in a same town. Geology is one major reason why.

A potentially destructive surge once every decade is why we spend no money on plug-in protectors. And spend massively on the 'whole house' solution. That massive expense is tens of times less money compared to what is spent on plug-in (point of use) protectors.

A plug-in protector is ineffective (can even make damage easier) if a 'whole house' solution does not exist. Yes, a plug-in protector without that 'whole house' solution can compromise what is superior protection already inside every electrical appliance.

How to make a protector more effective? Increase separation between a protector and hardware. Decrease separation between protector and single point earth ground. Impedance is relevant. Increased separation increases impedance; increases appliance protection.

Ignore concepts such as 'Type'. Appreciate why every meter of wire, every splice, every sharp bend, and metallic conduit seriously increases impedance (and hardly affects resistance).

Earth ground at a transformer can be the 'primary' protection layer. A completely (and electrically) different earth ground at the structure's service entrance is a 'secondary' protection layer. Safety ground in a wall receptacle is all but disconnected from earth (due to impedance) - does not provide a layer of protection. A plug-in protector has no earth ground - is not a layer of protection. For same reason, a protector in the roof plantroom does virtually no hardware protection (if a Type 1, 2, 3, or 4 protector).

If using a TNS or TNC system, then a single point earth ground must be installed at the service entrance. A neutral wire is not hardwired (does not connect direct) to earth. So it must connect to that earth ground via a protector. Every incoming wire must connect to that earth ground - either directly or via a protector. That is the 'secondary' protection layer.

A protector only does what a hardwire does better. TNCS systems can provide that single point earth ground only if installers are taught how to exceed human safety code requirements. That means even BT's wires, any TV coax cable, satellite dish, or OTA antenna wire must remain outside until it enters at the service entrance. Only then can it make a low impedance (ie less than 3 meter) connection to single point earth ground.

If the 'whole house' solution does not exist, then do not waste money on plug-in protectors. Those can only compromise (ie bypass) what is better protection already inside all electronics.
Statistics

New here?


See Also:



FuseTalk Standard Edition v3.2 - © 1999-2017 FuseTalk Inc. All rights reserved.

..