Unconventional termination of SWA

Talk Electrician Forum

Help Support Talk Electrician Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Again though, give me a reason why this is bad practice, other than you don't like it. I agree, made off badly it would look a mess. None of my work ever looked a mess, it was always exceptionally neat. Why is a direct connection to an earth block worse than going through several connections to get there? It is neither electrically, or in the case of being inside a plastic enclosure, entering from the rear, aesthetically worse than the official method of termination? Remember I'm only advocating this in the case of small cables in plastic enclosures in a residential situation where the official method isn't practical, or would aesthetically look a mess. Using this method, it can be done exceptionally neatly.


You state you are talking about small cables in residential situations. So I am assuming this CU you are terminating into, as it is plastic, and I guess other cables are standard PVC T&E?  is not actually installed in an area at risk of high impact, or other serious external influences? As such the SWA could easily have been terminated somewhere more suitable externally and the final length of cable brought into the CU as PVC.  As a small PVC cable will always be easier to dress and terminate into a small residential board and a single copper CPC will be far easier to work with in a confined space than multiple twisted steel strands of armour butchered into an earth bar. Other than a bit of laziness and possible speed to get away from the job ASAP on the installers side, there is no advantage to your method. Obviously if you can link a manufactures guidance where this method of terminating a SWA is considered acceptable then I am open to correction. But until such times there is no practical benefit to your solution. Additionally when it comes to periodic inspection and testing or any fault finding work where you may need to open up your twisted armour termination, it will be far more time consuming than a standard copper conductor. Your earlier comment about multiple connections being less safe due to HR joints etc. is irrelevant. Excluding the most basic of final circuits, you will always have multiple joints, or shall we limit light circuits to single luminaires just in-case we get a HR joint in the circuit? This is one of the key reasons we do dead testing before energising a modified circuit. Any competent person will be able to correctly terminate all cable types and ensure an electrically and mechanically sound joint in accordance with manufactures and industry standard guidance. It all pivots back to the "good workmanship and proper materials" guidance. 

Doc H. 

 
Last edited by a moderator:
@ Sidewinder,

Just looked up most of those regs, pretty subjective stuff to be honest...

@ Murdoch, Stop being obnoxious, who are you to call me a chancer, If you don't want to have a polite discussion, mind your own business. 

 
Last edited by a moderator:
Sure ................ but clamp down on the chancers ................................................but you NEVER do


You do not know what works are done by the horsemen, and, it is not of your concern.

If you turn this into a debate or slanging match, then you will find you will be on holiday from here.

If you wish to discuss forum moderation, or the like, use PM.

This is the last message on open forum discussing this.  Fact.

 
You state you are talking about small cables in residential situations. So I am assuming this CU you are terminating into, as it is plastic, and I guess other cables are standard PVC T&E?  is not actually installed in an area at risk of high impact, or other serious external influences? As such the SWA could easily have been terminated somewhere more suitable externally and the final length of cable brought into the CU as PVC.  As a small PVC cable will always be easier to dress and terminate into a small residential board and a single copper CPC will be far easier to work with in a confined space than multiple twisted steel strands of armour butchered into an earth bar. Other than a bit of laziness and possible speed to get away from the job ASAP on the installers side, there is no advantage to your method. Obviously if you can link a manufactures guidance where this method of terminating a SWA is considered acceptable then I am open to correction. But until such times there is no practical benefit to your solution. Additionally when it comes to periodic inspection and testing or any fault finding work where you may need to open up your twisted armour termination, it will be far more time consuming than a standard copper conductor. Your earlier comment statement about multiple connections being less safe due to HR joints etc. is irrelevant. Excluding the most basic of final circuits, you will always have multiple joints, or shall we limit light circuits to single luminaires just in-case we get a HR joint in the circuit? This is one of the key reasons we do dead testing before energising a modified circuit. Any competent person will be able to correctly terminate all cable types and ensure an electrically and mechanically sound joint in accordance with manufactures and industry standard guidance. It all pivots back to the "good workmanship and proper materials" guidance. 

Doc H. 


Thank you for the constructive feedback, good points.

 
@ Sidewinder,

Just looked up most of those regs, pretty subjective stuff to be honest...


Some are, some are cut and dried, remember, in the event of an issue then there would be an investigation, and witnesses would be called, and you would need to find a professional who was willing to back you in your method, I suspect that most would not.

 
Some are, some are cut and dried, remember, in the event of an issue then there would be an investigation, and witnesses would be called, and you would need to find a professional who was willing to back you in your method, I suspect that most would not.


I suspect you are correct, but I don't see how the way I did this would cause any sort of danger. I accept it's considered bad practice by many and it may not have complied with all relevant regulations at the time. If I were practicing today, I'd certainly not do it, but I've still yet to hear a valid argument as to why it might actually be electrically unsafe in the use cases I've suggested.

 
I suspect you are correct, but I don't see how the way I did this would cause any sort of danger. I accept it's considered bad practice by many and it may not have complied with all relevant regulations at the time. If I were practicing today, I'd certainly not do it, but I've still yet to hear a valid argument as to why it might actually be electrically unsafe in the use cases I've suggested.
OK, how were you restraining the cable entry to ensure that there could be no undue mechanical strain on the terminations?

When you split the armour into multiple connections, how did you calculate the distribution of the fault current?

How did you ensure that the fault current could not have EM effects on the armour such that iw would not become displaced in fault scenario, and cause other issues, i.e. come onto contact with other parts etc.?

 
so youve acknowledged that its not the correct way but its fine because youre retired?


I acknowledge it's not the accepted norm & today, due to the far more stringent regime, part P etc. I'd not consider doing it, but back in the day, yes I did it when the situation seemed appropriate. Maybe it was incorrect, but I don't see how, and no one has yet come up with a valid reason as to why, in the situations I've described, it could be considered unsafe.

I think a lot of people just blindly follow rules, or do things a certain way because that's what they were taught. Sometimes thinking outside the box and using common sense is appropriate. 

 
I acknowledge it's not the accepted norm & today, due to the far more stringent regime, part P etc. I'd not consider doing it, but back in the day, yes I did it when the situation seemed appropriate. Maybe it was incorrect, but I don't see how, and no one has yet come up with a valid reason as to why, in the situations I've described, it could be considered unsafe.

I think a lot of people just blindly follow rules, or do things a certain way because that's what they were taught. Sometimes thinking outside the box and using common sense is appropriate. 


Which is all well and good but it needs to be recorded in departures on the certs and you will be expected to show sound engineering reason as to why you have made the departure and the assurance that it isn't less safe. At the end of the day its your signature on the cert and you in court if it goes wrong and if you cannot prove that it isn't less safe than following the regulations then the usual defence of EAWR reg 29 will not apply and you will be hung out to dry. 

 
OK, how were you restraining the cable entry to ensure that there could be no undue mechanical strain on the terminations?

When you split the armour into multiple connections, how did you calculate the distribution of the fault current?

How did you ensure that the fault current could not have EM effects on the armour such that iw would not become displaced in fault scenario, and cause other issues, i.e. come onto contact with other parts etc.?


Good points,

I'd only do this were mechanical strain wasn't an issue, ie, cable was fully cleated to the location on the other side of a wall high up with no realistic chance of the cable being pulled. 

Armour would be split into 2 equal bundles, twisted, sleeved, run and terminated next to each other in the CU earth block, fault current would behave exactly as in a conventional termination except the CSA of the final earth connection is bigger than it would be with the usual tail from an earth ring/nut.

Quite rigid steel armour twisted and run directly together,  we're only talking protection by 20A/32A MCB's at most here, your suggestion is absurd. In any case it was all sleeved, any slight movement that may have occurred would have had no detrimental effect on surrounding components in the CU. 

Which is all well and good but it needs to be recorded in departures on the certs and you will be expected to show sound engineering reason as to why you have made the departure and the assurance that it isn't less safe. At the end of the day its your signature on the cert and you in court if it goes wrong and if you cannot prove that it isn't less safe than following the regulations then the usual defence of EAWR reg 29 will not apply and you will be hung out to dry. 


Like I say, different times, wouldn't be doing it today if I was still practicing.

 
I acknowledge it's not the accepted norm & today, due to the far more stringent regime, part P etc. I'd not consider doing it, but back in the day, yes I did it when the situation seemed appropriate. Maybe it was incorrect, but I don't see how, and no one has yet come up with a valid reason as to why, in the situations I've described, it could be considered unsafe.

I think a lot of people just blindly follow rules, or do things a certain way because that's what they were taught. Sometimes thinking outside the box and using common sense is appropriate. 
But going by that flawed logic we would still be using Screwits to make connections, which moved onto terminal block and now we have moved onto Wagos, all our screwdrivers would still have tape insulation or if you are posh pyro sleeve, we would still be doing daft things like trying to Earth window frames.

Things move on, standards get better (mostly), and hopefully as we get older and more experienced our own standards get better.

 
Good points,

I'd only do this were mechanical strain wasn't an issue, ie, cable was fully cleated to the location on the other side of a wall high up with no realistic chance of the cable being pulled. 

Armour would be split into 2 equal bundles, twisted, sleeved, run and terminated next to each other in the CU earth block, fault current would behave exactly as in a conventional termination except the CSA of the final earth connection is bigger than it would be with the usual tail from an earth ring/nut.

Quite rigid steel armour twisted and run directly together,  we're only talking protection by 20A/32A MCB's at most here, your suggestion is absurd. In any case it was all sleeved, any slight movement that may have occurred would have had no detrimental effect on surrounding components in the CU. 


OK, a few very common misconceptions in your answers.

If the cable is not terminated at the entry to the enclosure, it is not adequately terminated to prevent mechanical strain, thus it must have a gland.  No gland, inadequate termination.

Fault current does not beehave like "normal" current, the di/dt is such that there are phenomenon that are not normally observed, and area, length, bend radius, route, proximity to other materials all have an effect upon the current flow.

My suggestion is NOT absurd, the protective device has zero bearing on the potential fault current, and it is only the energy let through which limits the "harm" to downstream equipment.

How do you know that the movement would be slight, you are suggesting that this was a long time ago, but now you are suggesting 20/32A MCB's, would it not be more likely that we would be dealing with fuses, and 1 or 5s disconnection times.

What were the potential fault currents at the origins of the supply, 3kA, 4.5kA, 6kA, 10kA, higher?

Was this only domestic, or industrial/commercial?

 
But going by that flawed logic we would still be using Screwits to make connections, which moved onto terminal block and now we have moved onto Wagos, all our screwdrivers would still have tape insulation or if you are posh pyro sleeve, we would still be doing daft things like trying to Earth window frames.

Things move on, standards get better (mostly), and hopefully as we get older and more experienced our own standards get better.


Screwits were before my time! I'm not familiar with Wagos, still think in terms of terminal blocks... I assume they are some sort of fast to use, click a lever to close version of a terminal block?

I agree standards (sometimes) get better, sometimes just more bureaucratic! I do recall the earthing aluminium window frames nonsense.

 
OK, a few very common misconceptions in your answers.

If the cable is not terminated at the entry to the enclosure, it is not adequately terminated to prevent mechanical strain, thus it must have a gland.  No gland, inadequate termination.

Fault current does not beehave like "normal" current, the di/dt is such that there are phenomenon that are not normally observed, and area, length, bend radius, route, proximity to other materials all have an effect upon the current flow.

My suggestion is NOT absurd, the protective device has zero bearing on the potential fault current, and it is only the energy let through which limits the "harm" to downstream equipment.

How do you know that the movement would be slight, you are suggesting that this was a long time ago, but now you are suggesting 20/32A MCB's, would it not be more likely that we would be dealing with fuses, and 1 or 5s disconnection times.

What were the potential fault currents at the origins of the supply, 3kA, 4.5kA, 6kA, 10kA, higher?

Was this only domestic, or industrial/commercial?


OK, in order,

In the use cases I've stated,  in reality, barring a violent attack on the enclosure, this would not be an issue, in such an attack, I'd firstly be more worried about the T&E's that have no such strain prevention.

Fault current, fair enough, but I find it hard to believe there would be a problem there, and if so, why would this be worse than the tail from an standard gland installation?

I was using MCB's back in 1994. 

Domestic only, plastic consumer unit only.

 
Last edited by a moderator:
OK, in order,

In the use cases I've stated,  in reality, barring a violent attack on the enclosure, this would not be an issue, in such an attack, I'd firstly be more worried about the T&E's that have no such strain prevention.

Fault current, fair enough, but I find it hard to believe there would be a problem there, and is so, what about the tail from an standard gland installation?

I was using MCB's back in 1994. 

Domestic only, plastic consumer unit only.


What about the cable being manipulated, an SWA has much greater mechanical strength than a flat twin.

What is the difference between the bundled strands of the armour and the tail from a gland connection?

I thought you were talking about much earlier than 1994.  I would have to dig through my library to see if I have a copy of the Red & green 16th's I don't have them on computer, and I may not have them at all, though I should have a Red 16th, however, I could find contraventions in the Red 15th, so the 16th should be easier.

Domestic fault currents, can, be a lot higher than you may think.

A domestic CU is rated at 16kA with a suitable DNO backup fuse, and was back then under EN 60439-3 even though the individual breakers are only 6kA.

Back in 1994 I was building up and commissioning CNC machines in the automotive industry mainly. (Just as an aside!)

It wasn't an acceptable method then, and it's still not now.

 
I think we need to draw this to a close. My only argument was that in some specific domestic only instances were plastic consumer units were used, treating a smaller SWA cable leaving the CU as you would terminate a concentric could make a neater job than the rather industrial type of gland conventionally used to terminate SWA and that although maybe not considered good practice, was pretty much just as safe as a conventional method if done tidily.

OK, I get it, most of you don't agree. So I was probably wrong to do it. I was probably only ever an average sparky in any case to be honest, lol!! (although, I've yet to see anyone cosmetically neater in their work, to this day!) Anyway, I left the profession in 1994 at 25 to do something I liked better. I worked for the YEB at the time and all we ever did was domestic, I'd had enough of house bashing, chopping out chases all day and bashing my hands up, or grovelling around insulation infested lofts in summer heat finding dead rats... 

 
Last edited by a moderator:
Good points,

I'd only do this were mechanical strain wasn't an issue, ie, cable was fully cleated to the location on the other side of a wall high up with no realistic chance of the cable being pulled. 

Armour would be split into 2 equal bundles, twisted, sleeved, run and terminated next to each other in the CU earth block, fault current would behave exactly as in a conventional termination except the CSA of the final earth connection is bigger than it would be with the usual tail from an earth ring/nut.

Quite rigid steel armour twisted and run directly together,  we're only talking protection by 20A/32A MCB's at most here, your suggestion is absurd. In any case it was all sleeved, any slight movement that may have occurred would have had no detrimental effect on surrounding components in the CU. 


You have still failed to offer any industry recognised guidance backing up your method, you have referenced no regulations applicable either then or now that suggest your method is acceptable. By your own admission you consider it would not comply with BS7671. What happens a few years after your SWA has been fitted and A.N.Other has the board opened to do some alterations/additions/fault finding/testing etc.. the "inside" bit of your joint is open and accessible to mechanical strain whilst undergoing work inside the CU. Its not just about someone pulling the cable from outside. During the normal expected lifespan of the installation and the cable you have installed, it is quite probable that persons other than yourself could be working in or around your SWA. The whole concept of your joint is liable to mechanical damage, and/or damage to other cables in the enclosure as it is not properly terminated. It is bad enough working in a CU with a rats nest of PVC cables, the last thing you need is a SWA hanging around in the board as well, with loads of strands of relatively inflexible steel armour strands twisted and cluttering the earth bar. I would consider this to be a potential danger to any person working in the enclosure at a later date. So rolling back 20 odd years, you probably should have been applying, 130-01-01,  130-02-01 as a bit of basic good practice and common sense. Just because you cannot see a danger, does not mean there is none, and the very title of your thread implies you know that. The bottom line is that we do not have to prove it is electrically unsafe or dangerous etc. But you as the "competent installer" have to be able to prove it is safe, and compliant with the relevant regulations, which so far you haven't.

Doc H. 

 
Top