Omission of overload protection for fixed loads.

[spark 68]

DW is correct,
GN6 gives three examples of where OL protection may be omitted,


i) For a conductor situated on the load side of the point where a reduction occurs in the value of CCC, where the conductor is effectively protected against OL by a protective device placed on supply side of that point.


ii) for a conductor which because of the characteristics of the load or supply, is not likely to carry overload current, provided that the conductor is protected against fault current according to section 434. For example a domestic shower or immersion heater.

iii) At the origin of an installation where the distributor provides an OL device, and agrees that it affords protection to part of the installation between the origin and the main distribution point of the installation where further OL protection is provided. For example meter tails. (this also applies to the 3m rule).

Nowhere in this set of regs or GN6 suggests the Iz of the cable is less than the In of the OCPD, or less than the Ib of the circuit it is intended to serve.

 

[Outspoken]

Spark, that is Reg 433.3.1 that I quoted in the post DW was referring too, that is simply a repeat of the Reg and it does not state that it is related to showers, and it seems to me that people are arguing over nothing here as everyone seems to be of the firm conclusion that although the Regs are saying there may be situations where O/L protection may be omitted, no-one actually would omit it.

If this last point I have made is a correct reading of everyone's post I have to ask what was the point of the thread?

 

[spark 68]

OS,

The passages I quoted were directly from GN6, and this does mention showers and immersion heaters. :smile5:

As others have said, if I was designing a new circuit then obviously I would obviously not omit it.

 

[spark 68]

The point DW was making at the beginning of this thread was that the OCPD may be higher than the CCC of the cable provided that the Ib of the load is within the CCC, and is unlikely to overload the cable.

This is a very good point to bring up IMO, and is no doubt very useful reg to bear in mind when faced with an otherwise sticky problem.

 

[Engineer54]

Spark, that is Reg 433.3.1 that I quoted in the post DW was referring too, that is simply a repeat of the Reg and it does not state that it is related to showers, and it seems to me that people are arguing over nothing here as everyone seems to be of the firm conclusion that although the Regs are saying there may be situations where O/L protection may be omitted, no-one actually would omit it.

If this last point I have made is a correct reading of everyone's post I have to ask what was the point of the thread?

I think DW was just pointing out, that at the design stage of any new installation, all the usual criteria should be applied to those circuits being installed. As in most circumstances on these threads, this situation normally relates to existing circuits, such as cookers, ovens and hob installations, water heaters etc.

 

[Darkwood]

Would these include fluorescent fittings and LEDs?

Most certainly so, regardless of the piddly several watt output each driver has a inrush spike of which on it own isnt an issue but stacked together creates rather a large initial spike .... you will have to discuss with tech regarding LED and Electronic start batten fittings as they have yet to update their tables as the market is changing quite fast.

 

[Darkwood]

Spark, that is Reg 433.3.1 that I quoted in the post DW was referring too, that is simply a repeat of the Reg and it does not state that it is related to showers, and it seems to me that people are arguing over nothing here as everyone seems to be of the firm conclusion that although the Regs are saying there may be situations where O/L protection may be omitted, no-one actually would omit it.

If this last point I have made is a correct reading of everyone's post I have to ask what was the point of the thread?

The point of the thread is ive noted on several occasions that this regulation is been brought out like a joker card and thus sets up a firing squad debate where anyone that suggests say a shower without overload protection is presumed stupid and incompetent... i am merely expressing that according to the regs it is in fact permitted all be it compliance to protection against short circuit must be met.
You mention you find no reference of showers in your books so i assume you haven't got The IEE Electrical Installation Guide which is the same style and size as the OSG but is in a grey binder.... within this guide this regulation and the conditions of its application are mentions numerous times ....

Page 36 (4.2.2) Overloads currents

Overload currents do not arise as a result of a fault in the cable or equipment. They arise because the current has been increased by the addition of a further load.

Overload protection is only required if overloading is possible. It would not be required for a circuit supplying a fixed load, but fault protection is required except for exceptional circumstances (434.3).

The load on a circuit supplying a (say) 7.2kw shower will not increase unless the shower is replaced where the adequacy of the circuit must be checked.........

This is just a short snippet of the paragraph that does clearly suggest this situation can be put into practice in domestic but even in the light of this i express that i feel it shouldn't unless as i already mentioned you find yourself doing an upgrade and been a few amps short for overload protection and the newly fitted bathroom suite will be destroyed in pulling in a 10mm then it is possible to utilise the reg' and as others have mentioned when testing existing installs and you find this set up you shouldn't condemn it as long as fault protection complies.

The fact this thread has shown differing strong opinions is good point to its existence im trying to point out that whenever a shower thread arises this regulation is often misquoted as a Ace in the hand but i feel it shouldn't be used in shower threads as it throws more confusion to the OP about his original concern than it helps him.


PS next Friday i decided im going out

 

[Darkwood]

IEE Electrical Installation Design Guide.

I quote (this is a highlighted paragraph on p11)

Note: Overload protection is provided in practically all circuit designs in order to protect the cable should the load be increased, e.g. by adding further lights to a lighting circuit or changing a shower for one higher rated without proper checks been made.

However, for a fixed load e.g. a shower circuit, this is not an actual requirement of BS7671

The exact point ive been trying to make ...if you upgrade a shower and do all the necessary calc and checks and deem it impractical to try upgrade the cable due to restrictions then you are still in full compliance with BS7671 and shouldn't be called a numpty in doing so.

But ill stress even i would only implement as a last resort as it leaves the circuit prone to a real numpty upgrading without doing the necessary calcs, I would never use in design as such for a shower circuit or similar.

 

[Outspoken]

DW, I understand what your saying, and now you have stated where you actually got the reference to a shower I can understand the comments you have made, Yes I do own a copy of the IET EIDG and have it sat here, I would also take to task the statement they make in their opening statement...

"Overload currents do not arise as a result of a fault in the cable or equipment"

Whilst I would accept that under normal operating conditions this statement hold true, but then overload currents occur because of operating conditions outside of normal. Electrically we all know that environmental conditions can impact the cable supplying a circuit, I have encountered a shower feed that almost caused a fire because some numpty had cable tied the 6mm² T&E to a central heating pipe to make it "neat". This section of the cable was overheating and this changed the electrical resistance of the cable. In this instance the owner of the property smelt the cable breaking down and called me out.

Now I accept the chances of environmental impact are not common, but we all have to understand that environmental impacts outside of our control on the capacity of the cable can alter it's electrical properties and thus an overload current can be caused by the cable as well as the load attached to it, even if 999 times out of 1000 it will be the load causing the problem.

I am pleased to see that you have stated you would not rely solely on an RCD for protection of a shower. In the last 15 years when I have been involved in a shower install, albeit at home or in a commercial setting, I have always installed on an RCBO so that both types of protection are afforded the circuit.

I would agree that this debate has highlighted a difference of opinion on this issue, and in the cold light of day without the "affluence of incahol" on the thought process I can see why some have made the statements they have.

The problem I have is that if it becomes common practice for showers to be uprated and the cable to remain, without being uprated when perhaps it should be, with perhaps just an RCD protecting it, this has the potential to create a dangerous precedent that could become misinterpreted by the inexperienced and the bodgers to simply install showers without doing proper calculations and without proper due diligence on the circuit construction and create situations where it is then dangerous.

We are supposed to be increasing electrical safety and not leaving open the door to lowering it, which this has the possibility of doing.

 

[Darkwood]

I understand the angle you were taking outspoken and can agree to the the majority of the points you make but the truth is that if the environment changed it is likely that any negative effects of this would be a small overload for a long duration and as you must agree your protective device would not protect you from this anyway even if your circuit was provided with the correct overload hence we have a regulation making sure any designed circuit should ensure a small overload cannot occur for a long period of time. Your example of a cable strapped to the heating pipe would not in my mind operate the overload protective device as the increased load would be below the mcb threshold to trip in a given time i.e. <1.45 rating of device. I see your point but it still wouldn't make a difference if the mcb was design with overload protection of the cable or not, many forget that cable insulation is rated at 70 degrees and IMHO you need a really poor design and set of circumstances for a fire to occur and this will be unlikely to be because in our particular example that the shower had no overload protection.

Dont get me wrong im ot trying to create an open field day here for arguing for the omitting overload protection only having the debate to highlight that it is a last resort option on a upgrade if all fault current calculations are met and due to the installation methods it is highly unlikely to have its environment changed like lagging a loft etc.

 

[Outspoken]

Gents, there is an aspect of this that we have all, so far omitted, namely Regulation 433.1.1 that states that the operating characteristics of a device protecting a conductor against overload shall satisfy the following conditions;

I[SUB]b[/SUB] ≤ I[SUB]n[/SUB] ≤ I[SUB]z[/SUB]

Now we all now that;

I[SUB]b[/SUB] is the design current of the circuit
I[SUB]n[/SUB] is the nominal current or current setting of the protective device
I[SUB]z[/SUB] is the current-carrying capacity of the conductor in the installation (taking into account environmental conditions)

These conditions have to be met regardless of any other factors. I think people are getting slightly confused because some have used motors as an example where BS88-3 fusible links are used to give fault protection and not overload protection because motor starters already have O/L protection installed, and thus this aspect of protection is already accounted for. For the most part the O/L, if set correctly, will protect the supply circuit by disconnecting the motor from the supply before it becomes over loaded. However it is possible for the O/L to either be set incorrectly and/or be faulty/bypassed and thus the fusible links will provide both fault and overload protection so long as they are selected correctly, and the same is true for circuits with fixed resistive loads, such as showers.

The shower in the example that originally started all this debate is a fixed load of lets say 5.3Ω, which at 230V gives a power rating of 9.98kW, but how the shower is constructed is vitally important and something that has been lost in this debate.

I have an electric shower in an extension where visitors stay, it is a Mira 10kW unit and very nice too. Before posting this I went back to my college/apprentice days and took the thing apart, completely and with some manufacturers literature discovered something very interesting.

The unit has THREE fixed resistance coils;

• Low 9.8Ω, @230V = 5340w (5.4kW).
• Medium 6.75Ω @230V = 7837w (7.8kW)
• High 5.25Ω @230V = 10076.19w (10kW)

The switching is completed by a simple 3 position switch.....

Now investigations into the switch can see that is is possible to break down (although I accept they are tested fully and this is likely rare) and connect two or more of the heater elements together...I think straight away anyone looking at this can see the problem.

I was curious as to whether the manufacturers had used a single coil that was broken into three sections that were ultimately connected together to form the maximum rating or whether these where in fact three separate coils...They are 100% separate coils.

Obviously if the coils were connected in series we don't have much to worry about as the rating at 230V would be R[SUB]1[/SUB] + R[SUB]2[/SUB] +R[SUB]3[/SUB] which is thus 9.8+6.75+5.25 = 21.8Ω and a power rating of a mere 2426w (2.4kW), however if the switch breaks apart it is highly likely that the coils can be connected in parallel and and thus we have a potential overload issue.

In parallel the combined coils have a resistance of;

Low and Medium. 0.25Ω @ 230V = 211,600w (211.6kW) = 920A
Medium and High 0.338Ω @ 230V = 156,508w (156.5kW) = 680V
Low and High 0.292Ω @ 230V = 181,164w (181.1kW) = 787A
Low, Medium and High. 0.44Ω 230V = 120,227w (120.2kW) = 522A

The above is wrong, forgot to divide 1 by the final resistance and missed my mistake. Thanks Spark 68 for the nudge!!

Should be this...

Low and Medium. 1/0.25Ω @ 230V = 57.5A
Medium and High 1/0.338Ω @ 230V = 77.96A
Low and High 1/0.292Ω @ 230V = 67.17AA
Low, Medium and High. 1/0.44Ω 230V = 101.23A

Now I do not know how typical a construction/control this is in showers, I cannot say I have taken too many apart over the years and really looked at them, but what this demonstrates is that even with fixed loads it is possible to cause a circuit overload when there are clearly separate coils with different resistances are providing different power levels.

I appreciate that many of you here have more experience with showers than I do, but do you know exactly how the manufacturers switch the loads for the different power settings in each and every shower manufactured?

 

[spark 68]

You may want to re-check your calcs here:

In parallel the combined coils have a resistance of;

Low and Medium. 0.25Ω @ 230V = 211,600w (211.6kW) = 920A
Medium and High 0.338Ω @ 230V = 156,508w (156.5kW) = 680V
Low and High 0.292Ω @ 230V = 181,164w (181.1kW) = 787A
Low, Medium and High. 0.44Ω 230V = 120,227w (120.2kW) = 522A

I think you have forgotten to reciprocate the sum of your addition

Low 9.8Ω, @230V = 5340w (5.4kW).
Medium 6.75Ω @230V = 7837w (7.8kW) High 5.25Ω @230V = 10076.19w (10kW

 

[Outspoken]

Spark..

E²/R = Power so therefore;

230² = 52900 divided by the resistance of a coil gives the power output = 52900 /0.25 = 211600w

Current drawn on the circuit as a result is P/E = 211600/230= 920A

This can all be verified by turning the equations around...

Thus:

Power = 211600
E = 230
I = 920

Thus resistance
E² / w = 52600/211600= 0.25
or E/I = 230 / 920 = 0.25

Confirm E by √wR = √211600 * 0.25 = 230

Which ever way you want to work it out it calculates correctly...

In case your talking about adding resistances in parallel..

1/(1/R[SUB]1[/SUB])+(1/R[SUB]2[/SUB])+(1/R[SUB]3[/SUB])

1/9.8 = 0.102Ω
1/6.75 = 0.148Ω
1/5.25 = 0.190Ω

Add the resultants together in the combination required and you get the combined parallel resistance...

Or do you know a different way that has escaped us?

 

[spark 68]

ok using your figures:

Low 9.8Ω, @230V = 5340w (5.4kW).
Medium 6.75Ω @230V = 7837w (7.8kW) High 5.25Ω @230V = 10076.19w (10kW)

low and medium in parallel, 1/9.8 + 1/6.75 = 1/total, = 0.1 + 0.15 = 0.25, 1/0.25 = 4 ohms total.

230/4 = 57.5A , = circa 13kW

 

[Darkwood]

Are we forgetting that showers have a thermal cut out for any occasion when the elements overheat due to lack of water if the shower doesn't have a pressure switch or the breakdown and part shorting of the coils of the element....