Diversity
Diversity in this context is all about how one establishes the maximum demand for an electrical installation or an electrical circuit. The guidelines shown here are some of those as presented in the On Site Guide. Note however that these are just that - guidelines. They are based on statistical probabilities and typical usage. As a circuit or installation designer you may choose to make different assessments of current demand based on your knowledge of the actual installation.
Note also that the figures presented here are for domestic properties only.
Allowances at point of use
The following guidelines are used when estimating the load for a particular final circuit, and shows how appliances and power points etc fed from the circuit should be accounted for:
| Point of use / appliance | Current demand to be assumed |
|---|---|
| Other Socket outlets (i.e. not 2A or 13A) | Rated current of the socket |
| 2A Sockets | at least 0.5A |
| Lighting outlet | Actual rating of lamp or 100W (whichever is greater) per lamp holder.
For circuits (also) supplying discharge lighting (i.e. linear fluorescent tube lights, sodium / mercury vapour lamps, or any lighting technology that requires the use of a current limiting ballast) then use VA load rating if known, since this will allows for any ballast losses and the power factor. If the VA rating of the lighting units is unknown, then use the rated wattage of the actual tubes/lamps used multiplied by 1.8 to arrive at an estimate of the VA rating for the lamp and ballast. (you can omit this step if you know all the fittings you are using contain modern HF ballasts or power factor correction). e.g. 10 x 58W Single tube linear fluorescent lamps, gives a total real power consumption of 580W. Multiply by 1.8 to give 1044 VA, and hence 4.5A |
| Shaver socket, bell transformer,
electric clock point, or any appliance under 5 VA |
May be neglected for the purposes of assessment. |
| Cooking appliance | 10A of total load plus 30% of remainder. Add additional 5A if cooker point socket is fitted. |
| All other fixed equipment | Use nominal current as specified on the appliance ratings plate. |
Examples A lighting circuit with 6 traditional 1 lamp fittings, plus a 500W Halogen flood lamp, would be treated as 6 x 100 + 500 = 1100W or 4.8A A 2.6kW double oven, plus 6.8kW hob, and a cooker point with a fitted socket. Total load is 9.4kW or 41A peak load. Hence the diverse load is 10A + 30% of 31A + 5A for the socket = 24.3A
Allowances for Diversity
The following table is typically used for when assessing the load of a group of circuits fed from a single distribution board or CU. Note that the total current value used is the total current after diversity has been applied - not just the nominal rating of the MCB or Fuse
| Purpose of circuit | Diversity applicable (domestic) | Notes |
|---|---|---|
| Lighting | 66% of total current | |
| Heating and Power | 100% of total demand up to 10A, + 50% of the remainder. | Where space heating is connected to a general purpose power circuit. |
| Cooking appliances | 100% of total demand up to 10A, + 30% of the remainder. | Add a further 5A if a socket is fitted to the cooker point. |
| Water heaters (instant) | 100% of first appliance, 100% of second, and 25% of any remaining appliances. | |
| Water heaters (storage) | No diversity allowable | |
| Under floor heating | No diversity allowable | |
| Storage Heaters | No diversity allowable | |
| Standard Circuits | 100% of current demand of largest circuit. + 40% of demand for every other circuit. | These are standard final circuits as described in appendix 8 of the OSG |
Example
A domestic consumer unit with the following circuits:
| Circuit No. | MCB Nominal Rating | Circuit Function | Design current (Ib) | Diversity Allowance | Demand after diversity |
|---|---|---|---|---|---|
| 1. | 32A | Cooker (12kW) | 32A | 100% | 23A |
| 2. | 32A | Ring 1 (kitchen) | 32A | 100% | 32A |
| 3. | 32A | Ring 2 (downstairs) | 32A | 40% | 13A |
| 4. | 32A | Ring 3 (upstairs) | 32A | 40% | 13A |
| 5. | 16A | Immersion heater (3kW) | 13A | 100% | 13A |
| 6. | 6A | Lights (1kVA) | 4.3A | 66% | 3A |
| 7. | 6A | Lights (600VA) | 2.6A | 66% | 2A |
| Total | 148A | 99A |
You will note that the final load including diversity is within the nominal capacity of a domestic 100A supply, even though the maximum load including individual circuit diversity alone is significantly in excess of it.
When even diversity is not enough
Imagine we added an additional circuit to feed a 40A shower to the circuits listed in the example above. We would then arrive at the situation where even applying all the diversity calculations described here, the total demand including diversity would be well in excess of the apparent supply capacity. This is an increasingly common situation that seems to rarely cause any problem in practice. It also worth noting that there are still many many properties with only 60A or 80A supplies rather than 100A, where it is even easier to apparently exceed the supply capacity.
Part of the cause of this apparent problem are the assumptions upon which the calculations are based. These are now somewhat outdated and don't reflect modern usage for many properties. The guidance in the OSG for example has remained unchanged for many years, and yet in that time we have seen:
- an "explosion" of use of small low load devices in the home;
- a large growth in the numbers of socket outlets deemed adequate
- The introduction of the 17th edition of the wiring regs which has tended to further increase the number of circuits used due to the increased reliance on RCDs and the need to maintain discrimination under fault conditions; i.e. Although there has been a general increase in the number of circuits provisioned, there has not been a corresponding growth in electrical demand.
- a general fall in the numbers of houses that rely on electricity for extensive day time heating as oil or gas fired central heating has become almost ubiquitous.
Couple this to the reality that the incomer fuse will typically support quite significant overload for short durations, and the supply impedance in the majority of properties is high enough that there will be a small amount of load shedding[1] at high loads, and it becomes apparent why they are not fusing on a regular basis. (note that under no circumstance should the load shedding effect be relied upon as a design parameter since it is not under the designers control, and is subject to change without notice - it is included here only as an example of a mitigating factor)
- [1] Most properties will have a measurable supply impedance of a few fractions of an ohm. This will mean as the load climbs toward the maximum, the available voltage will sag. This has a small natural limiting effect of reducing current drawn by the majority of domestic appliances. e.g. a property with a 240V supply, and 0.2 ohm impedance, will drop 20V at full load, reducing the current demand by nearly 10A.
Alternative Method: Applying a judgemental approach
An alternative approach to assessing load is to make an engineering judgement based on more detailed knowledge of the installation, and can take into account:
- the operating time profile of the load
- the coincidence or simultaneous operation of individual loads with other loads
- the seasonal demands of heating and cooling loads
- the allowance, if any, for spare load capacity based on the users needs.
Examples of these judgements could be:
- Where socket circuits are concerned, further reductions in load could be assumed after the first two are accounted for at 100% and 40% - say counting the remainder at 20%. Alternatively one might wish to count all socket circuits at only 40% loading.
- A load like an electric shower one might assume is unlikely to run for more than 12 mins out of every 15, and hence a reduction of 12/15ths would be arguable.
- It may be practical to ignore complete circuits where their loading is likely to be mutually exclusive (i.e. heating and cooling circuits are unlikely to run concurrently)
Any such judgement will be specific to a particular installation, and will take a little more time and investigation than a straight numerical exercise like that presented earlier.