Tramps in your toilet? – waste and consumption

‘Energy saving’ still tends to imply the classical three-to-five-year-payback

capital project, with its attendant disruption, technical specialism, and perceived

risk. But managements today are less disposed than ever to contemplate these

‘hard’ solutions. So, this article introduces a desktop information-management

technique which, if properly applied, will in almost any circumstances yield

significant savings on energy and other consumables with little or no expenditure.

Such factors would include vehicle mileages, hours of darkness, production

throughputs, attendance figures, and how cold (or hot) the weather was. I am

happy to concede that it is sometimes difficult to pin down what factor affects

what, and to determine quantitatively what the link is between the two; but

the investment and effort necessary to establish these basic relationships will

be amply rewarded.

To take an obvious example: if we know the achievable mpg figure for a car,

then from the miles driven in any given month we can estimate what would be

a reasonable quantity of fuel to have used. Likewise, if we had taken the trouble

to establish how our electricity consumption varies with available daylight,

we could (having measured the hours of darkness over the month) estimate a reasonable

ration for electricity.

And so on, for every stream of consumption for which we have previously been

able to establish (a) an appropriate driving factor and (b) the formula by which

the two are related. Some anecdotes may give you a better feel of where this

is all leading.

The energy manager of Taunton Deane Borough Council noticed that patterns of

water and electricity consumption were spontaneously changing in the borough’s

public lavatories. On investigation he found one case of a technical fault in

the urinal flush controller; the remainder were caused by tramps dossing down

at night and triggering the automatic flushing systems.

Repeatedly aroused from their slumbers by Niagara Falls going off, they were

jamming matchsticks into the pushbuttons of the hand dryers in an effort to

keep warm. And there was one other exceptional circumstance: one of the lavatory

attendants, who was nesting in a service duct, had wired in a fan heater to

keep himself warm.

Another example is from a sewage works that ended up with the cleanest sewage

screens in the world. Raw sewage passed though a coarse screen between two open

chambers. Periodically, as the screen became fouled, the rise in upstream chamber

level triggered an electrically-powered rake, which scraped off the debris and

then (like a car’s windscreen wipers) parked at the end of its return stroke.

A loose bolt on the limit switch caused this parking action to fail, and the

5kW rake motor, plus about 8kW of associated pump and macerator motors, began

to run continuously instead of only occasionally.

Finally, one data processing centre in Swindon lost £9,000 over an eighteen-month

period because its heating and air conditioning systems had become locked in

contention through a minor control fault.

Nobody is immune from systems being incorrectly used or maintained, or leaking,

or going out of adjustment, or running when they are not needed. I have seen

a brand-new building exhibiting symptoms in at least three of these categories,

so what hope for a building which has had five, ten, or 20 years to develop

hidden waste?

Ration calculations

Again, it is possible to estimate the ‘correct’ ration of energy, water, etc,

if one has a measurable driving factor with a known relationship to consumption.

One classically difficult commodity to manage is gas, since its monthly consumption

is usually so very variable. How would you estimate an appropriate allocation

of gas at the end of each month? Paradoxically, it turns out that the demand

for space-heating fuel is relatively easy to gauge independently, through the

medium of degree-day data.

Published on the internet (www.DegreeDaysDirect.com) degree days provide an

index of how cold each month was in each region of the UK. Thus, in a month

which registers 300 degree days in a particular region, a given building will

have needed twice as much heat as in a month registering 150 degree days; three

times as much as in month registering 100 degree days; and so on.

Monthly fuel requirements for space heating are proportional to the monthly

degree-day value. The only slight complication is that there may also be a fixed

element of fuel consumption attributable to domestic hot water, catering, or

other non-weather-related uses. But this is not an insurmountable difficulty,

since we can say that the building will use so much fuel per month, plus so

much per degree-day.

This we can call its caracteristic pattern of consumption. Knowing the regional

degree-day figure for the month we can use this characteristic pattern to estimate

the fuel consumption target.

For example: suppose a building’s history of fuel consumption has been critically

examined and that its minimum achievable consumption characteristic has been

assessed as 3,200kWh per month (fixed) plus 45kWh per degree-day (the weather-dependent

component). Then suppose that in a particular month when the prevailing weather

yielded a resultant 275 degree days, the building used 17,700kWh. Is this good,

bad or indifferent?

Answer: the expected consumption for the month can be computed as 3,200 + (45

x 275) = 15,575kWh. As this is only 2,125kWh less than was actually used, we

might conclude that it is not too far adrift. We would repeat the assessment

each month, and would expect the deviation to be random; sometimes more than

expected, sometimes less. Sometimes a larger deviation, sometimes smaller.

The point is that we can make an objective assessment of the cost of each month’s

deviation from expected performance, and if it breaches a certain threshold

we would be prompted to take action. The same thinking can be applied to many,

if not most, other consumable commodities, since they are purchased for a purpose

and that purpose may be quantifiable (mileage, hours of darkness, etc). Figure

1 summarises the general principle.

Plotting week by week or month by month the quantity of resource consumed (a)

against the appropriate driving factor (b), which represents the work done,

one would expect to see the points exhibit a rational pattern (c) such that

more resource is consumed to do more work. An idealised straight line (d) can

be superimposed on the scattered points to represent an activity-related target.

High consumption (e) can then be quantified by reference to the ‘expected’ consumption.

There is a free government publication covering these principles, Fuel efficiency

booklet 13: waste avoidance methods, which goes on to describe a graphical method

called ‘cusum charting’ which enables its users to tease out information about

subtle, low-level adverse shifts in performance, working out when performance

changed for the worse, how much it is costing in the long run, and even sometimes

what the physical nature of the fault is. It can also be used to verify, objectively,

the effect of energy-saving measures.

The technique is relatively simple (within the reach of a competent advanced

spreadsheet user) but remarkably powerful. Most importantly, it is universally

applicable to all forms of consumable resource, not just energy and water; think

particularly about environmentally damaging ones like volatile organic compounds.