Highlights from Energy Management Register bulletins

The bulletin goes out about once a week with advice and information related to energy-saving methods and products. Why not sign up? You can cancel at any time.

24 AUGUST 2023

REVERSIBLE HEAT PUMPS

Reader John C. called up to talk about an energy audit he had been doing on a building which had ventilation heat recovery using a reversible heat pump. Reversible heat pumps present a minor problem for monitoring purposes because unlike with a gas-fired heating system, you cannot calculate expected consumption using a regression model based on heating degree days as the sole driving factor ('independent variable'). Because both cooling demand and heating demand affect consumption you should, in theory, use a multivariate model in which expected energy consumption E is given by

E = c + m1.HDD + m2.CDD

where HDD is heating degree days, CDD is cooling degree days, and c, m1 and m2 are constants appropriate to the system in question.

But John said he does something different which surprised me: he uses a simple linear regression with the sum of HDD and CDD as the driving factor. Can this be right? Heating and cooling degree days are loosely complementary to each other but not actually mathematically related, so I have always considered them as unconnected and adding them together feels odd.

Now for John's method to be correct in theory, the building would need to have an expected-consumption model of the form:

E = c + m.(HDD+CDD)

This will be true in the special case where m1 and m2 in the first formula are equal to each other. Let's pursue that idea. Suppose that a building is mechanically ventilated such that all its makeup air has to be heated or cooled to a particular set point, with the same makeup rate all year round. Such a building's sensitivity to changes in outside air temperature will be the same whether heating or cooling. To take a numerical example if you need to inject, say, 2.4 kW of heat for each degree that the outside air is below set point, you will need to extract 2.4 kW of heat for each degree that it is above set point. If you do this with a heat pump that has a coefficient of performance of 3 when cooling and 4 when heating, the electrical power input relative to temperature difference will be 2.4/4 = 0.6 kW/degree when heating and 2.4/3 = 0.8 kW/degree when cooling. Converting 'kW/degree' to 'kWh/degree-day' this gives us m1 = 14.4 and m2 = 19.2. They aren't equal, but they are close, and that is probably why John's method works for him. He could perhaps tweak it by applying weightings to the HDD and CDD values but it might not make much practical difference.

Neverthess I would still recommend using a multivariate model when electricity is used for both heating and cooling, because it will cope with buildings that don't fit the convenient stereotype. Thanks to John, though, for a thought-provoking question.

Sign up for future bulletins | Energy management training | Ask a question

9 JUNE 2023

CHILLER MAINTENANCE

One common cause of poor chiller performance is failure to keep the condenser coils clean. The 'coil' is the air-cooled heat exchanger in which the hot refrigerant condenses back to liquid and if its airways are clogged the refrigerant temperature (and hence condensing pressure) rise, necessitating higher compressor power to extract the same amount of heat. As a rough rule of thumb a one-centigrade-degree rise in condensing temperature will add about 2% to the power demand in a low-temperature industrial refrigeration plant or 3% in the case of a chilled-water circuit feeding an air conditioning system. In extreme cases the high pressure associated with excessive condensing temperature will trip the chiller out.

For guidance you should expect to see condensing temperatures 12 to 15 degrees above ambient air when a chiller is under reasonable load. A higher difference suggests a problem: cleaning the condenser coils properly could be the answer, although there are other possible causes, like for example discharged cooling air short-circuiting back to the inlet.

Have a look at your maintenance records. They will show the ambient (or "on-coil") air temperature and the condenser (or "discharge") pressure, which corresponds to a particular temperature depending upon what refrigerant you are using (tip: photograph the pressure gauges, because they may have scales showing temperature as well as pressure). You can then do a "what-if" analysis of each chiller to check the effect of lowering its condensing temperature using this estimating tool.

Sign up for future bulletins | Energy management training | Ask a question

15 MAY 2023

REVERSE ROTATION

Here's an aspect of energy saving in motor-driven systems that had never occurred to me until I went on a training course last week about industrial dust extraction systems. Our instructor, Christoph Ritter of Osprey Corporation, guaranteed his audience that if he went to their factories he would find that some of their vacuum fans would be running backwards This may sound crazy, but it can and does happen. It only needs two of the motor power connections to be swapped accidentally. Centrifugal fans do still work in reverse but their efficiency becomes diabolical. If they have straight radial blades the fan-wheel itself is no less efficient but the air leaving the volute has to turn through 180 degrees, with the consequent loss of head. If the fan has backward-curved blades (normally more efficient) these are forward-curved when reversed, introducing even more loss.

The problem tends to be masked in direct-coupled fans with variable-frequency drives. One reason is that you cannot easily see the direction of rotation when there are no belts to observe; the other is that the drive system will compensate by speeding up the fan (if it can) drawing much more power to deliver the required air flow. On Christoph's course he uses a rig to demonstrate this and a fan current of 5 amps had to go up to 22 amps to deliver the same flow when the fan motor was running backwards.

Don't assume it cannot happen to you.

Sign up for future bulletins | Energy management training | Ask a question

27 APRIL 2023

COMPARATIVE CHILLER EFFICIENCIES

I've just completed two days of training for a group of energy auditors and I think a couple of points from the sessions are worth highlighting. The first concerns chillers. On a recent survey I pulled out the latest maintenance report for the site's twin-circuit chillers, one of which showed different discharge pressures on its two circuits. The discharge pressures tell us the condenser temperatures, which were 39°C and 43°C respectively. This is reasonable relative to the ambient air temperature of 27°C at the time of the test (my chiller expert friend Lawrence Leask says he'd expect a 12 to 15 degree differential under high load), so the results gave no cause for concern in themselves. What was more interesting was the fact that they were different. The circuit condensing at 39°C will be using almost 10% less electricity than the one condensing at 43°C to achieve the same output and therefore running it in preference to the other one will yield savings genuinely at no cost. This is not unlke the recommendation in my last bulletin regarding boilers with differing efficiencies.

The other point concerned control, and it is something I have mentioned here before: when you're faced with some energy-using system or object and you're lost for inspiration, just ask the question “how is that controlled?”. This is not just because improving control tends to be cost-effective. It's also just a good first step to understanding the thing you are studying.

Sign up for future bulletins | Energy management training | Ask a question

19 APRIL 2023

A NEGLECTED OPPORTUNITY?

Visit many boilerhouses and you'll find printout slips from test apparatus taped to the boiler casings. They're worth a closer look for the opportunities that they reveal. Here's why.

Every time the boiler is serviced, it will be tested to determine what proportion of the input fuel energy is going up the chimney. The service engineer's test equipment infers those losses from the exhaust temperature relative to ambient air and the concentration of either oxygen or carbon dioxide in the flue gas. It then reports the result as a percentage efficiency (which is one of the numbers printed out on the slips) and you can compare that with the figure given in the manufacturer's literature. If the observed efficiency is less than it should be, then either the air/fuel ratio needs to be adjusted (to minimise the residual oxygen level) or the boiler surfaces need cleaning (to bring down the exhaust temperature). Your maintenance contract schedule should put the onus on the contractor to maximise the efficiency.

Moreover, regardless of where a boiler sits relative to its target combustion efficiency, if you have two or more identical boilers on a range one may be more efficient than the others, and simply by making that one the lead boiler in the sequence you will save fuel at no cost. Furthermore, you could track the reported efficiency of each individual boiler through time. That enables you to challenge the quality of maintenance when a test result is worse than previously reported, and to ratchet up your expectations when a test result is better than you've had before.

Don't let those test result slips fade out of sight in the boiler house. Put them to work.

Sign up for future bulletins | Energy management training | Ask a question

27 MARCH 2023

ESTIMATING VENTILATION RATES

In a fully mechanically-ventilated building with air conditioning, you would expect the regression lines for gas (against heating degree days) and electricity (against cooling degree days) to be related. This is because, to a first approximation, the rate of heat flow into or out of the building per degree of inside-outside temperature difference will be the same or at least similar.

Let's put some hypothetical numbers on this: suppose the building's thermal power demand varies by 7.5 kW for each degree that the air temperature changes. If it has a boiler efficiency of 80%, that means its gas demand will vary by 7.5/0.8 = 9.375 kW/K. And if its chillers have a seasonal coefficient of performance of 3.0. electricity demand will vary by 7.5/3.0 = 2.5 kW/K.

There is something interesting about that kW/K (kilowatt per degree) unit of measurement. Multiplying top and bottom by time in we get kilowatt-hours per degree-hour. A degree-hour is 1/24 degree-days so multiplying numerical values in kW/K by 24 converts them to kWh/degree-day. For our numerical example for gas the 9.375 kW/K becomes 225 kWh/degree-day and for electricity 2.5 kW/K becomes 60 kWh/degree-day. When we carry out regression analyses, these numbers in kWh per degree day are the slopes that we would expect to see.

Of course it would probably be rare for us to know the building's heat loss or gain coefficient, but what we can expect is that when comparing regression lines the gas slope should be three or four times the electricity slope (in theory the ratio is the chiller coefficient of performance divided by the boiler efficiency: 3.75 in my example). If the ratio is substantially different, it points to a problem. Maybe the building has a much higher ventilation rate when cooling than when heating; perhaps the chillers' coefficient of performance is much lower than it ought to be.

Sign up for future bulletins | Energy management training | Ask a question

13 MARCH 2023

THE CALORIFIC VALUE TRAP

Every hydrocarbon fuel has two measures of unit energy content: its gross calorific value (GCV, also known as ‘higher heating value’) and its net calorific value (NCV, otherwise known as ‘lower heating value’). The difference relates to the water vapour which is inevitably generated as a product of combustion, with GCV accounting for all the chemical energy contained in the fuel while NCV discounts that part of the fuel's chemical energy which will be lost as latent heat in the water vapour in the exhaust. Methane's GCV is 55.4 MJ/kg for example compared with an NCV of 50 MJ/kg.

I was trained always to think in terms of GCV and that is the usual convention in the UK: most significantly, when we are billed for natural gas, that's the basis on which suppliers state the energy content in kWh.

A problem arises, however, when we are accounting for an enterprise's total energy consumption (as we are required to do for compliance with ESOS and SECR) if part of that consumption is road fuel inferred from vehicle mileages. This is because the UK's Department for Energy Security and Net Zero (DESNZ) provides official conversion factors (note 1) which convert miles or kilometres into kWh based on NCV. Who knows why? Anyway, the kWh figures so obtained are not compatible with those you will be citing for gas, or indeed for other fuels where you know the quantity and have tabulated their kWh totals on a GCV basis. The true gross energy content of forecourt petrol is 5.5% higher, and forecourt diesel 6.3% higher, than the net values unhelpfully given in DESNZ's tables.

A similar consideration applies when checking the maintenance records of boilers. Combustion test apparatus is very commonly set up to report efficiencies on a net rather than gross basis, and results based on NCV will be systematically higher than those based on GCV. This isn't a problem as long as you are aware of the basis of reporting. Just remember, if you are checking a natural-gas boiler whose manufacturer's declared full-load efficiency is (say) 93% on a GCV basis, test results on an NCV basis should indicate around 103% (note 2).

Note 1: see their “Greenhouse gas reporting: conversion factors 2022” spreadsheet, “SECR kWh pass & delivery vehs” tab
Note 2: a value over 100% is possible because efficiency is useful energy output divided by energy input, and net calorific value does not represent the total input energy.

Sign up for future bulletins | Energy management training | Ask a question

8 MARCH 2023

DRY BULB, WET BULB

Our recent technical briefing on energy saving in swimming pools naturally focussed a lot on ventilation as a means of controlling the humidity in pool halls. It is a very good example of where we need to understand the physics of moist air if we are to avoid wasting energy or causing other problems by over- or under-ventilating the space.

Suppose I want to maintain an internal dry bulb temperature of 30C in the pool hall when the outside air is at 5C and 90% relative humidity (RH). If that outside air is heated to 30C its relative humidity would fall to about 17%, and adding too much of that will not only absorb excessive heat; it will bring the pool hall RH down to uncomfortable levels. For instance at 30C and 33% RH the wet-bulb temperature is 19C. The wet-bulb temperature is what a swimmer feels when they get out of the pool and if the pool water was at 29C, air that feels like 19C is a bit of a shock.

The basic tool that we use for these calculations is the psychrometric chart, a diagram in which the vertical axis shows the humidity ratio (grams of moisture per kg of dry air) while the horizontal axis shows dry-bulb temperature. Assorted curves, lines and external scales are provided such that, at any point on the chart, parameters such as enthalpy, relative humidity, dewpoint, vapour pressure and wet-bulb temperature can be read off. I have posted a downloadable psychrometric chart at https://vesma.com/downloads/psychrometric_chart.pdf . It is in SI units and uses coloured grids and curves to improve readability when printed on A4.

Sign up for future bulletins | Energy management training | Ask a question

23 JANUARY 2023

POWER FACTOR

One of my clients has been approached by a company selling power-factor correction equipment with a claim that it will save them energy. The reality is that it wouldn't; but it might reduce their electricity bills. Here's how and why.

So what is power factor, and why might it need correcting? The essence is this. Ideally, the alternating current in your supply varies exactly in phase with the alternating voltage, peaks and troughs coinciding. But if you have inductive equipment (motors being the prime culprit) the variation in current will lag behind the variation in voltage and when you multiply the two together at any instant the resulting power will be less than it would be if the current and voltage were in phase. If this ‘real’ power is 93% (say) of what you'd expect (given the nominal voltage and current), the power factor is 0.93. The corollary of this is that to supply any given amount of real power, you will have draw a higher current if the power factor is less than 1.0. For example if your building has a power factor of 0.8, the current you draw from the supply network will be 25% higher than is strictly necessary (I have an explanatory video in case that didn't make sense).

Because it increases the current that you draw, poor power factor increases distribution losses in the public supply network and reduces your headroom in terms of your agreed supply capacity. But it does not affect how much real power is delivered, and that's what your electricity meter registers. However, some items on your electricity bills are denominated in kVA and because these reflect the supply current they will be higher than they need to be. That is why it may be important to 'correct' your power factor, which you can do by installing devices called capacitors. A capacitor is an antidote to all those pesky inductive loads: it is passive component which counteracts their effect. And don't forget that as well as reducing kVA charges, you will win back some supply capacity which may enable business growth without investing in new transformers etc..

If you have a capacitor bank at your main intake, your electrical equipment will continue operating as it did before and your internal cabling and switchgear will still be carrying excess current, so there will be no reduction in losses on your premises and no energy saving. It's all about what your site looks like to the distribution network.

Sign up for future bulletins | Energy management training | Ask a question

5 DECEMEBER 2022

COMBINING CONSUMPTIONS

A trainee on a recent measurement and verification course claimed that she had sometimes added together gas and electricity meters for the purpose of regression analysis because it gave better results. The instructor subsequently asked me for guidance on whether it's legitimate to add together different metered utilities in that manner. The answer is 'yes' in certain circumstances but the subject merits a little further explanation.

Firstly, 'meters' are not always meters. For bulk-delivered fuel consumption might need to be gauged from the change in stock level. Meanwhile, the unmetered difference between a main meter and the total of its submeters would constitute a 'virtual' meter. And virtual meters can be the sum of two or more measured flows: as an energy manager I found that the 'day' and 'night' consumptions on a two-rate meter could be analysed as separate 'meters' with their own characteristics, although usually one would analyse the total. Equally in a dual-fuel boilerhouse the gas and oil consumptions would be combined into a composite flow called 'boiler fuel' and as a further example a glass-melting tank using a combination of gas burners and electrode heaters would be treated likewise.

Anyway, in these last two instances, combining two different utlility flows is not only legitimate but necessary, because the energy sources in question are either substituted for each other in an unpredictable manner or are used in randomly-variable proportions, making the isolated analysis of either energy source meaningless. However, this approach is only warranted when the energy sources are combined for a common purpose, with their joint consumptions assessed against the same driving factor or factors.

Sign up for future bulletins | Energy management training | Ask a question

30 NOVEMBER 2022

PLAN, DO, CHECK, ACT

On two recent assignments supporting consultants, my clients specifically homed in on the concept of the 'plan, do, check, act' cycle (PDCA) which underpins the approach to energy management espoused by ISO50001 (a management-systems standard against which some energy users seek to be certified). PDCA was originally a quality-improvement philosophy that was widely adopted in post-war Japanese manufacturing industry. As a successful approach to quality improvement it became the framework for the ISO9001 quality-management systems standard, and later for other similar standards including ISO50001.

Leaving aside ISO50001, it is important to recognise that there are three quite distinct manifestations of the PDCA method in energy management. Firstly, it has a role in the implementation of individual energy-saving projects. Here technical audits, design, specification and procurement constitute the PLAN element; installation and commissioning are the DO element; verification of savings is the CHECK element and the ACT element covers remedial adjustments if the intervention fails to meet expectations. The ‘check’ and ‘act’ elements can be repeated as a loop thereafter to ensure that savings persist. The essence of PDCA in this context is that it will have multiple singular instances.

Secondly PDCA can be used as the framework for an organisation’s overall energy-management scheme. In this role the PLAN element consists of deciding what resources, procedures, documentation and skills the organisation will need. DO signifies putting those elements in place, assigning roles and setting the scheme to work. CHECK in this context will mean periodically auditing the execution of the scheme and ACT signifies rectifying shortcomings in the operation of the management scheme, and introducing refinements and improvements. We repeat the CHECK-ACT loop to keep the scheme working as intended, but this is on a completely different time-base (say annually against weekly) from the check-act loop that would be appropriate for monitoring energy performance itself, which we will come to in a moment. Indeed the fact that it is energy that is being managed is neither here nor there. The exact same PDCA process would apply to the organisation’s management schemes for quality, safety, or environmental impact, for example.

The third manifestation of PDCA is in routine exception reporting. This is an essential process for detecting and diagnosing (often hidden) excess costs arising out of equipment malfunctions, inappropriate user behaviour and other random adverse occurrences. In this application the CHECK element is repeated regularly and frequently (weekly or even daily), analysing multiple individual streams of consumption. It compares actual consumptions with what they should have been under the prevailing conditions in terms of weather, production throughput, or other relevant measurable driving factors. Deviations can then be reported in order of priority for remedial attention. The ACT step is thus triggered only when warranted. This is the process known as ‘monitoring and targeting’ (see footnote) which, incidentally, also facilitates the objective evaluation of savings, the computation of meaningful performance indicators, and more incisive benchmarking. The PLAN and DO elements of monitoring and targeting consist of cataloguing the available data on consumptions and their associated driving factors, making robust arrangements for data collection, deriving the formulae which enable accurate calculation of expected consumptions, and tuning those formulae to reflect historical best achieved performance.

For the minority seeking compliance with ISO50001, you need to understand that although it claims to be based on a PDCA cycle it does not clearly recognise, or distinguish between, the three PDCA applications outlined above. They are quite distinct and have completely different purposes and requirements. If you are developing or refining an energy management scheme (certified or otherwise) for your organisation, bear this in mind.

Sign up for future bulletins | Energy management training | Ask a question

20 SEPTEMBER

INTERSTITIAL CONDENSATION

When you insulate a building internally, you introduce or exacerbate the risk of interstitial condensation in the outer layers of the structure, for example the brickwork of a masonry wall. The problem arises because water vapour diffuses readily through most building products. When the outer layers of the structure are deprived of heat by the insulation, they end up much closer to the outside air temperature and vapour diffusing from inside the building will then on occasion condense within the thickness of the brick (or other material). Subsequent freezing and thawing of this interstitial damp can cause masonry to crumble, while even without freezing, moisure collecting in concrete can corrode reinforcement bars with the expanding rust cracking the concrete.

To avoid these risks, it is important to put a vapour barrier on the warm side of any internal insulation. This impedes the flow of moisture into the structure, and by reducing its vapour pressure within the outer skin, reduces the dew-point and with it the likelihood of dangerous condensation. Conversely we need to avoid vapour-impermeable layers on the cold side because if vapour does get through for any reason, it will condense on the cold impermeable surface. Hence the preference for breathable roofing felt over insulated lofts: it is usually impracticable to incorporate a vapour barrier with retrofit ceiling insulation.

Now regular readers will know that I am not a fan of "multi-foil" insulation blanket, the thermal effectiveness of which is often exaggerated. But it would have one factor in its favour; the inner foil layer would act as an effective integral vapour barrier. So you can imagine my horror on seeing a new class of product on the advertised on the web: "breathable" multi-foil insulation blanket, which combines unexciting thermal performance with perforated foil layers allowing moisture through to the cold layers of the structure. Unbelievable.

Sign up for future bulletins | Energy management training | Ask a question

5 SEPTEMBER 2022

TAKEN FOR A FUEL?

Five years ago this bulletin featured a bogus energy-saving product which injected hydrogen into internal combustion engines and claimed to improve their performance by a frankly ludicrous 33%. Then last week regular reader Clare C. sent me a link to an advert for a fuel additive which claimed more modest ("up to 20%") savings in automotive and marine engines, and oil-fired appliances. Is this plausible? Let's just think what that implies. As ever, when faced with claims for high percentage savings, we need to ask ourselves where any supposedly lost energy is going. To cut fuel consumption by 20% for a given output implies that we start from a position of using 25% more than we should (the 25 units that we save going from 125 to 100 units of fuel is 20% of the 125).

That's a lot of wasted fuel and could really only occur through incomplete combustion, implying huge quantities of carbon monoxide, black smoke, or indeed unburned fuel pouring from the vehicle tailpipe or boiler chimney. To put it in perspective, if 25% excess unused diesel all came out as soot you would get over a kilogram of it from every six litres of fuel. But even small amounts of black smoke or unburned fuel would be evident, and would hopefully trigger immediate remedial maintenance actions, as should high carbon monoxide levels (although those will only be picked up by routine inspection tests). The point is that all these signs and symptoms can be cured with normal diligent maintenance. The problems that these snake-oil products supposedly address should not exist in the first place.

Sign up for future bulletins | Energy management training | Ask a question

7 JULY 2022

AIR-CON BOLT-ON

Reader Adam F. is plagued by emails from a company selling a bolt-on thermostatic control for split-system air conditioning units. They claim 'up to 40%' savings. Is this plausible?

Now the rate of heat flow into an air-conditioned space is proportional to the outside-inside temperature difference (barring changes in ventilation rate and ignoring solar gain, which I will come back to). Let’s suppose you are maintaining 18°C indoors: the rate of heat inflow when it’s 28°C outside will be double what it is when it is 23°C (a ten-degree differential compared with a five-degree differential).

To maintain steady internal conditions the heat inflow must be balanced by an equal amount of cooling. There are two ways to reduce the energy used for cooling:

  1. reduce the rate of heat inflow; or
  2. improve efficiency or reduce losses in the refrigeration plant which provides the cooling
Solutions based on improved thermostatic control address the first option, and they claim to do so by preventing overshoot whereby the evaporator (indoor unit) continues to cool the space after the set point has been reached and it has turned off. The effect of such overshoot, if it occurs, will be to depress the internal temperature slightly. The heat flow into the space will accordingly increase slightly, balancing the excess cooling that has been supplied. But how significant will the effect be? Ultimately it depends on the impact on average internal temperature over time. Remember that the overshoot will be transitory, but let’s be pessimistic and suppose that it gives an average space temperature that is 0.2°C lower than it need be. With an outside-inside differential of 5 degrees, that would imply only 4% excess heat flow and corresponding cooling load. But this is 4% of quite a low load; if the system were sized for a 20-degree differential a 0.2°C offset in space temperature would be adding only 1% to the load when running at design conditions.

But there is a twist. Overshoot can only occur as the thermostatic control commands the cooling to turn off. This may be quite frequent at low loads, but becomes less so as the load increases and the cooling spends a greater proportion of its time running. So the hotter the weather and the harder the cooling has to work, the less waste there will be in absolute terms, and this smaller absolute waste becomes an infinitesimal percentage of the higher demand. Solar gain, when it occurs, increases the load on the cooling system, which reduces the number of start-stop cycles by lengthening the ‘on’ periods and hence cuts down the opportunities for thermostatic overshoot.

The final thing to bear in mind is that although we have, in this analysis, a range of potential savings from maybe 4% at low load to essentially nil at full load, not much consumption occurs at low load so the potential year-round savings are skewed well away from the 4% figure.

My verdict: plausible savings might be of the order of 1% but only if thermostatic overshoot actually occurs.

Sign up for future bulletins | Energy management training | Ask a question

23 JUNE 2022

BLOWING YOUR PROFITS

This month's 'Star Spot' award has gone to a consultant who spotted completely avoidable waste of compressed air in a factory. Compressed-air systems are both costly to run and prone to hidden losses, so it pays to do what our reader did, namely a walk-round check with his client while the factory was shut for the weekend. They suspected a serious problem because outside working hours there were enough leaks to keep one compressor running almost continuously. As they walked around they found the odd poor joint but when they went outside there was quite a significant hissing from one of the factory's dust collectors.In this installation the life of the filter bags was prolonged by timed short pulses of compressed air inside the individual bag in order to blow or knock the dust off of the outside of the filter. Unfortunately one of the air lines had become detached upstream of the solenoid valve and was continuously discharging compressed air. With this fixed, the load on the duty air compressor dropped dramatically.

Remember also that all the time the air line was disconnected, the filter bag wasn't ever being back-flushed: another perfect example of how energy waste often goes hand in hand with loss of service.

Incidentally, it would have been possible to improvise a measurement of the compressed-air savings quite simply. During a period of no demand, you can shut off the compressors and time how long it takes the pressure of the stored air to drop by one bar. Knowing the volume of receivers on the system allows you to compute the rate of air loss. When the pressure drops by one bar, you have lost the receivers' volume of free air. So for example if you have a 690-litre receiver and it drops from 7 to 6 bar in 30 minutes, the air loss rate is 690/30 litres per minute or the equivalent of over 12,000 cubic metres per year on a continuously-running system. As an efficient compressor takes about 0.1 kWh per m3 of air compressed, that would be wasting 1,200 kWh per year. Repeating the test after attending to leaks would show you how much you had saved; repeating it periodically allows you to monitor for deterioration.

Sign up for future bulletins | Energy management training | Ask a question

10 JUNE 2022

SAVINGS IN ELECTRIC MOTORS

Reader Matt contacted me after seeing a presentation on replacement electric motors. The sales person claimed they could save 50% energy on a like-for-like replacement of an IE3 motor. Matt quite correctly challenged them to explain how they could save 50% on a motor that's 90%+ efficient, and of course they could not give a scientific answer.

They might have meant that their motor technology halved the losses in the motor, taking it from say 90% to 95% efficient. But that would result in about 5.3% saving, not 50%. The only way to reduce the energy consumption of a motor by an order of magnitude is to reduce how much work it does. Actually that is entirely possible; it's what variable-speed control does. On centrifugal fans, for example, a 20% speed reduction almost halves the mechanical power absorbed by the fan and that, thanks to the principle of an energy balance, translates into a corresponding reduction of the power delivered by its motor and hence the power that the motor draws from the mains. So maybe the vendor was not talking about a like-for-like replacement but the replacement of fixed-speed with variable-speed motors. In which case speed-control on the existing motors could be considered as an option.

Anyway, speed control would be one way that the vendor might have achieved their 50% savings. But that would be savings on motor power alone, while their web site trumpets 40-60% savings 'overall' on heating, ventilation and air conditioning systems. Also not completely impossible, but (again invoking the principle of an energy balance) it implies that the building's demand for heating or cooling was reduced by that much. However, short of using the fan-speed control aggressively to cut back the ventilation rate drastically, it is hard to see how that could be achieved.

Sign up for future bulletins | Energy management training | Ask a question

23 MAY 2022

FRIDGE AND FREEZER DOORS

It has always been a staple of energy training related to catering that the doors of fridges and freezers should have tight seals and effective closers, and that walk-in freezers should have insulated strip curtains to supplement the proper door when it needs to be kept open temporarily. Most of us would assume that this advice relates to preventing the ingress of ambient air, but that's not the whole story. When room air gets into a freezer, something like a quarter of the energy needed to cool it down goes into condensing and then freezing the water vapour it was carrying. The amounts involved are not huge: something like 0.02 kWh per cubic metre of air overall. What is significant is that the internal vapour pressure will plunge. So even after the door is closed, ambient moisture will pour in through any gaps in door seals, adding continuous cooling load as the condense-freeze process continues. Meanwhile the resulting ice build-up will be clobbering the energy performance.

It's atmospheric moisture that you need to keep out.

Sign up for future bulletins | Energy management training | Ask a question

5 MAY 2022

DO-IT-YOURSELF ENERGY AUDITS

In an earlier bulletin I introduced the three levels of energy audit (opportunity scans, detailed studies and investment-grade audit), the three aspects (technical, human-factors and procedural) and three approaches (checklist-based, product-focussed and opportunity-led).

Focussing for a moment on technical opportunities one tends to find, in general terms, a spectrum of interventions ranging from quick, unintrusive and cost-effective through to radical and disruptive. The order goes something like this:

If you are an energy end user, there is a lot to be said for doing your own audits at the opportunity-scan level, and bringing in outside advice only for the deeper studies that subsequently result. These are the advantages of doing your own audit: Needless to say your audit can span technical, human-factors and procedural aspects, and if you are subject to ESOS, your work will contribute to compliance.

Sign up for future bulletins | Energy management training | Ask a question

25 APRIL 2022

APRIL'S "STAR SPOT" AWARD

The circulating pumps on large heating systems are generally in pairs, with one running and the other available as a standby. It is then common practice to swap the duty and standby pumps periodically to even out wear and tear. April's award went to an energy consultant who had noticed a sticker on a heating-system control panel instructing people to leave both the duty and standby pumps running continuously. This regime, which had been applied to both primary and secondary circuits, effectively disabled time control and increased the pumping power requirement (and wear) unnecessarily. The justification for the instruction was that during automatic duty-standby changeovers there was a short period with no flow, which caused the boilers to lock out. This had evidently been going on for three years when it was noticed, but the problem was easily eliminated by amending the controls so that outgoing duty pumps did not stop until the standby pumps had started.

Our awards judge said it reminded him of seeing a temporary 'do not switch off' notice on a bank of lights which had been in place 20 years because the facilities manager had retired and nobody wanted to challenge the status quo. If you have a similar story about avoidable energy waste that was easy to resolve please visit http://EnManReg.org/starspot. We'll give a £50 donation to the your chosen charity if you win.

Sign up for future bulletins | Energy management training | Ask a question

1 APRIL 2022

NEGAWATTS AND NET ZERO

Thanks to reader John S., who alerted me to a company which has been trying to raise finance to develop an alternating-current (AC) rechargeable battery technology based on its concept of a ‘biode’, a battery electrode that switches continually between being an anode and cathode. While an AC battery is an intriguing concept in its own right, it is the application to mains supplies that interests me more.

Because the electrons in an AC supply continually flow into and out of the load, rather than continuously in the same direction as with DC, the cumulative net current flow over a given time interval is actually zero. The only reason that energy is consumed is that the voltage also alternates: power is voltage times current, and negative voltages multiplied by negative currents give positive power. But what if one adapted the biode principle for mains power? If you were to blend DC voltage with alternating current you would get alternating power, with the customer feeding back as much energy to the grid as they absorbed, 50 times a second (when negative current is multiplied by positive voltage).

At the power station this blended alternating/direct (BAD) supply could cause problems because obviously the returning energy is never going to recombine CO2 from the atmosphere into fuel (that’s entropy for you). Admittedly a wind turbine for example might be more reversible, and here the returning power could perhaps be absorbed with the blades working half the time as a fan. However, there is another possibility. If we think about existing three-phase AC distribution networks they already work with a net current flow of zero, which is why a star-connected load does not need a neutral wire. That gives us the germ of an idea. If BAD distribution systems were two-phase rather than three-phase, half the loads could be on one phase and half on the other, and they would take turns to feed each other. Alternatively, by dropping the AC frequency to 0.00001157 Hz (one cycle per day) and adopting seven-phase distribution, you could spread the load between customers over the course of a week.

These ideas are going to need massive investment and eliminating power stations would occasion huge disinvestment, but this dichotomy is entirely in line with the “net zero power” philosophy I have described. Alongside the technical breakthrough we can expect major innovations in financing, also based loosely on the biode principle: novel bank deposits that fluctuate between credit and debit but average at nil (so-called alternating current accounts). I shared these thoughts with Extinction Rebellion (motto: “stick the man to it”) and they confirmed it makes net zero sense.

Sign up for future bulletins | Energy management training | Ask a question

25 MARCH 2022

DESK-TOP ANALYSIS PRIOR TO ENERGY AUDITS

Rocketing energy prices are prompting users to ramp up the search for energy-saving opportunities. But before you reach for your hard hat and clipboard, don't forget that there are desktop analysis techniques which could prove invaluable as a precursor.

Let's start with regression analysis, with which we establish the historical relationship between consumption and its driving factor(s). This will give us clues if we see anomalous patterns. Does consumption appear to be weather-related when it shouldn't be? Does it fail to respond to production throughput when logically it ought to vary? Do we seem to have unreasonable levels of fixed consumption? Regression analysis also enables 'parametric' benchmarking which is a simple but more effective variation on the theme.

Cusum analysis, meanwhile, shows us whether past performance has been consistent, and if not, when it changed. When combined with regression analysis it will also show in what manner performance changed. Did we add (or lose) some fixed demand? Or did sensitivity to a driving factor change?

Next, the concept of expected consumption enables the computation of 'performance deficit', meaning the absolute quantity of energy that we are using in excess of achievable minimum requirements. When translated into cost terms this gives us a clear view of where our most valuable opportunities lie.

And finally I could add visualisation of fine-grained consumption patterns. But that is costly, whereas everything else can be done with information collected at daily or weekly intervals.

(a similar article with illustrations and links is at EnManReg.org/prior-analysis)

Sign up for future bulletins | Energy management training | Ask a question

7 MARCH 2022

DIGITAL TWINS

Last week I attended a thought-provoking presentation on digital twinning (DT) by the energy manager at Glasgow University, which has built digital twins for five of its buildings. It's not a topic I know much about but I was interested because, going by what it says on the tin, it sounded like potentially a good tool for what I would call 'discrepancy detection' as a way of saving energy. In other words, spotting when a real building's behaviour deviates from what it should be doing under prevailing circumstances, which will nearly always incur a penalty in excess energy consumption. The other potential benefit of DT to my mind would be the ability to try alternative control strategies on the virtual building to see if they yielded savings, and what adverse impacts there might be on service levels. This would be less intrusive than the default tactic of experimenting on live occupants.

Unfortunately I came away with the impression that we are still a way off achieving these aims. The big obstacle seems to be that DT is not dynamic - it only provides a static model. That surprised me a lot, and if any readers have evidence to the contrary, please get in touch. Another misgiving (and to be fair, the presenter was very candid about these issues) was the cost and difficulty of building and calibrating a detailed virtual model of a building and its systems. Then there is the question of all the potential influencing factors that you cannot afford to measure.

My conclusions are in two parts. One is that simulating the effect of alternative control strategies would have to be done with software short of a full DT implementation, in other words, using much-simplified dynamic block models. The other is that discrepancy detection is probably still best done with conventional monitoring-and-targeting approaches using data at the consumption-meter level, with expected consumption patterns derived empirically from historical observations rather than from theoretical models.

Sign up for future bulletins | Energy management training | Ask a question

28 FEBRUARY 2022

ALL HANDS ON DEC

Reader and former trainee Richard C. wants to know how much energy his prospective client uses but they won't tell him. Not a problem for a UK establishment: as it is a college, it must have a Display Energy Certificate and by visiting this government web site * he can get a copy instantly. Their DEC won't show the total kWh figure but it will give the floor area and kWh/m2 separately for electricity and gas, enabling him to reverse-engineer the totals he wants for every building on their estate that is visited routinely by the public. Better still, if he repeats the calculation not with the actual kWh/m2 figures but reasonable benchmark values, he will have plausible estimates for achievable consumptions and hence the possible savings to be had in kWh terms.

* Thanks to reader Mel M. for providing this link

ADDITIVE AFTERTHOUGHT

Something struck me recently during a boilerhouse survey. There are numerous products which claim to improve the efficiency of wet heating systems by improving heat transfer into the system water (we can ignore heat transfer beyond the boiler, because the quantity of heat delivered is dictated by what the building needs). Some of these products supposedly increase the effective surface area for heat transfer. But even if this were true and a significant effect, what would be the point if the boilers are oversized anyway, which most are? Shortage of heat transfer area is rarely a problem.

Sign up for future bulletins | Energy management training | Ask a question

8 FEBRUARY 2022

YOU CAN'T WIN...

Reader R.S. from South Africa is frustrated that some of his clients have bought into the idea of fitting solar panels into chiller circuits. He fully agrees with my article from five years ago explaining why it cannot work, and also an ASHRAE article from 2020 that comes to the same conclusion but "is not succinct and compelling enough to convince the layperson". When I wrote my piece, the vendor in question was implying (ludicrously) that a solar collector in the refrigerant loop would work as a supplementary compressor. Interestingly, possibly because I pointed out that the solar energy would superheat the refrigerant and reduce the condenser's capacity to transfer latent heat, the product's promotional material is now pushing this superheating effect as if it were beneficial. You can't win.

Sign up for future bulletins | Energy management training | Ask a question

24 JANUARY 2022

ENERGY WASTE AND LOSS OF SERVICE

Quite often, energy waste is associated with loss of service. When I was an energy manager with schools on my patch, we would sometimes get calls to say that kitchen calorifiers (hot-water tanks heated on a loop from the main heating system) were not working. Quite often it was because one of their boilers' burners was locked out, causing that boiler to contribute unheated water to the circulating mix, which reduced the heat supply to the calorifiers. But while this was happening the dead boiler would be acting as a heat exchanger dumping system heat up the chimney. Waste and loss of service together. Next example: de-icing heaters under the front steps of a prestige office, but running continuously all year because of a control fault, resulting incidentally in one of the heaters burning out prematurely. Or how about this, from reader Roger H.: the dust extract system at a factory went through bag filters which were cleaned automatically by periodic pulses of compressed air. The air hose came off upstream of the solenoid valve, causing continuous undetected waste of compressed air... And loss of service.

I could go on, but I want just to mention another link. When equipment is wasting energy by running unnecessarily, it is also wearing out faster than it needs to, risking premature failure with inconvenient nd potentially costly consequences. That's one good argument for having an effective monitoring and targeting regime.

Speaking of which, let me finish on a positive note. I once implemented energy monitoring and targeting on some distillation columns. These were high-pressure, high-temperature vessels which ran continuously between nine-monthly strip-downs for inspection. However, once they could track deviations from expected energy consumption, the owners realised that they could safely increase strip-down intervals by not taking columns out of service unless they had internal problems. And how did they know when a column had internal problems? By spotting incipient loss of energy efficiency.

Sign up for future bulletins | Energy management training | Ask a question

17 JANUARY 2022

A BETTER WAY TO BENCHMARK COMPRESSORS

Someone posed a question on LinkedIn last week related to the efficiency of his compressed air system. He intends to measure the specific energy consumption (SEC, the overall kWh/m3) of each compressor. He has a flow meter that he can deploy and he wanted guidance on the procedure. Now this is interesting because a compressor's SEC is inherently variable. It will be worse at low loads, because there is likely to be some fixed consumption unrelated to throughput. So if he just measures electricity consumption and air output over any given interval the answer that he will get won't be reliable. A compressor that was heavily loaded during its test will tend to outscore one that did not have as much work to do.

He can get around this problem by taking frequent measurements over an extended interval (daily over the course of a month, or hourly over the course of a few days, etc) and then conducting a regression analysis for each machine. This will give him for each compressor (a) its fixed 'overhead' consumption, represented by the intecept on the y axis, and (b) its marginal SEC (how many kWh per additional cubic metre, represented by the slope of the line). Nearly all individual compressors exhibit a straight-line relationship with a gradient of around 0.1 kWh/m3, often with very little scatter.

The point is that the gradient is a constant. That means you can compare the gradients of different compressors' regression lines and it matter not a jot what pattern of load they were under, how big they are, or what ancillary equipment is connected to their electricity meter. If you'd like to know more please see my article on compressor benchmarking at http://EnManReg.org/airbm.

Sign up for future bulletins | Energy management training | Ask a question

Earlier articles...

Click here to see items from earlier years

VESMA.COM
all about saving energy
Pound House
Market Square
Newent
Gloucestershire GL18 1PS

Tel: +44(0)1531 821350

(C) VESMA.COM. Excerpts from this material may be reproduced on condition that you include (a) a link back to this page and (b) this clause.