Why - Fan Speed Control?

In this series of posts of "Why?" - I hope to answer the questions that people in our industry might have regarding why we do things. And one of the I have wanted to discuss for quite a while, are the Affinity Laws. but perhaps I'm getting ahead of myself, because to discuss those, I have to discuss everything else.

For anybody who has ever wondered why condenser fans (and evaporator fans) are fan speed controlled – this "little" piece is for you.

In this post we’re going to cover:

• Why do we need fans in the first place?
• Why do we need to regulate condenser capacity?
• How do we regulate condenser capacity?
• Why we might want to choose variable speed fans over fixed speed fans?
• What other benefits speed controlling fans have over fixed speed fans (if any)?

So lets jump straight in.

Why do we need Condenser Fans?

We need fans on an air-cooled condenser to draw (or blow) through cool(er) air over the condenser fin pack (secondary heat transfer area), to enable the condensing circuit to reject sufficient heat from the refrigeration cycle so that the refrigerant discharge gas condenses into a saturated/subcooled liquid.

If we’re not taking away enough heat at any given time, the greater the percentage of refrigerant is a gas, and the discharge pressure increases until equilibrium is found as the saturated condensing temperature increases the condenser capacity*, or the system hits the safety buffers of the High Pressure Safety Switch, and all refrigeration stops. The same principle will apply to water cooled systems.

There may be the occasional time in retail refrigeration (very low ambient temperatures, and low duty) where natural convection is draws air through, well, naturally, but we’re not discussing that or for that matter, a domestic fridge/freezer which relies entirely on natural convection for its entire life span and is effectively unregulated.

Unregulated condensers will work by having a relatively constant environmental conditions, such as a kitchen. Where we have a much wider range of environmental conditions – we need to regulate condenser capacity.

Here we see the annual hourly temperature range for three locations range from -7°C through to above 29°C, which an air-cooled condenser may be exposed to. The under-counter fridge in my kitchen spends the majority of its 8760 hours a year at temperatures between 16-25°C.

At last years IIR Extreme Heat Briefing - one of the presentations demonstrated that the UK despite our preconceptions, has one of the widest range of temperatures, and our systems need to be designed accordingly.

Why do we need to regulate condenser capacity?

Why do we need to regulate condenser capacity, why can’t we simply have condenser fans switched on and off, by the same relay is turning the compressor on and off? In a lot of cases, such control may suffice - but in retail refrigeration this can (will?) create problems – around a condition called “over-condensing”.

When I was squeaky voiced applications engineer – I dealt entirely with compressors, and to a lesser extent, air-cooled condensers. As we know – the lower the condensing temperature, the higher the CoP – so how the can “over-condensing” be an issue? When I first came across the time, I was a little stumped. Then I forgot about it and got on with my life (which was mostly playing Grand Theft Auto San Andreas), until I had to start selecting expansion devices.

Expansion devices, which can be anything from a length of pipe, a Thermostatic Expansion Valve (TEV), a “Stepper” electronic expansion valve, or a solenoid valve and ejectors, have capacities which depend largely on the pressure differential between the inlet and outlet.? You have a very high of pressure on the inlet and a very low pressure on the outlet, you are going to get more flow through the orifice, than you would if you have a much lower pressure differential.

And it’s that flow of refrigerant that ensures that we keep food cold. Too little refrigerant through the expansion valve – and the product gets (too) warm, regardless of what the compressor is or isn’t doing.

There are also other effects of over condensing where liquid refrigerant remains in the condenser, which may cause nuisance low level liquid alarms or in the worst case, potentially cause liquid starvation to the evaporating fixtures. With saturated gas defrosts, there will be the requirement for a minimum condensing temperature, which in turn will compromise defrosts if that minimum condensing temperature threshold is breached.

To appropriately regulate our condenser operation so that we work safely, efficiently, and effectively in all expected conditions - we need to define:

• A maximum condensing temperature (so we don’t break things, or more bluntly - kill or inflict life changing injuries on people);
• A nominal condensing temperature (so we can define a system’s efficiency);
• A minimum condensing temperature (so our expansion devices always have sufficient capacity for our duty conditions, etc).

The above PH Diagram showing illustrative condensing envelopes for a typical system. As always, things will be different depending on system design.

Over-condensing** and over-subcooling can create other issues, but frankly, I’m sure that’s all a little bit of Dick Simmel’s black art. Just define your system and environmental conditions accurately, and pass it to the next person, who happens to know how to circuit evaporators and design refrigerant distributors!

How do we regulate condenser capacity?

In much the same way as we regulate compressor capacity. In its most basic form, we will have condenser fans switched by a pressure switch, that cuts in (on) at a certain condensing pressure and cuts out (off) at another condensing pressure (i.e. minimum condensing temperature). The more air you have going through the condenser, the greater the condenser capacity.

Multiple Fixed Speed Fan Control

If we have more than one fan, you could have all fans switched by the same pressure switch or have them switched by individual switches (set to different pressures) to form stages of regulation. With the advent of PI&D controllers (Proportional, Integration, & Differentiation), condenser capacity could be much more intelligently regulated, should there be a gentle increase in condensing pressure, the fan response might be gentle. Should there be a rapid increase in condensing pressure, fan response should be rapid (e.g. instead of bringing on one fan stage, a PI&D controller may bring on all fan stages at once for example).

This approach is simple, cheap, and effective for decades.

Types of Variable Speed Controls.

The other approach to switching a fixed speed fan (or groups of fans) on and off, is to vary the rotational speed of a fan. What you might see on a desk fan is slow, medium, and fast – we can also see on refrigeration system, or the use of TRIAC voltage choppers, Variable Frequency Drives (VFDs), or Electronically Commutated Fans, and all three measures through different mechanisms vary the speed of fans.

But why? Let’s have a look of an example. We take the same fan (that is capable of variable speed), and we take the same average fan speed. One operates at 100% speed for 30 minutes out of the hour, and the other runs at a constant 50% speed for 60 minutes.

?Which consumes the most energy? We will assume that the hourly airflow is the same (it’s not, but bear with me)?

If you answered the variable speed fan consumes the least amount of energy – yes, you are correct. There is, after all, a reason why we spend more on fans and VFDs than we do on simple switches and relays. And in this example by halving the fan speed, the shaft power*** is reduced by 87% and over the hour, despite the fixed speed fan being off for 30 minutes, the energy consumption would be reduced by 75% for the same total airflow per hour.

Why is this? Because of the Affinity Laws.

The Affinity Laws and Fan Speed Control

The Affinity Laws, otherwise known as “The Fan Laws” or the “Pump Laws” (as it applies equally to pumps) is a series of laws that predict impeller (the thing that pushes the fluid) performance approximately.

There are three laws, which with everything else constant, they are:

1. Flow is proportional to impeller shaft speed (or diameter).
2. Head pressure developed is proportional to square of impeller shaft speed (or diameter).
3. Shaft power is proportional to the cube of impeller shaft speed (or diameter).

If you’re a visual learner, look on!

Law 1 -Flow is proportional to impeller shaft speed

So here is Law 1 shown on a graph. What you see is what happens when we speed up our fan from 600 rpm to 1200 rpm, we will get approximately twice the airflow, the change is linear. And frankly for most applications, heat exchanger duty or temperature difference is proportional to the amount of air going over the condenser.

Law 2 - Head pressure developed is proportional to square of impeller shaft speed

With Law 2 - we see that the increase in head pressure developed by the impeller exponentially increases with fan speed (to the power of 2). At twice the fan speed, we have a fan that develops four times the head pressure than its original 600 rpm speed.

Law 3 - Shaft power is proportional to the cube of impeller shaft speed

Law 3 shows an even greater change. Doubling the fan speed increases the fan shaft power (not motor power) by 8 times. See here for James May talking about the relationship that speed has to drag and power. The general concept is the same.

Applying the Affinity Laws from Nominal Design case to variable load, variable ambient.

But perhaps this graph shows the benefit better as we of course look to see if we can reduce design fan speeds, rather than increase them as we normally start from the nominal full load and hot weather assumption, rather than the low ambient part load usage case.

A potential 99% power reduction to push 20% of the air flow. We can also see the absolute savings in the mid speed range.

Fundamentally, that is why we have fan speed control.

Other factors not readily apparent

There are many other benefits to speed controlling fans, that aren't just physics.

Design and Specification

Not only do we save energy (a lot of it), and that saves operating expenditure, but we also, as a result are likely quieter**** (if not just based on absolute sound power levels, but also by a reduction in intermittency), but we should also have a lower average velocity onto the face of the condenser.

This lower velocity through the condenser fin pack has two implications.

1. Calculating pressure drops, and in particular the velocity pressure has the velocity of the fluid squared in the calculation (much in the same way one might calculate kinetic energy. This means that if our actual fluid velocity through a fin pack is less than the design velocity, (or any obstruction), the related pressure drops significantly more (by the square root, so if the velocity halves, the pressure drop is quartered.) And if we don't need the same head pressure from the fan, as per Affinity Law 2 - the shaft speed can reduce and furthermore, Law 3 shows than any reduction shaft speed reduces power by the cube root.
2. The lower the face velocity of air, should reduce the rate of pollen and particulate matter accumulating on the condenser fin surface. As you might not know, condensers (and evaporators for that matter) will gradually become clogged in operation by particulate matter and the amount of air drawn through it becomes reduced (increasing condensing temperature), or the fan has to increase shaft speed accordingly to overcome the increased pressure drop (as per Law 2).

With regards to this specific issue issue - multi-fan condensers (excepting some packaged refrigeration units), have the fans installed in modular sections. Along with the other benefits that brings, sections are designed to stop short circuiting on air when a single fan malfunctions. As a result, this will mean that those sections of the condenser will likely suffer from worse clogging and further more require even more fan power (by switching on more fans) or increased condensing temperatures due to a reduction in air flow.

Operation and Maintenance

Other factors which aren’t readily apparent from a design or procurement perspective is that variable speed fans will significantly reduce wear on the electrical switchgear (due to less switching) and the stress on the fan motors themselves is reduces, as the starting current and frequency of starts are removed (the same can apply to many variable speed compressors).

Appropriately commissioned variable speed fans will also improve system stability, as it more closely matches condenser capacity to the actual Total Heat of Rejection (THR), rather than having say condenser capacity regulated in 20% steps, which improves reliability and efficiency.

Concluding thoughts

When it comes to regulating the speed of fans on condensers, it is really is a nailed-on choice for many applications, including condensers, especially when we factor in diversity of load.

We now know, if we didn’t already:

• Why do we need condenser fans on air cooled condensers.
• Why do we need to regulate condenser capacity, to prevent high pressure events and issues with over-condensing and compressor envelopes.
• Typical methods of regulating condenser capacity.
• Why the industry has moved towards variable speed fans over fixed speed fans.
• Some of the other benefits speed controlling fans will have over fixed speed fans.

Thank you for reading to the end, hopefully it you found it interesting and illuminating. I would also like to thank Kevin Mullis for his input to this post. Do give him a follow on LinkedIN, something interesting or useful will appear your feed.

End Notes

*or vice versa if we’re taking away more heat than being put into the system.

**Sometimes natural convection and a large condenser surface area will create over-condensing even with the fans off. In some installations, you might find condensers with two circuits, and with one circuit isolated during winter, or, for condensing units a high tolerance part is machined to fit a portion of the condenser (cough, a piece of cardboard box).

The over-condensing caused by natural convection phenomenon can affect any system that effectively has a wide capacity range, and a wide environmental ambient profile to operate it. An example may be blast freezing or chilling, where the blast duties are potentially an order of magnitude or two above that of “hold” operation, where chilled/frozen product is kept at temperature until unloaded from the freezer

***Shaft power is the power to turn the shaft. The motor imparts this power to the shaft, but there will be losses in the motor and controls. The actual energy input to the motor and controls will always be greater than that consumed at the shaft.

****Certain fan speed control may create noise issues at certain fan speeds, and poorly commissioned systems can create their own noise issues. But in the main, and certainly with EC fans, variable speed is lower noise than their fixed speed counterparts.