Every fan’s air perfor­mance, noise and effi­ciency depend on the geom­etry of the impeller, the housing compo­nents, the speed and the diam­eter. Theo­ret­i­cally, both centrifugal and axial fans can be used in air handling units (AHUs). In both designs, the air is drawn parallel to the axis of rota­tion. While the outflow for axial fans is primarily parallel to the axis, the air in centrifugal fans mainly escapes outwards and radi­ally from the center, i.e. centrifu­gally.

In both designs, the air flow is also directly propor­tional to the speed, while the pres­sure increase rises with the square of the speed. Which is the better choice depends on the appli­ca­tion involved. The instal­la­tion and oper­ating condi­tions are impor­tant factors for the fan to work at optimum effi­ciency and with minimal noise emis­sions.

Axial fans: a sensible solu­tion?

Fig. 1: Differ­ences in the char­ac­ter­istic curves of centrifugal and axial fans. (Graph | ebm-papst)

Axial fans work most effi­ciently when they convey air – for example via a heat exchanger – into the open at low back pres­sures. To achieve the best possible effi­ciency, an axial fan’s impeller should be posi­tioned in a fan housing that has been aero­dy­nam­i­cally opti­mized. Together with a front plate, this ensures the correct flow control and provides the sepa­ra­tion required between the intake and outlet sides.

The char­ac­ter­istic curve of a typical axial fan is marked in blue in Fig. 1. There­fore, axial fans achieve optimum effi­ciency at high air flows and do so with minimal noise emis­sions. Axial fans are sensi­tive to inflow fluc­tu­a­tions without addi­tional measures, such as an upstream guide vane. It is also often useful to have a discharge vane on the outlet side to opti­mize effi­ciency, which means more mechan­ical effort during the assembly and an increased overall length.

The outflow char­ac­ter­is­tics of axial fans are very focused compared to centrifugal fans, which is a disad­van­tage when applying air to down­stream filters or heat exchangers. However, if the outflow goes directly into a duct network, this can be advan­ta­geous (Fig. 2).

Centrifugal fans: the better choice

Centrifugal impellers are intrin­si­cally less sensi­tive to influ­ences on the inflow and outflow sides. The devel­op­ment of RadiPac centrifugal fans from ebm-papst, specially designed for instal­la­tion in AHUs, not only involved opti­mizing the energy effi­ciency and noise emis­sions of the impeller, motor, control elec­tronics and housing: it also involved consid­ering the actual instal­la­tion condi­tions in AHUs. The result is clear: this choice of fan does not require large reserves to be prepared for instal­la­tion losses. Further­more, centrifugal fans without a scroll housing are partic­u­larly flex­ible when it comes to outflow vari­ants out of the AHU. Since the fans apply air to a pres­sure chamber (pres­sure plenum), it is possible to connect to a duct network in virtu­ally all direc­tions without signif­i­cant losses.

Fig. 2: Axial and centrifugal fans can be used in air handling units (AHUs). Centrifugal fans (left figure) ensure that the air is applied more evenly to down­stream fittings (filters, heat exchangers) than with axial fans (right figure). The axial speed is presented in color. (Image | ebm-papst)

What is the defi­n­i­tion of effi­ciency?

Theo­ret­i­cally, effi­ciency is defined using the quotient of output over input. In venti­lation tech­nology, the para­me­ters are air conveying perfor­mance (air flow x pres­sure increase) divided by elec­trical power consump­tion. However, this infor­ma­tion alone does not guar­antee that the manufacturer’s spec­i­fi­ca­tions will be compa­rable. First of all, it is impor­tant to deter­mine which compo­nents the fan concerned contains. If there is only a fan impeller, the effi­ciency values cannot be compared to the values of a complete fan consisting of control elec­tronics (VSD), a motor and a fan impeller.

To obtain real­istic infor­ma­tion about effi­ciency levels, the entire fan unit has to be measured as a whole.

Simply multi­plying the indi­vidual effi­ciency levels of the various fan compo­nents at their optimum point is not enough either. Although this is often done in prac­tice, you cannot expect all the compo­nents used to work at their optimum effi­ciency when they are put together, espe­cially consid­ering that compo­nent manu­fac­turers often only provide optimum effi­ciency values. It is diffi­cult to obtain values for partial load behavior at a reduced speed. To obtain real­istic infor­ma­tion about effi­ciency levels, the entire fan unit has to be measured as a whole.

η_Fan≠η_max⁡Motor *η_{maxIm­peller} *η_{maxCon­trol elec­tronics}

Back to the defi­n­i­tion of effi­ciency. Air perfor­mance is defined by multi­plying the air volume and the pres­sure increase. The air flow, i.e. the air volume, is provided by the air conveying task. The pres­sure increase required is found by deter­mining the compo­nents that the air flow passes through, such as filters, heat exchangers and the connected air path. As a sum, this is the overall pres­sure. It is spec­i­fied as static pres­sure and is used to select the fan.

Nowa­days, the terms total pres­sure and total pres­sure increase are also common. The total pres­sure is the sum of the static pres­sure and the dynamic pres­sure and, there­fore, is always higher than the static pres­sure alone. There­fore, this infor­ma­tion should be treated care­fully when comparing the effi­ciency of fan systems. Because:

q_V*pstat<q_V*(p_stat+p_dyn)

There­fore also

η_stat<η_tot

There­fore, different fan systems have to be made compa­rable before you can real­is­ti­cally compare their energy values. First of all, as described above, this involves how the fans are composed and the defi­n­i­tion of the pres­sures used in the calcu­la­tion. Instead of talking about percent­ages of effi­ciency, it is better to compare fans for a defined air conveying task using the expected power consump­tion. It is also impor­tant to eval­uate how the fan will react with its surround­ings when installed. These vari­ables, called instal­la­tion losses (system effects), can become rather impor­tant for different fan types and instal­la­tion condi­tions, and must be added to the required (static) overall pres­sure increase when selecting the fan.

Fig. 3: The figure shows an axial fan’s char­ac­ter­istic curve with and without a diffuser. (Graph | ebm-papst)

Here, axial fans perform much worse with a very high partial speed level than centrifugal fans without a scroll housing. Measure­ments by ebm-papst have shown that even the best axial fans on the market do not achieve the overall effi­ciency levels or the low noise levels of centrifugal fans (Fig. 3). However, the effi­ciency and acoustics of axial fans can be signif­i­cantly improved using combi­na­tions of sound absorbers, diffusers or guide blades mounted on the outlet side. Yet, even with these measures, which signif­i­cantly increase the overall length, you are best off with a common centrifugal fan.

Centrifugal fans with modern EC tech­nology

The power consump­tion values spec­i­fied in tech­nical docu­ments are impor­tant when selecting fans. You also need to correctly inter­pret the spec­i­fied effi­ciency levels. With axial fans, you also need to consider the fact that the aero­dy­nam­i­cally instable oper­ating range (stall area) is very close to the fan’s optimum effi­ciency. If the system curve changes to higher pres­sures, this can have a devas­tating effect on the device’s oper­ating safety and, there­fore, on the entire system.