System Effects

System Effects – Are They Real, or Just the Fan Manufacturer’s Excuse?
 
Almost every week, we get a call from a customer saying that their fan doesn’t perform to the catalog rating. Once we have the customer do normal trouble-shooting – confirm the fan speed, confirm the fan rotation, confirm that all system dampers and devices are set at their correct locations, etc., we then try to get some system information. This includes the flow and pressure readings they took, along with any assumptions and non-standard operating conditions. These readings are then plotted on the fan curve in an attempt to determine what may be causing the fan performance variation.
 
Everyone knows that the fan catalog rating is based on the optimal test laboratory conditions. After all, it is a competitive world, and catalog ratings need to be at their highest allowable levels. The first thing to check about the catalog rating is the fine print at the bottom. It will include statements that describe how the fan was tested.  It might say, for example, Installation Type D: Ducted Inlet / Ducted Outlet. If the fan was installed in an alternate configuration, then an adjustment must be made to the expected ratings.
 
There are a couple types of problems that often show up at this point in the analysis.  The first is that the plotted operating point falls within near the fan curve, but not at the expected flow point. Most of the time, it is at a higher pressure value and lower flow rate, which indicates that the system resistance is higher than expected. Once the customer checks his or her system, and corrections are made, the fan performance usually comes back to the expected levels.
 
In some cases, the measured flow and pressure indicate that the fan is operating in an unstable portion of the fan curve. When that happens, either the system resistance must be reduced, or a different fan installed.
 
If one of the measured points doesn’t fall on the fan curve, a system effect is taking place. Many people think that system effects are something that the fan industry has created to allow poorly performing fans to be manufactured. Rather, it is the response of the fan to uneven airflow distributions, and is very real. As a matter of fact, some of the highest performing fans are the most susceptible to system effects.
 
The detailed information on system effects is documented in AMCA Publication 201 – Fans and Systems. This document is available from the AMCA website (www.amca.org). It includes many variations of inlet and outlet ductwork connections, and explains how to estimate the effects on the fan performance. Most of the effects in the field are a result of inlet elbows or outlet transitions.
 
Let’s look at a few examples. Photo 1 shows a discharge elbow on a centrifugal fan that turns opposite the direction of the wheel rotation. In the photo, the shaft is rotating clockwise. Air has momentum, and it wants to continue to rotate up and to the right as it discharges the fan housing. Placing a counter-rotating elbow on the discharge not only breaks up that momentum, but also makes it almost impossible to estimate the airflow through the elbow. The published elbow pressure drop values assume uniformly distributed airflow into their inlet, and this installation will have anything but that condition. This condition may degrade the fan performance by as much as three velocity pressures!!
 
Photo 1
 
Photo 2 shows a severe horizontal drop into the fan inlet. This will push all the air towards the bottom of the inlet duct, and load the wheel unevenly. Remember, the catalog rating is based on an even airflow distribution into the fan inlet.
 
Photo 2
 
Photo 3 shows another type of discharge elbow, but rather then turning opposite the direction of rotation, it turns out over the inlet connection. This condition, although not as severe as photo 1, can be as much as a two velocity pressure loss. Also, if the system pressure drop was calculated based on two elbows, this back-to-back setup will have a much higher pressure drop than expected. Like a fan catalog, the performance ratings on duct fittings are based on clean inlet airflow.
 
Photo 3
 
Photo 4 has all kinds of problems, on both the inlet and outlet.
 
Photo 4
 
Earlier this year, we got a call from a customer that his performance was about 50% of what was expected. This application was for a material handling fan, with inlet ductwork that was correctly sized, and had the first elbow almost 40 feet upstream of the fan inlet. After reviewing all other options, we asked the customer to do a pitot traverse of the fan inlet, hoping that it would show us what was wrong. When the results came back, we found that all the air was coming into one side of the fan inlet.  The customer couldn’t understand why that was happening, as the inlet ductwork was “picture perfect.” After some arm twisting, we convinced him to put a horizontal splitter immediately downstream of the last elbow before the fan inlet. The splitter was about three duct diameters long, oriented so that the splitter kept the airflow from bouncing back and forth across the duct. The customer said they would try it, but if it didn’t work, they wanted us to pay for the work, as well as conduct any more ‘experiments’ at our expense. Luckily, the splitter solved their problem.
 
System effects are a well-known phenomenon, but not well accepted in the marketplace. If the system designer takes them into account, the system will perform closer to the desired value. If not, be very careful about speeding up the fan without checking with the manufacturer first. You may exceed the allowable tip-speed, and have problems much greater than a low airflow!!