What happens to CFM with one fan out in a 2-fan system? 2

[aeroUB]

If you have a two-fan system that are pulling air at a known CFM, what happens when one is shutdown and blocked off so that air is not allowed to enter through the stopped fan?

There are two identical fans drawings air through a unit with an internal (before the fans) static pressure of 0.832" wg. The fans are run at a given rpm now matter what the static pressure is. Is there a way to approx. (with info to back it up) on how much air (even % wise) only one fan would pull with everything else the same? Obviously the static pressure will drop and I believe that a single fan will move more than 50% of the original air based on the fact that Static pressure is proportional to Flow^2. At 1/2 the flow, static pressure would be about 1/4.

Any ideas? Or is this totally dependent on the design of the fans themselves?

Thanks
aero

 

[KiwiMace]

Fans in Parallel, constant RPM: This depends on how steep your system curve is when plotted on the fan curves. I would say with your application data, you will be getting much less than a 50% reduction. Helps to have a curve in front of you as you read the below:

In free air (low gradient system curve, low static) situations, you start to approach doubling the airflow for turning on the second fan. Plotted on a fan curve, there is a big gap (on the x-axis) between where the system curve crosses the one and two fan curves. Therefore, a large change in flow for two, going to one fan operation, up to 50% reduction.

As you add longer ductwork or otherwise increase the static pressure, the system curve gets steeper and crosses the fan curves much closer together (x-axis). Much less change in airflow for two going to one fan operation.

 

[ko99]

Yes, there is an easy way to estimate CFM. On a single spreadsheet, draw the following:

1. The fan manufacturer's performance curve (pressure vs flow)

2. The performance curve of your two fans in parallel. It will have the same shape as a single fan except the flow is doubled (pressure doesn't change).

3. The system curve. Mark the operating point (0.832") on the fan curve and draw a line from the origin to this point. System curves have the shape of dP = CFM ^N. N = 1 for laminar (straight line), N = 2 for tubulent. I suspect your flow is primarily turbulent.

Your new operating point (CFM and pressure) is where the system curve passes the single fan curve.

Please beware this assumes the flow distribution through your system isn't altered due to a fan failure. If there are parallel paths to the fans, a disproportionate amount of flow loss (and temperature rise!) will occur in the paths nearest the failed fan. That effect is much harder to estimate.



ko (

http://www.ecooling.biz)

 

[aeroUB]

Great info guys. Ko, that is exactly what I was looking for and after messing around with it for a while I think I got it. With the fans i'm using for this example I'm coming up with approx. 71% of the airflow with one fan compaired to the two fans. Also a TOTAL pressure (static + velocity) change from 1.01" wg to 0.853" wg. Seems like reasonable answers.

PART 2:

Say the second fan stopped and was NOT isolated. If i assumed it was just an open 'hole' to the atm. with known dimentions...how could I go about finding the new internal pressure in the unit then? Obviously the TOTAL pressure would really fall off then but how could you approx. this situation?

Thanks a lot for your input guys...very helpful!
aero

 

[KiwiMace]

This is a matter of estimating the duct system pressure drop based on the new arrangement of one fan supplying air to a parallel duct network with two paths:
1. The existing duct
2. The short-circuit back through the other fan

This will be a new system curve for a single fan. If the fans are located adjacent to each other, the short circuit will dominate, and you'll lose most of your airflow back through the stopped fan.

Backdraft dampers almost always required in this situation.

 

[quark]

Good suggestions by KO and CinciMace. You can make some trial and errors to estimate what will be your new cfm when you go with a single fan, provided you know the static pressure and flowrate at common discharge of the two fans, when both are running.

Once you get the values, use DP1/DP2 = (CFM1/CFM2)² to check DP for any lower flow rates. From the performance curve of the fan, check the maximum possible cfm that a single fan can deliver and check the corresponding static pressure. If this matches with that you obtained from the proportional equation, you are close to your answer. If the static from the performance curve is lower, go for lower values of cfm and check corresponding static pressures. A small calculation in excel may take about a minute time.

The main drawback of this method is that we assume only dynamic losses in the system. Nevertheless, this will give you an approximation.

ko99,

Can you please explain, why for laminar flow dp is proportional to cfm and not cfm²