Maximizing Airflow: 2 4" Fans on 1 Duct - A Cost-Effective Solution?

[derekpearcy]

Here is an interesting question that arose during some random things I was doing: If you have a 4" duct, and have TWO 4" fans that you want to connect to it, how does that work? Both are 200 cubic feet per minute, and I want something that will do 400 cubic feet a minute. Its about 4x cheaper to do two fans that do 200 cfm as one that does 400 cfm.

The question is if say you have a "T" or a "Y" split of the duct and both fans are running simultaneously, will the two slow each other down by creating some sort of back suction or not?

 

[russ_watters]

Your instinct is correct. Velocity (volume) and pressure are related by Bernoulli's equation, so to get twice the air down a duct without increasing its size requires four times the pressure. That also tells you why the bigger fan is 4x as expensive...

 

[derekpearcy]

I understand the idea with the pressure. So ideally, you want a duct the same size as the fan for it to work properly right?

Also, another aspect of it I was wondering was this: Does having the fan pushing air thru a duct versus pulling air thru a duct make a difference? I noticed that if the duct was even 1" smaller than the fan, I would get a backdraft coming out from the fan if I tried to push it. Pulling the air thru it seemed to work better, but it could be the same in terms of output.

 

[stewartcs]

Hmmm...I would think that if you have two identical fans operating in parallel (assuming free delivery) you should get twice the flow in the common duct and a relatively constant static pressure.

I tend to think of this as being the same as if you have two hydraulic pumps operating in parallel and discharging into a common header. If the pumps output 20 GPM each at 1000 psi, then the total flow rate in the header would be 40 GPM at 1000 psi.

However, the fans (or pumps), in any given system, can only deliver as much air (or liquid) as the system will pass for a given pressure.

Is my reasoning flawed??

 

[russ_watters]

Yes, your reasoning is flawed - and this does work the same for ductwork as it does for pumps. If you header together two pumps without increasing the size of the pipe, you will not get double the flow.

I'm not sure what you mean by "free delivery" though.

 

[stewartcs]

If you header together two pumps without increasing the size of the pipe, you will not get double the flow.

I'm not sure what you mean by "free delivery" though.

I was assuming that the size of the header can accommodate the output of both pumps. Of course if you are limited by the throughput of the header due to the cross-sectional area then the flow will also be limited. Hence I used the term "free delivery" to indicate that you are not limited by the duct or pipe area.

So if you assume the duct or pipe is only capable of flowing the output of one fan or pump then you are correct. However, the OP didn't state that.

 

[russ_watters]

I understand the idea with the pressure. So ideally, you want a duct the same size as the fan for it to work properly right?

Well, you want the duct sized for a certain pressure drop and airflow. The actual diameter of the fan isn't necessarily related to this (ducted fans usually aren't axial anyway, they are centrifugal since axial fans can't generate much pressure).

Also, another aspect of it I was wondering was this: Does having the fan pushing air thru a duct versus pulling air thru a duct make a difference?

Sort of. The pressure difference isn't great enough for there to be a difference in pushing or pulling except when the fan is right at the inlet or outlet. The airflow is shaped kind of like a mushroom - it gets pulled in from all sides and pushed out straight. If you disrupt the suction it can make a difference, but the application really decides what you should do.

 

[RufusDawes]

what type of engineering is this '?

 

[derekpearcy]

I'm making a solar heater, which produces heat inside of an insulated box with a glass window on the front. The fan is to take air from in the house, run it thru the heater, then back into the house. So I have an input and an output duct.

One of the things I did notice is that having the fan pulling air through the ducts means that the hot air is flowing directly over the fan. This lead to fan malfunctioning, as I think it overheated. The input air from the house was about 70 degrees, the output is about 143 degrees. And at 70 cubic feet per minute, that's 5,000 BTU's. =) This unit is only 16" wide and 48" tall. I plan on making bigger units that can push a lot more air. So that is where I'm stumped with the fans, which led to my original question posted.

I'm planning on making one that is 6x bigger. So I need more air flow. I found that 70 cfm cools the heater down a little too fast. So I'm looking at about 10 cfm per every square foot would be more effecient. That way the fan doesn't kick off when it starts pulling the air thru and cooling it down too fast.

I'm planning on building a couple of big 4 foot by 8 foot solar heaters. But that means I'm going to need a fan that's roughly 320 cfm in output. The real thought in question is:

1- using only one fan (big, about 400 cfm and $90)
2- making multiple smaller solar heaters and using 200 cfm fans on each (only $12 each)
3- making a single big solar heater, but linking several fans onto 1 duct (this is the thing that's not going to work I guess)

 

[what-how-why]

If you look at performance curve of a fan you see that delivery volume decreases as back pressure increases. Specification of flow say 400 gpm or 600 gpm corresponds to just one point on the curve, which means the spec throughput is defined just under one particular pressure.
The actual back pressure depends on the down stream system, ie, the duct, the what ever after the delivery header. this facter is independent from the fan itself.
When both fans operate in parallel, there is a tendency that double volume flow will be pumped through, but as throughput increases, back pressure will increase from down stream ystem, which in turn tends reduce throughput. So by operating two fans in paraleel you won't get double throughput. Actual throughput you can get in the end depends on 1) pressrue-throughput curve of down stream system and 2) pressure throughput curve of fans. The intersection of both above curves is what pressrue and throughput you get in the end

 

[russ_watters]

Sorry, this thread kinda fell of my radar...

what type of engineering is this '?

Fluid dynamics is part of aerospace and mechanical engineering.

 

[russ_watters]

3- making a single big solar heater, but linking several fans onto 1 duct (this is the thing that's not going to work I guess)

If you are going to make a big heater and want to use multiple fans, you simply need to have each fan in its own duct, with a separate connection to the heater.