How to measure CFM from a 3" Fan

[mcrstang]

I am trying to see what the best way to measure the cfm generated from a fan with a brushless motor. The fan is in a 3" circular housing. The fan system is used inline with brake ducting in a race car. I have two issues in trying to measure CFM.

1. Measuring CFM through the 3" ducting statically.
2. Measuring CFM through the 3" ducting under pressure.
Simulating a 200 mph wind force at the enterance of the
fan.

The fan is pulling air from the enterance of the 200 mph wind force, and generating additional force to the brakes through the 3" ducting.

Any ideas on the best approach?

 

[MikeHalloran]

Remove the fan.

At 200 mph, it makes things worse.

Probably also at 40 mph.

Substitute a brushed fan, connect an LED across it with the supply, then disconnect the supply and see at what speed the LED comes on.



Mike Halloran
Pembroke Pines, FL, USA

 

[rb1957]

try a NACA scoop intake (to minimise drag losses)

 

[MiketheEngineer]

At 200 mph - why do you need a fan??

 

[rb1957]

i guess the point is how much ram effect you're getting. maybe try it without the fan and if you need more airflow then add the fan. i haven't designed a NACA inlet but i imagine that there is something in the design that says if you want Xcfm from a Y airspeed then you need an inlet with Z in2 (which hopefully will be close to your 6in2, = 3" dia tube).

the only hassle would be if the scoop isn't facing forward for some reason. then, of course, no ram effect, and you need the fan.

note, mike's point is that the fan is an obstruction to the flow in the tube.

 

[MikeHalloran]

Several cars ago, the harness around the electric cooling fan's relay burned out. ('low-force' GM spade connectors, intergranular corrosion of tin).

I repaired the harness, replaced the relay, added a manual switch (auto/on/off), and stuck a resistor in the wire carrying power to the fan, connected that to an LED, and grounded the other lead of the LED. So the LED would come on whenever the fan was powered.

Surprise: At >~30mph, with the switch 'off', the LED came on; the wind was motoring the fan.

Conclusions:
You probably only need the fan if you're braking hard while going slow, and it will clearly be an impediment at high speed, even when it's on, because ordinary fans do not produce 200mph airstreams.

But measuring the voltage across the terminals of an unpowered fan in the duct would give you a rough approximation of the velocity in the duct, from which you can infer how well your scoop is working, sort of.




Mike Halloran
Pembroke Pines, FL, USA

 

[mcrstang]

The air inlet is a fixed position on the front nose of the car. therefore a NACA duct is not a possiblily.

The idea of using a brushed fan with an LED is a good one, though the brushless fan spins at 20,000 rpm which I do not have the resources to find a brushed motor with the same specs to check this theory.

What I would like to measure is the output ability of the fan alone, Isolated to see if it has the capabilities to overcome the inlet airflow at speed. Or if it is working against the airflow. Also, I would like to make a fixture to be able to back to back test multiply versions of fans.

One thought would be to construct a set dimension box and make a plastic baloon like bag inside the box. I would then hook up the fan to the bag and run it until it creates a positive pressure in the bag. Observing the time it takes to fill the bag, I should be able to determine the CFM.

Any thoughts?

 

[IRstuff]

??? Wouldn't you just have the fan with an open duct on both sides and measure the mid-stream velocity?

Filling a balloon introduces a back-pressure, which was not stipulated in your OP.

TTFN

FAQ731-376

 

[mcrstang]

How would you recommend to measure the wind stream velocity? The fan creates a vortex in which has different velocities at different points in the air stream.

 

[thruthefence]

Go to the pick'n'pull, find a car with a Bosch L jetronic EFI, get the MAF box & splice it into your 'duct'. Heck, get two, one pre fan, one post fan. There should be plenty of Bosch info out there to do the calculations.

 

[thruthefence]

Correction; should have said "LH Jetronic" the 'hot wire' version.

 

[rb1957]

i like the idea of the balloon, but then it'll need to be pretty damn big ... a 200 mph = 300 fps; 3" dia = 0.05ft2 ... gives you 15cfm, and presumably you want more than that

 

[imcjoek]

Calculate pitch speed?

On a 3.0" diameter fan with ~30 deg blade angle (like a typical CPU fan), I am finding a pitch speed of about 50mph @ 20,000 rpm.

So unless the motor can produce a positive net torque at 80,000 RPM, the fan will be acting as a regenerative system...

My math may be fuzzy, but the concept I think will show that you need to go a lot faster, or a lot bigger.

 

[IRstuff]

That balloon needs to be a lot bigger, anyway. 200mph*3in_diam_area = 15 cfs = 864 cfm.

TTFN

FAQ731-376

 

[FeX32]

Disregarding the fact that you are using the fan for the application you indicate, your question was simply how to measure the flow speed.
Why not use an anemometer? The turbulence (or unevenness as you say would not matter really)

If you can't get an anemometer then you can use a manometer. This can relate the pressure change to the dynamic pressure which in turn is related to the free stream velocity.



Fe