How to match a motor to a fan?

My application is a robot with an upward-pointing fan on top, which tracks a balloon up overhead, and herds it to a specified corner of the room.

I’m assuming (randomly) that a 100cfm fan would do the job, so I found some 4" ones that produce that at 3000rpm:

I found some interesting articles about the conversions and thought process to do this here

The first one basically says, “hp = ft.-lbs. * RPM / 5252”, which I’m guessing gives ~1/4 hp. at 3000 RPM (with a 4:1 gearbox?) for this fan:

The fan web pages say they only need a motor with 1/250th of a horsepower, so it looks like my calculations above are off by 2 orders-of-magnitude :frowning:.

Please let me know how to identify what motor+gearbox ratio will be appropriate to drive a 4" fan such as the one above at 3000 RPM as a constant load (until my batteries go dead :sunglasses: ).



Would you also need to consider how you are directing the air? Is there a shroud or cone directing the air?

I can put the fan inside a 3-6" piece of clear 4"-diameter tubing to focus the fan’s output a bit.

I calculated that 1.3-3 oz.-in. @ 3000 RPM is the torque the fan manufacturer recommends (converted from 1/250th of a horsepower).

Which motor (likely geared down) will be a good match to generate a continuous 1.5-3 of torque @ 3000 RPM without overheating?

And would (well-lubricated) plastic gears last very long for continuous use, or is a metal gearbox the only way to go?



You could look at ducted fans for the RC world. Electric Ducted fans or EDF’s are used for jets. It might give you a lot of options. In my mind these could give you better response and can work off the shelf. Some come in cool colors too :slight_smile: … tegory=153

Thanks much. I asked a co-worker about this, and he recommended a RC store, which I’ll visit this morning to see and learn about what EDF’s they have there.

If I can find a “weak” EDF, I’m imagining I’ll get 4 of them and mount them in a box-configuration pointing up and outwards from a common center, like the fingers of a hand reaching up to catch a basketball.

I’m imagining that this arrangement would be the most stable for catching/holding/suspending/guiding a balloon.

More news on this after my visit to the RC store,


Actually, you want just one fan blowing straight into the balloon.

You can prove this to yourself by blowing up and into the balloon with just your breath. Blow steadily, not it spurts. You will find that a steady stream of air exerts a great deal of control, such that you can even have it 30deg or more off center and the balloon wont fall out of the stream. (This makes a great variation on the common kids’ party game of “keep the balloon up.”)

Here is a really great explanation video:

And yes, you can float a ping pong ball with just your lungs also. It is notably more difficult than a balloon.

Thanks for the tip on using a single fan.

I returned from the store with their cheapest ducted fan, (tiny) controller board, and LiPo battery.

The Controller is called a ICS-300, made by GWS.

Their website is They assume RC use, so they don’t say what waveform their ICS-300 expects. Instead, they just say, “Plug it into your RC receiver.”

What voltage and PWM range do these RC receivers generate?

The guy at the at the store suggested ~50Hz at 5V, but I thought I’d seek a second opinion what I should have my microcontroller send it before accidentally smoking the part.

To cut to the chase, which sort of uC output pin should I hook this to, and what should my program direct that pin to generate?

Thanks much,


You control it the same as a servo. 50Hz is correct, with 1-2ms pulses. 5V is fine for the signal, and it should work down to (nominal) 3.3V.

The ICS-300 is also a BEC, so it’s power wire on the three pin header is designed to supply power to the receiver. If you don’t want that, you need to disconnect that line.

This ESCs has a simple safety features built in, such that it wont start running until it receives lowest throttle position (probably ~1ms) for about a second.