Rough environment rotation speed sensor

How about a reed switch and a magnet? If the operating temperature range of your sensor doesn't extend to 100 degrees, you can insulate the sensor and it will still work.

Thanks for the feedback Tekmeister

I would need to attach the magnet to the blade, which would bring the fan out of balance.

Although placing 3 or more magnets 120º each could solve this, the fan performance/airflow would be influenced. As this is a fan to be run at a determined airflow, the magnets would spoil the function.

Are there other ideas?

Optical might work, but I'd be looking at using a laser.

Any chance you can swap the fan motor out to one that provides feedback via a 3rd wire?

If you do choose to go down the Magnetic route, you might want to look at Hall Effect magnetic sensors over reed switches. The solid state nature of a Hall effect device will typically outlive the mechanical operation of a reed switch, Especially with the number of operations you are expecting.

A reed switch may have issues keeping up with the high speed switching if operating three times per revolution.

I am worried that the steam will interfere

If the pipe is that full of steam, I think the plastic blades will be the biggest issue.

Many of the sensors use IR and work at very close distances.

You might need to provide an IR reflective surface on the blades, which will ensure the contrast to the background is increased.

I'm sure that a sender on one side, and receiver on the other side of the blade will be more than suitable to overcome any steam..

As this is a fan to be run at a determined airflow

Out of interest why are you worried about the speed ... surely measuring the flow is more important.

Mark

Thanks to all of you for the feedback.

I have a device called "Duct Blaster" http://products.energyconservatory.com/minneapolis-duct-blaster-system/

It is used to measure pressure/airflow within a system. (They sell for determination of leakage in housing ventilation systems)

Their meters are set to standard ambient conditions 20ºC, 1013.25hPa

As I use the Duct blaster to measure the air movement of a hot airflow, the density will change. Hence I need to work out a corrective factor for the indicated airflow.

It is quite easy for the temperature and the barometric pressure, but the steam (vapor and condensate) will effect the density as well.

By knowing the rotation speed of the blades, I can calculate the real airflow without any corrective factors.

(N₁/N₂) = (D₁/D₂)^-3 (Q₁/Q₂)

D = Fan size

N = Rotational speed

P = Pressure

Q = Volume flow rate 

Sounds like something that the TI LAUNCHXL-F28027F LaunchPad with the BOOSTXL-DRV8301 Motor Drive BoosterPack may be able to solve. TI seems to say this setup can be used to measure the position of a motor without the need of any external sensors. Would be intresting to see if this really does work, good luck.

Kas

Have you considered using a small microphone and using the sound to determine fan speed?

You should get a measurable difference when the fan blade passes by the mic, which would allow you to directly measure the fan speed.

DAB

Hi Kas,

Yes the TI board has the ability to self measure the stator position using the EMF on the coils.

TI uses this information to control the current into the wires to rapidly spin the motor and control the speed.

I think this technology only works with the new Brushless motors though.

DAB

I was going to mention hall effect sensors. The reason I went with reed switches was because they operate over a wider temperature range, but I suppose for steam (~100 degrees C) hall effect will be fine.

A previous project I was involved in needed to measure rotational speeds in environments of 300+ degrees C. We used multiple reed switches for redundancy/reliability.