Wind Sensor values

Interfacing hardware, sensors, output drivers, chips.
Post Reply
blixem
Posts: 3
Joined: Mon Aug 02, 2010 7:20 pm

Wind Sensor values

Post by blixem » Mon Aug 02, 2010 7:30 pm

The output of the sensor is logarithmic, which means the sensor can capture very slight air movements at the low end, but also not saturate at full output until the air-flow reaches about 30 miles per hour. We’ll post some more data on the sensor soon.
I noticed on the Wind Sensor page at the bottom the air-flow reaching about 30mph would be the approximate top end. Do you have any graph or notation on the logarithmic output showing voltage -> airspeed relationship?

I have ordered a pair of these devices and am planning on testing the airflow inside of some equipment racks at different points. My first tests really just need to have a comparison between locations in the rack, showing where the airspeeds fluctuate. Future tests would need to describe the actual air speed differences in numbers I can relate to management.

paul
Site Admin
Posts: 735
Joined: Mon May 12, 2008 4:19 pm

Re: Wind Sensor values

Post by paul » Mon Aug 02, 2010 8:02 pm

blixem,

The wind sensor makes a great hobbyist sensor and is internally consistent. To get to some harder numbers with it, really requires me to buy some moderately expensive test equipment (hot wire anemometer) that I don't have access to right now.

I have also had a hard time getting the sensor to behave over temperature (it varies more than a little over 50 degrees or so). So calibrating it seems a bit like a moot question until I can iron out the temp variation, at least better than it is now. The gain (governing top end value) can be easily adjusted by swapping on a feedback resistor on the final op amp stage.

If you cared to buy a hot wire anemometer (I think you can find them for around $300) you might easily be able to calibrate my sensors to within 10% or so, and by doing the same experiment two or three times while swapping sensors, you could probably convince yourself (and maybe your management) about the wind flow. Having a real anemometer with someone's name on it a engineer would recognize, such as Fluke maybe (not quite sure they make them - and theirs might be $600), might be a real advantage in making your case.

If you buy the pro model first and drill some holes in your case you may not need to mess with my sesnors at all. I realize the size of my sensor might be some advantage. (no holes)

Sorry to shuck the hard work off on you. I'll get to revisiting this in the near term (with some other designs I think), but that's about all I can offer you right now.

You might also look at the (newish) Honeywell airflow sensors. You have to know the direction of airflow, and they're a little more pricey than my sensor (~$50) but then you get Honeywell's engineering and a real datasheet.

good luck and don't hesitate to write back if you think I can help,
I'll put getting the wind sensor schematic up on the site, on tomorrows todo list. I've been meaning to do this for a long time.

Paul

blixem
Posts: 3
Joined: Mon Aug 02, 2010 7:20 pm

Re: Wind Sensor values

Post by blixem » Mon Aug 02, 2010 8:26 pm

Thanks for the quick reply. Once I get the sensors, get everything hooked up, and my code running, I will see if I can find if anybody local with a "real" tool and do some calibration.

Funny thing is, this project is something I'm doing as a side project on the cheap to prove an airflow issue causing high temps, vs. managements initial (unscientific) assumption there is nothing we can do to fix the problem. We have multiple rack based units venting side to side in an enclosed rack, and I'm trying to show that once you put "x" amount of these in the rack there isn't enough air volume to handle the equipments internal fan. My two concepts were to either use a barometric pressure sensor on the intake/outtake to show increasing pressure as things were added, or to show a restriction of airflow (due to back pressure on the fans). I decided on the air flow measure due to the potential re-usability of the wind sensors in future projects.

Since management doesn't believe there is a problem we can easily fix, I have to front the project until I show results enough to convince them to buy professional grade testing equipment. Maybe worst case I can find a fan with a known CFM and calibrate from there...

paul
Site Admin
Posts: 735
Joined: Mon May 12, 2008 4:19 pm

Re: Wind Sensor values

Post by paul » Tue Aug 03, 2010 7:42 am

I actually like the barometric pressure idea though, since a rise in pressure would mean that resistance (and hence airflow) is increasing somewhere in the system. Most fans are characterized with a fan curve that shows how performance falls off with increasing pressure, so this might actually be a way to develop ideas about airflow without knowing what the actual airflow patterns are.

Paul

blixem
Posts: 3
Joined: Mon Aug 02, 2010 7:20 pm

Re: Wind Sensor values

Post by blixem » Tue Aug 03, 2010 10:52 am

Good point, thanks for the input Paul. I'll give you an update if I get anything.

mixio
Posts: 1
Joined: Mon Oct 18, 2010 1:21 pm

Re: Wind Sensor values

Post by mixio » Mon Oct 18, 2010 1:23 pm

Hello

I have calibrated the wind sensor and I'm reading its values.
Now, how can I map a value to an actual wind speed measured as km/h or m/h ?

Post Reply