Air flow measurement sensor

Thread Starter

embpic

Joined May 29, 2013
189
I need any air flow measurement sensor/circuit so i could measure the amount air is flow.
I searched on google but i got air flow detector using opamp but i need measurement of air flow.
 

KMoffett

Joined Dec 19, 2007
2,918
I need any air flow measurement sensor/circuit so i could measure the amount air is flow.
I searched on google but i got air flow detector using opamp but i need measurement of air flow.
What velocity of air flow are you wanting to measure? HVAC? Wind? Breath?

Ken
 

cork_ie

Joined Oct 8, 2011
428
Mass Airflow Meter from a car ? Sorry, im a mechanic by trade.
Absolutely the best way to measure airflow in an engine is to use a second MAF (Mass airflow meter) as a MAF will give you a far more accurate and linear output than any alternative. Automotive MAFs have specially circuitary built in to compensate for changes in density and are designed to be linear for the kind of air flow ranges in an engine.There are two main types
1) linear 0-5V output for stated range.
2) Variable frequency types - output signal frequency is depenant on airflow. If you look at a workshop manual for the type you intend to use it will generall give you figures for the signal voltage or frequency V's airflow in g/s or Kg/hr
 

wayneh

Joined Sep 9, 2010
17,496
The biggest problem you have is that the flow rate is not constant across the area the flow is passing through (a duct?). It has a velocity profile that depends on position. This is not as bad a problem if you only need to watch relative changes, but it's a huge problem if you need an actual volume estimation, cfm for instance.

What are you trying to do?
 

sirch2

Joined Jan 21, 2013
1,037
Bertus, my thoughts exactly. I've done quite a lot of work on measuring low velocity airflows over the last 6 months and was planning on posting some details but I usually look through to see if the OP has bothered to reply to anyone before I put a lot of effort into a post.

So I'm happy to post what I have if the OP or anyone else expresses an interest.
 

sirch2

Joined Jan 21, 2013
1,037
OK, I tried one of those and wasn't that impressed so I rolled my own. I'll get some details together and post something later.
 

wayneh

Joined Sep 9, 2010
17,496
I'll say it again: Measuring a point velocity alone is almost useless in measuring flow (the word chosen by the OP). They are not the same thing. Could be a language problem here, it's hard to tell.
 

sirch2

Joined Jan 21, 2013
1,037
wayneh - I don't think anyone is disagreeing with you, but sampling across the airflow and averaging is one approach.

For pilko I have attached a schematic of a thermistor based anemometer. It uses self heating of a thermistor in one leg of a bridge (TR2). TR1 is there for ambient temperature compensation which works OK for small variations but it is hard to get thermistors with the same coefficient of resistance for 10k and 220R so the bridge does go out of balance if the temp changes too much. Half of U1 and T2 form a constant current supply for the bridge. U2 is an instrumentation amp to amplify the bridge output. U3 is just there to allow a MCU to switch the unit on and off and could be left out. The other half of U1 drives an LED to indicate bridge balance and if airflow is detected. Basically I adjust the bridge via R5 until the LED lights and then back it off until it just goes off. This means the output of U2 is at around 0.6V, the voltage being set by the junction of D1, I found this to be good for my application which was detecting very slight drafts.

Was going to attach the Eagle schematic and board files but they don't seem to be allowed formats
 

Attachments

wayneh

Joined Sep 9, 2010
17,496
I'm just pointing out that we're blind without more info from the OP about what he needs. The best solution might be a streamer hanging from the grill of his fan, a wind sock. Or he may need a temperature and pressure compensated sensor array and a microprocessor.
 

sirch2

Joined Jan 21, 2013
1,037
I found the Modern devices wind sensor quite noisy when detecting low velocity flows so it was hard separate draft from noise and I wanted to input the data to an MCU so wanted 0-5V at the output in the range of flows (around 0.3m/s) that I was interested in.
 
Top