Airflow measurement - again...

[Darrell Norquay]

At 02:44 PM 1/10/97 -0800, Tom Cloud wrote:

>> How do MAF meters work?  I understand they typically use heated wires??
>> Why not use thermistors (slower response?? -- but larger R change and
>> easier to instrument .. ??).  I suspect that they have long ago been
>> tried and discarded -- true??
>> 
>> Tom Cloud <cloud at peaches.ph.utexas.edu>

And Mazda replied:

>I don't have the response curve to prove it, but thermistors are 
>generally much too slow for such an application.  Also, a thermistor 
>changes resistance with temperature, so you cannot use it by itself for 
>flow measurement.  You would have to have a mass and a heater to heat it, 
>so you can measure a temperature.  After all that you end up doing what 
>the hot wire does, except much slower and at higher cost.

Thermistors don't need to be slow.  It depends a lot on the size and thermal
mass of the thermistor body.  Fenwal and YSI sell lots of units that are
only 1mm in diam, very fast response.  In any case, a hot wire anemometer is
just a more rugged type of thermistor, which uses more current and has less
sensitivity.  Also, a little slower response may give better averaging of
transients and a smoother signal...

Thermistor mass flowmeters are very common in the instrumentation world, and
can be very accurate indeed.  Usual method is to have two identical
thermistors in a "bridge" configuration, one in the air stream and one out.
This compensates for any ambient temp effects.  The thermistor in the
airstream is purposely heated to raise it's temp, and the drive current is
"servoed" to maintain a constant voltage across the sensor.  As airflow
increases, it takes more current to keep this voltage (and therm temp, BTW)
constant, and the change in drive current is simply measured to determine
airflow.  Exactly the way a MAF works, but a MAF is a mass produced, limited
accuracy unit made to be cheap and not particularly accurate.  (BTW, there
was some discussion I saw somewhere on how to make a hot-wire anemometer
from a couple of #52 panel lamps, one with the glass envelope broken off...)


regards
dn
dnorquay at iul-ccs.com