Air Flow Measurement

[Stephen Dubovsky]

  alright, someone who knows alot about this stuff, so here are some questions!

>
...
  Some of us dont have access to a dyno (hence the car window idea) ;)
although I might be able to schedule some time in the wind-tunnel...
...
>The problem is that steady 
>flow conditions only exist at very low throttle openings. 
>
>As you open the throttle the steady flow starts to become pulsating 
>flow and if you're unlucky it will turn into reverse flow at some point.  
>To give you an idea of the magnitude of this, at WOT, 2250 rpm on the 
>engine I'm working on (4-cyl) flows about 37g/s average, but between 25 
>and 54 g/s peak-to-peak.  The frequency is twice engine speed for 
>obvious reasons

Ok, why the heck does the airflow change directions?  Some sort of shock
wave like in a 2cycle tuned pipe exhaust?  Wouldn't this just indicate lousy
intake manifold/runner (whatever you all call them) design?  Seems it would
give peaks and valleys to the torque curve (like resonances in the exhaust do).

>The problems become apparent when you know that the sensor output is
>non-linear, the frequency response is asymmetrical, and any reverse
>flow signal will be rectified.  This means that the true average 
>(voltage) signal will not represent the true average airflow.  Any 
>digital approximation to this will only be worse unless you use a few
>tricks.

Understand that reverse flow would be rectified, no problems there.  Yes,
the sensor is non linear, watts vs m/s is ideally square or cubic in nature
(can't remember w/o my books in front of me.)  What do you mean by the freq
response being asym?  So the sensor has some frequency response, all sensors
do.  I would expect a low pass filter (LPF) that MIGHT be resonant somewhere
(although I cant think of a single reason why this would happen).  This
could still be dealt w/ easily.  I would expect the freq response of a MAF
to be really poor compared to computational ability.  The fastest
thermocouple I can find in the Omega catalog is 0.002" thick (or was that
0.001"?) which means it has VERY little thermal mass and still has a several
ms (1-10) response time (slow in my book).  I havent seen a Pt element any
smaller.  This gives a response bandwidth for 100Hz to 1kHz @ -3db which
brings me to my next question...

>To convert this analogue signal to a useful byte or two, you must 
>perform some A-D conversion.  So you are quantising it not only in 
>the range, but also in the (time) domain.  If you don't do this syn-
>chronously (ie you sample in the time domain) you will wind up with 
>a very noisy (and aliased) signal.  If you do it synchronously (in the
>crank domain) you have to decide what angle gives you the represent-
>ation of the average airflow.  Your sample angle cannot be calculated
>(easily) and varies significantly with engine speed (and to a lesser
>extent with throttle position)

  Ok, if I sample async and the sensor looks like a LPF w/ a corner at 1kHz
it will roll off at -20dB/decade.  If it is resonant somewhere then put an
electrical filter between it and the ADC (a good idea anyway) to make it
look like a LPF.  I've got 10bit A/Ds which mean I get about 1/1024 = 60dB
dynamic range (which is a pretty good approx).  The signal would be at -60dB
in 3 decades, 1KHz -> 1MHz.  Nyquist then says I sample at twice that, 2
MHz, and the aliasd signal gives less than 1 bit error across the entire
1MHz bandwidth.  This would be the WORST case, as you could use a multi pole
filter to attenuate the signal faster and sample much slower (or digitally
decimate to avoid the problems that the phase shift from high order analog
filters would cause).  Anyway, sampling wouldn't be a problem (and I could
always use the signal in the analog domain).


>
>I guess what I'm trying to suggest is that (from my experience at
>least) the MAF is no slap-it-on-and-go solution.  A MAP sensor on
>the other hand is linear, can be throttled to give a physical average
>or can be sampled and averaged/filtered in the time domain without
>difficulty.  It has it's disadvantages, but...
>
>There's heaps more to be said about this subject but I hope it's
>useful to share this with you guys.  If anyone has any other ideas or
>experience on getting MAF's to work well, I'd love to hear from them.
>
>Andrew Rabbitt
>Orbital Engine Company
>Perth, Western Australia

  Ok, if the flow reverses direction, can't you just move the sensor
'upstream' (even past the air filter)?  I know it would delay the response
time, but if you looking for accurate measurements it seems that a MAF can't
be beat.  Sure seems a lot better than using MAP and some magical 'volmetric
efficiency'.  Could always use the TPS or MAP to help response time (as
discussed in some other thread to get TPS to help the MAP).

  Another question/problem.  How affected are MAFs by the relative humidity.
Since it works on an energy loss principle, the reynolds number and other
stuff are important, but so is the specific heat of the fluid being
measured.  RH changes the sp. heat of air.  When the humidity goes up, so
does the sp.heat, and the energy lost from the sensor, giving a false high
reading of the air mass.  It actually measured the air mass + some
desired/undesired H2O(g).  Curious to know if this is a big problem.

  Plz dont take any of this the wrong way, Im just looking for a couple of
answers and you seemed to know too much practical stuff about MAFs;)
SMD

--
Stephen Dubovsky
dubovsky at vt.edu

95 Yamaha FZR600
83 Porsche 911SC
84 Jeep Cherokee