bluff body sensor?

Anything that does not fit into pitot or conventional orifice flowbench design
Post Reply
Dan_DiM
Posts: 27
Joined: Sat Mar 03, 2012 9:45 pm

bluff body sensor?

Post by Dan_DiM »

ive noticed that all the flowbench discussion on measurments is some sort of pressure differential setup, either a pitot, venturi, or calibrated orifice, and ive wondered about elbow meters for bidirectional flow but hear they are not repeatable. the problem i see is that there is a range that these setups work within, while bruces design works fantastic with a single orifice plate that will work for 95% of the jobs out there i was wondering if anyone has explored a bluff body type sensor? from what i understand the output is 100% linear and the conversion and calibration is based on only the width of the bluff body. the output is the frequency of vortices that alternate off the body and can be measured with air movement on a diaphragm through the use of differential pressure sensors(not the differential but how often the pressure reverses), or with a piezo or other electromechanical sensor in or behind the body. these sensors apparently have a huge range and the linearity seems like it would be an advantage as far as calibration and repeatability go.

the issue i think you might see is noise in the cabinet making it's way into the signal that would need to be filtered out. these meters are used in industrial applications and are readily available. i also believe, though it's uncommon, it might be used in a few cars as an airflow meter(similar to a MAF but without the mass part, the density would be figured out from temp and map).
Tony
Posts: 1438
Joined: Wed Mar 03, 2010 3:40 pm
Location: Melbourne, Australia

Re: bluff body sensor?

Post by Tony »

Vortex shedding flow meters are fairly common in industry for measuring liquid flow.
They are especially critical of up stream turbulence and require a very long run of straight smooth pipe up stream of the flow meter, typically twenty to fifty pipe diameters for liquids are recommended.

Vortex airflow meters have also sometimes been used for measuring airflow for vehicle electronic fuel injection systems, but they too are very critical of upstream flow conditions, and any change to upstream pipework or air box can throw them way off initial calibration.

When taking measurements on an airflow bench, swirl, tumble, and all manner of flow instabilities can be produced by what is being measured.

Maybe a large settling chamber, a flow straightener, and then fifteen feet of dead straight four inch plastic pipe ahead of the vortex meter, and five feet of straight pipe after might work. It is going to be a lot more critical than you might think to get really good results with air. Much easier with more dense liquids. Resonances and pipe vibration must also be avoided.

Certainly possible, but I would expect the sheer size of the whole thing would make it rather impractical for many of us.
I suggest you Google "installing vortex flow meters" to get more of a handle on how these work, and how very critical these are to install.
Also known as the infamous "Warpspeed" on some other Forums.
1960FL
Posts: 1338
Joined: Fri Jan 08, 2010 10:36 pm
Location: Maryland

Re: bluff body sensor?

Post by 1960FL »

Dan,

It sounds like a viable option for flow but the insertable models 8" pipe and up seam to not have any better take on accuracy at 2% Full Scale, the PTS on a single orifice does better than this.

The ideal is a venturi (nozzle) array but at $700 to $1200 a pop that can be expensive and they are only rated to 1% FS unless you purchase aerospace quality at about $4000 for a 2" nozzle.

The orifice can produce these results the key is well designed chambers and balancing the flow across the orifice on intake and exhaust ( air flow sees the same thing in either direction) if you do not want to use different CD'S based on flow direction.


For thos interested here is what Dan is an example of and industrail model but they do come in many shapes and sizes.

http://ovalthailand.com/vortex3101.html


Rick
Post Reply