Calibration Question

[dougk]

[color=#000000]A little history:
I recently built an averaging pitot tube bench (thanks Bruce) and did some flow development on Ducati motorcycle heads(100+tests). I use Megnehelic gauges because I happened to have a pair laying around. My 0-5" velocity gauge would fluctuate so I would pinch the static line to get a reading. I use a flush mounted static pickup.

When almost through I decided to see if I could calm down the fluctuation and I inserted a small brass tube inside the static pickup about 1/8-1/4" into the pipe. This immediately calmed the guage but I also picket up 9 cfm!! Dwyer shows this as one type of static pickup.http://www.dwyer-inst.com/htdocs/airvel ... uction.cfm
I called Dwyer for their input and he couldn't tell me which way would be correct. My feeling is that the inserted tube produces more syphon pressure (negative) on the low pressure input to the gauge causing a higher reading. He suggested a snubber which I purchased ,which worked,and I finished the tests.
BTW, my heads are flowing comparable numbers to what others get on the heads, so I basically believe my bench.

I made a spread sheet using the following formula with the addition of figuring out vapor pressure and adding that into the correction factor:
I'll try and show the formula I've used in the past:

Static pressure (ps) vertical manometer
Barometric pressure (pb) barometer gage
indicated temp (ti) thermometer

absolute temp (ta)=ti+460
Dry air density (d)=1.325*(pb/ta)
Correction factor (cf)=sqrt(.075/d)

Pressure velocity (vp) from your inclined meter
Velocity feet per minute (vc)=(4005*(sqrt[vp]))*cf....(I use 4000.4 per Dwyer)
CFM's = (inside dia of your testing pipe in square feet)*vc

I had an orfice plate with a 1" hole (chassis punch) in a .025" copper plate I decided to flow. It flowed 104.5 cfm corrected to 28"

I then calculated orfice flow using the following formula:
Formula:
: Q = C*(3.14d

[Mouse]

Doug,

A velocity profile in a round conduit has a D shape, the higher velocity existing the the very middle. The D shape also shows a lower velocity near the bounderies. As you move your static pressure measuring point toward the center of the conduit, the static pressure will increase, assuming the end of the static point tube is at a perfect 90 degree angle to flow. This changes the pressure relationship to your impact measuring point, which is fine, you just need to correct for that. You will need to flow test several calibration orifices and experiment with a correction factor until your range of orifices all jive, and assuming there are no other issues with your design.

Here a a section from the Flow Performance FP1 manual:

"Calibrating
The user can enter a calibration factor for the FP1 processor to use in calculating the volume of air that is being measured.
The exact figure depends a lot on how much space your pitot tube is occupying inside the conduit, the aerodynamic disturbance created by the pitot tube, and the location that the static pressure is being measured in relation to the impact pressure measurement location. If impact and static pressure are measured on the same plane perpendicular to the conduit, at the same distance from the nearest boundary, and the pitot tube is of modest size in relation to the conduit and has an aerodynamic shape, the calibration factor will be very close to the actual cross sectional area of the conduit in square feet."


Do yourself a favor and buy a calibration orifice from Bruce. Then there is no question about Cf and they flow right smack on the number. Anyone trying to build a flow bench should have at least one.

I would not recommend your calibration orifices be punched, as this may create a rounded edge and a peaked edge on either side. AOL CD tin lids and bottoms make good material for an orifice as they are only .010" thick and the rolled/folded edges help keep them flat. Make sure the edges of the orifice are perfectly square. Takes some time, but worth the effort.


John[I]

[dougk]

Mouse, are you on the Speedtalk forum also?

I'll buy an orfice plate from Bruce,should have bought one when I bought the averaging pitot tube.

I don't understand how the D shape would lay in the pipe for the pressure gradient. I guess you're saying the closer you get to the center of the duct the higher the pressure,but that would give me the opposite result from what I observed.

My biggest problem is the math. There is a 45% difference . My orfice can't be off by 45%.

Like I say, my numbers on the heads are comparable to others work . There is no way I'm reading 45% high.

Cheers,Doug

[Mouse]

Doug,

If a conduit were horozontal, the D shape applies.

You are right, I do have it sort of backwards. When I said the static pressure would increase, I didn't mention in a negative pressure relationship. So, the static pressure would decrease, meaning it is going more negative as the end of the tube is seeing higher velocity nearer the center of the tube.

I have posted on Speedtalk a couple times. Great site!

John

[Mouse]

Aha, I found it. I knew I had this somewhere in my pile of papers:

Velocity Profiles in a round conduit



John

[dougk]

Thanks John,

I'd pretty much figured out the "D" shape . I guess my port is turbulent flow from the pressure gradient I'm seeing, that and the way a string "flies".

I still can't reconcile the difference in calculated and measured flow though.

[Mouse]

How are you measuring your test pressure, a plenum or settling chamber?

What diameter pipe is you pitot tube in?

[dougk]

Plenum...very basic setup. I built a cube out of 1x 12" pine with the motorcycle cylinder and head bolted to the center of the top with the 4 studs. 1 1/2" PVC (nominal dia 1.595") comes off the center of one of the sides with the test pressure sampled from the upper corner of the same side. The cylinder protrudes about 1 1/2" into the box.

I may have had an air leak at the base of the cylinder with the test orfice, I'll retest.

I flow from 30" to 12" depending on lift.

[RRBD]

Doug,
For my static taps, I drill 5/32" almost thru and then make the actual pressure tap hole using a 1/16" drill then I insert my short piece of 5/32" brass tube and epoxy it into place. Try to make sure there is'nt a burr on the inside as well.

Scott