Fabricating velocity probes

[jsa]

Thanks for testing and posting the results John, I was wondering how the probes were going.

[slracer]

John, I would like to see a pic of the setup to be sure I understand what you did, but from what I understand your of your test procedure, I would concur with your analysis of what is happening. I also thank you for all the effort you put into this testing! Without your support, the results could have been suspect!

I was really hoping it would turn out better as I have very small ports and need to get a probe into them and be able to position it accurately. My exhaust valve is approximately .78" dia and the port before the valve is even smaller so pretty tight. Intakes are closer to .96" dia.

Back to the drawing board! -- Doug

[jfholm]

I am going to test you're smaller probe tonight and see if it gives different results. It is tighter and closer together so it may not be as bad - also if you always have the main tube pointing to the center of the port you may at least get some repeatability. I will try to give some better information tonight.
John

[jfholm]

Well I noticed that my original clamping device was not holding the velocity probes and letting them twist and move. I got some clamps that clamped tighter. Sorry about crushing your probes Doug. (just kidding) I'll try to get some pictures tomorrow.

All tests done at 10" H2O.
Better results but still off the same way a little bit, but the smaller tight tube is closer and may be usable. You be the judges:

Small tight velocity probe: (I was more careful this time)
against pushrod wall about .015" off the wall, 3/4" in from intake surface and half way between floor and roof.
#1-with main tube up and pressure tip right below it = 175.2 fps
#2-with main tube horizontal pressure tip same spot = 180.6 fps
#3-with main tube down below pressure tip in same = 173.3 fps

same probe but moved to center of port 3/4" in:
#1-with main tube up pressure tip down = 180.1 fps
#2-with main tube horizontal to right tip in center = 178.1 fps
#3-with main tube down under pressure tip in center= 172.7 fps

I retested Doug's first probe that has a larger radius but is still what I would call small, I will just put fps as it was identical to tests above:

Against pushrod wall:
#1 = 182.2 fps
#2 = 179.2 fps
#3 = 177.0 fps

Center of port:
#1 = 185.6 fps
#2 = 173.6 fps
#3 = 172.7 fps

[jfholm]

Well guys I just ran tests on my conventional probe again. It has a larger radius that Bruce's and I need to make another one.

Things I noticed at 10" H2O with Doug's probes my port flowed 144 cfm with my probe in place 141 cfm. So my larger probe hurt flow 3 cfm. The velocities were slight higher. Was this caused by my probe being slightly larger?

Once again the same way as I tested Doug's. I only tested against the pushrod wall.

#1 - 206.9 fps
#2 - 215.2 fps
#3 - 215.3 fps

Edit ps: I think I need to make a new probe for myself also. I made mine a long time ago. I noticed that the static ports are .600" from the pressure tip and is 1/8" tubing. I think the static ports should be closer to the pressure tip. In the back of my mind I think they should be around 300% of the tube diameter which would be .375" instead of .600". Any input? This may be why mine is reading higher velocity.

[jfholm]

I would really like to come to a resolution on this as if Doug's is the correct fps reading at 28" H2O that would be 300 fps and with mine it would be 359 fps. That is quite a difference.

John

[jsa]

[jfholm]

I have a tube I built to calibrate pitot tubes. It is a known diameter so if it is flowing x cfm of air at a certain test pressure we should know the fps mathmatically. The probe should match that fps then, correct?

John

[bruce]

Correct, I eluded to this a few pages back

[jfholm]

Tonight I tested all three velocity probes in my "tube". It has an inside diameter of 1.600" and flows 144.3 cfm at 10" H2O. This works out to a Cd of .815.

Doug's velocity probes do not seem to affect the cfm much as the are so small. With Doug's the tube flowed 144.2 cfm. When I flowed with mine, the tube dropped to 141.9 cfm. I used the orifice calculator and subtracted the area of my velocity probe from the area of the tube and that is indeed what it flows. So my velocity probe is large enough to affect air flow.

Now the velocities. Using Ed's pitot calibration spreadsheet it said that a diameter of 1.6 at 10" H2O should be 209.2 fps.

w/my probe 3" in the tube in the center = 199.3 fps/9.03" H2O
w/Doug's larger probe same conditions = 203.3 fps/9.42" H2O
w/Doug's smaller probe same conditions = 203.3 fps/9.42" H2O

now that was 209.2 without correcting for temp and barometric pressure in the spreadsheet. When I set Ed's spreadsheet to calculate taking in consideration for the lower barometric pressure and higher temp then the velocity went up to 228 fps.

When I put the 9.42" H2O in Larry Cavanaugh velocity calculator it corrected right to 203 - it does not go out decimal places.

From Larry's velocity calculator, which at one time I checked against my Port Flow Analyzer and they were in agreement, I would say Doug's probe's were right on in the center of the tube.

[stef-1]

I had a chance to try out mine. Although I couldn't use the same setup as John I did observe the same trend. In hindsight it's not suprising considering that a duct will have a velocity profile - that's why you have averaging velocity probes! Still, it was an interesting idea to try out. Making a new probe should be easier now I have some "practice" at it :p

[jfholm]

I finally found the pdf file again on the design of the original pitot tube and in that article is where it states that the static holes should be 3 tube diameters behind the pressure tip.


btw I am constructing a new tube to test with meeting this criteria.
John

[slracer]

John, Is that 3D the inside D or outside D? I have made "cases" for both, so decided to ask! -- Doug

[jfholm]

Good point Doug. I don't really know. Since the brass tubing is so thin I just am going to use the o.d. My first probe had the static holes .600" from the pressure tip. I am excited to make another and since I am using 1/8" brass tubing the static ports will be 3/8" from the tip. I hope this makes a difference in my readings.

The one thing I like about yours Doug is the static ports are the same location as the pressure tip.

John

[slracer]

John, I've been thinking (it doesn't hurt too much) and was wondering if the static port(s) on the outside of the tube (away from the crook side) could be taped over and retested. That would position the (average) static pickup and the dynamic closer together which "should" reduce the variance in pressures. If that does help, I'll try to put together an even smaller crook and see if gets an acceptable variation. I still believe that this type of design could be more accurate as the staticc pickup is not in disturbed flow and the flow CSA is the same at both the static and dynamic ports (especially in small ports). It my still be that it can't be done, but the first results look encouraging to me! If you would prefer to stop testing, let me know as you have gone far further than was expected! Thanks again! -- Doug

PS - Can you post the link to the discussion about static port location, I remember some tables that had "K" factors for variation in location that I would like to look at again?