More accurate method

[chicosaurus]

Hi

What's a more accurate method of flow measurement:-

1. Orifice Plate

or

2. Pitot tube

[84-1074663779]

Neither, both have advantages and disadvantages, but the problems involved are quite different.

Each has its proponents. The accuracy you finally get will probably be proportional to the care and effort you put into it.

[98-1074649673]

Build for repeatability, accuracy will come after that. U want a bench that repeats from test to test and day to day.

You will find some like pitot style and some like orifice, personally I like pitot, but thats what I started out to build before I knew anything about flowbenching. My mind is to ingrained on the pitot style to change now . . .

[chicosaurus]

[86rocco]

Perhaps I'm mistaken because I don't have a complete understanding of exactly how a pitot type flow bench is constructed but it would seem to me that the orifice type, by virtue of the fact that you can switch orifice sizes, would be useful over a wider range of flow values than a pitot tube type. Or am I missing something?

[84-1074663779]

Hehehe, its like politics and religion, there are pitot bench people and orifice bench people. Each can see definite advantages in doing it their own preferred way.

Personally I have only been interested in orifice style benches, but pitot benches have some real advantages. The biggest is probably the large reduction in required blower power.

[98-1074649673]

The key to pitot style of benchs I have found is the velocity in the testing tube (duct dia) I had to decrease the dia of my testing tube to get my readings steady. Why spend big bucks on a Dwyer Pitot when I can machine up annubars for any size duct dia u need? I can make them for any standard PVC pipe, 3", 2" 1.5" checkout my parts page.

The test tube dia can be changed based on what u are testing. I have one in 3" dia, 2" dia and 1.5". I use the 1.5" dia one currently. I'm only flowing at 8" of water on the bench I now am using. If and when I get my new bench built I plan on flowing at a higher flow depression. I never really sat down and figured out what test tube flows what CFM's. Haven't had that much free time to do testing on my flowbench . . . just testing parts, which is what I built it for.

[Mouse]

My 2" flow element (pitot tube) reads well within 1% from about 40cfm to over 500cfm. I'm only pulling 2"H2O depression at 500cfm, but it reads right on the money using sharp edge orifices for testing (526cfm actually).

John

Edit: Opps, better clarify that. That is 500cfm calculated to 28"H2O depression. The actual cfm is 142 @ 2"H20.

The 40 cfm is also calculated to 28" H20, and is actually around 50 @ about 37"H20

John

[chicosaurus]

Been fiddling with calculations on flow & I think for my present needs (& budget) the Orifice Plate should do the trick. Going to use a 4 " pipe with various size orifices (using a 0-1.1 inchH20 Manometer). Want to do all measurements at 28 " H20.

Going to building the cabinet for the flowbench this weekend. Need to shop around for machining the orifice.

The reason I was considering the Dwyer Pitot tube is because it can be easily obtained in South Africa. (Might up(down)grade the flowbench to a pitot tube at a later stage)

Thanks for the help & the healthy discussion

PS Any other tips/suggestions will be welcome.

[Mouse]

[84-1074663779]

The ideal orifice is mounted flush in the middle of a broad flat surface, with still air (room air) upstream, and has a reasonably high depression across the orifice.

This is pretty easy to do when you place a sharp edged test orifice flat on top of your bench. It is a lot harder to do inside the bench with the actual measurement orifice. The problem is that the measurement orifice always sees disturbed air, and this effects the flow coefficient.

Running a higher delta p across the measurement orifice makes it less susceptible to upstream turbulence, but it can greatly increase the required blower power.

An excellent test is to construct two identical orifices, one fitted to the test hole, and the other as the measurement orifice. The pressure drops should be identical. How much they differ will give you an idea of how far you are from the ideal. Then swap them over just to be absolutely sure.

If you can get your largest orifice to flow within one percent of ideal, all the smaller sizes will fall into line. You can then just make your smaller sizes according to diameter, and the flows will agree in step. The largest measurement orifice is always the one that causes the most grief, and is the place to start.

[chicosaurus]

[84-1074663779]

The pipe mounted orifice flowmeters used in industry are usually used for water or steam, that type of thing. They are not precision instruments, but usually just a low cost basic flow indication for process monitoring.

In the gas industry laboratory where I used to work, where precision gas flow measurement is required for revenue purposes, positive displacement flow meters are always used. These either use a bellows that fills and empties, or a partly submerged turbine where inverted buckets on a paddle wheel fill and empty as the thing slowly rotates. These are highly accurate and have low pressure drop and work right down to zero flow.

For laboratory (high) air flow references we used sonic nozzles. What you do is have an accurate calibrated nozzle, and increase the depression across it until the flow goes sonic. Any further increase in pressure differential across the nozzle gives no further increase in flow beyond that point.

You connect up the sonic nozzle to a monster vacuum pump and open the valve. You hear the air rushing into the hole with a mighty roar. As you open the valve further the roar gets louder, then suddenly it goes dead silent ! ! As the air rushing in gets to the speed of sound, no sound can escape back out. Fascinating stuff.

You then know exactly how much air volume is rushing down that hole, and you can put your test piece in front of it and know the flow will not change from the exact sonic nozzle calibration figure as long as it remains sonic.

I could tell you 101 ways to measure flow, but putting an orifice in a pipe is probably the worst idea possible. It is only done because it is cheap and convenient in industry. It hs NO virtues apart from that.

You can do it, and it will work, but you will find that you end up with a series of orifices that vary in size in a totally unpredictable way. Make a one inch orifice and a two inch orifice and put them in series in your pipe. In theory the larger hole should have four times the area and one quarter the pressure drop. The ratio you measure will be nothing like four.

Swap the locations around in the pipe, the ratio will still not be four but will vary to something else quite different.

Now try to figure out how you are going to calibrate it.

The best way to calibrate a flowmeter or orifice, the way it would be done in a proper laboratory is with a gas bell. This is a large chamber like a gasometer suspended over a deep tank of water. The chamber has a very accurate known internal volume. It can rise or sink into the water as you either blow air into it, or let air escape. By weighting the bell you can accurately control the internal air pressure.

Now by measuring how far it rises or falls, and the time taken you can very accurately measure the airflow through anything, at a fixed known pressure drop. We had one about ten feet in diameter and ten feet high.

Certified orifice plates are also used, but great care is required to ensure nothing upsets the upstream airflow. For this reason they are usually only used as secondary flow standards.

[Mouse]

Tony,

I was wondering about calibrating with sharp edge orifices. When I place an SEO on my fixture and pull air through it I get a flow reading of 161cfm. Now if I lay a short section of 4" pipe on the SEO the flow increases to 166cfm. So if I flip the SEO and blow through it, the fixture will now represent that 4" pipe ring. How do I handle this situation. Should I calibrate to 161 or 166 when blowing out the 4" fixture?

I noticed that the Efunda calculator has an input for upstream pipe size, but doesn't seem to be used in the calculation (it doesn't matter what size I enter, the result is always the same).
Thanks

John

[84-1074663779]

Interesting question John.

The flow coefficient seems to be a magical number steeped in mystery. It probably lies between 0.6 and 0.7, but that is one heck of a difference.

A very great deal seems to depend upon what the air upstream of the orifice is actually doing, and how much energy it takes to push/pull a certain volume through the hole.

Does the air have to accelerate or decelerate ? is it turbulent or laminar ? and so on.

My own preference is to use still air and a large flat surface with a sharp edged orifice. At least it is a condition that can be readily duplicated, and it should be repeatable.

As soon as you put a pipe upstream it is going to direct flow towards the orifice, and maybe increase the flow. Or it might create turbulence and decrease the flow.

If the pipe was very nearly the same size as the orifice, and it had a flared entry, it might increase the flow a lot. Place a sharp bend in the pipe just before the orifice and it would probably upset things rather badly.

It is all too difficult. My head hurts..........