Tractorsport Flowbench Forum Archive

Posted: **Fri Apr 29, 2005 5:36 am**

I was having a night of no sleep and my mind has been hashing over alot of stuff. I keep thinking about the ora plate and bench mods. Why couldn't you just make a plate with a given hole size and mount it right below the cyl head fixture and attach your manometer to the adapter plate.when you wanted to change what your flowing unmount the fixture change the ora plate to another size and re attach the fixture
??? another 64$ ?
If we test a head are we really getting accurate readings because each engine had a diff bore size and stroke. So IMO the hight of the fixture should match the length of the piston travel and the inner diameter should match the engines bore size. To prevent a false reading due to cylinder wall schrouding. If the fixture is bigger then the actual cylinder then the reading would be off because there would be more surface area around the valve when it is open as there would be less area if the fixture was smaller.
On my quest for insanity I understand that the ora plate needs to be a certain thickness to get an accurate flow through a given hole but does any of the formulas add for deflection of the plate.I know the figure for the edge of the plate. Because if the plate does flutter or deflect that will change its flow charastics. in my lack of sleep mind then if a plate was made thicker say 1/2" (I understand it might not flow as much) shouldn't the math formula be able to correct for the thickness and calculate a lower flow number? Then all you'd need to do was enlarge the hole to a given size to return back to the given cfm you were wanting.
sorry if im not making much sence being awake for 24hrs does that to me

PS found an interesting link

Posted: **Fri Apr 29, 2005 7:29 am**

Are you talking a true sharp edge orifice or a plate with a sharp leading edge and trailing edge? I think that alot of the differences everyone is seeing is due to the different contruction techniques being used to make an orifice. A true gas/air orifice has one sharp leading edge, this is made by a 45* angle intersecting the face plane with the dia being held to .001". This is a gas/air industry standard (sure they use .0001 tolerences). And there are standards for the construction of a measuring orifice. I have not been able to get a copy of these standards as of yet. I did have a quite lenghty discussion with a manufacturer of orifices for the gas industry a few months back which was quite enlightening on their construction.

This sharp edge needs to be "dead" sharp or your readings are going to be off. It should be sharp enough that it will shave the surface of your fingernail off.

Posted: **Fri Apr 29, 2005 9:13 am**

Posted: **Fri Apr 29, 2005 11:39 am**

My idea, if I were to build an orifice bench would be to place my orifices on a strip of sheet metal that could be pulled across an aperature, much like film in a camera. I have never liked the disk idea too much. To air seal the orifice, a gasket could be made of rubber tubing that could be inflated with air pressure to form the seal on both sides of the orifice.

I have found that .01" sheet metal will make an acceptable sharp edge orifice if the hole is cut cleanly. The hard part is to do so without distorting the metal. I have also cut orifices in .02" metal and then beveled the trailing edge with a fine rat-tail file and flow test them until they flowed correctly. Not too difficult to do.

Posted: **Fri Apr 29, 2005 5:55 pm**

still havent really slept yet LOL
Mouse I was thinking of the same idea with using a slider plate instead of a disk. Then all youd need is a weak spring and an old axl bearing to make a detent to hold the plate in the same position everytime you slid it to another position. Hmm now im thinking of a worm gear that is attached to the plate so you could crank it left or right like used on a mill table.
ok now the wheels are turning who has an extra x asis mill motor lol

Posted: **Fri Apr 29, 2005 7:49 pm**

This is along the same lines of what I want to do for my orifice plates in the bench. What I have thought of is to use a carrier and a plate. Kind of like a drawer sliding into a cabinet on its guides, the carrier and orifice plate would be in the horizontal plane.

The bottom of the carrier would be an aluminum plate, with a 6-inch hole, and about 2 inches further to the end and each of the sides. The distance from the center of the hole to the front of the cabinet would be made so that the center of the hole is centered over the flow chamber, and plate would extend outside of the cabinet far enough to have a handle across its width. The back of the handle that faces the bench would have a weatherstrip or similar seal on it, so that when fully inserted a seal would be maintained.

The top of the carrier would be made of 1/2-inch aluminum with a rabbet machined into it. The dimension with the full 1/2-inch thickness would be drilled and tapped for the bottom plate to be fastened to. The portion that had been rabbeted away would make for the recess that the 1/8-inch orifice plate would slide into.

Sorry that's a lot to visualize.

But my thoughts are if the orifice plates are all machined with the same length and width, with the orifice centerlines the same, then to change ranges you would merely slide out one plate and insert another one.

I've seen pictures in industrial flow measurement handbooks that show a "Daniels orifice carrier" which allows an orifice plate to be changed without breaking the flanges. This would be similar in nature.

Best Regards,

Terry Terezakis

Posted: **Fri Apr 29, 2005 8:12 pm**

I think you guys might be worrying about the wrong things.

First up, the air pressure across the measurement orifice is only going to be a few inches of water, not hundreds of pounds per square inch. A thin metal plate is hardly going to bend or deform under such a small mechanical force, especially if it is supported properly from behind.

The plate needs to be thin with respect to the diameter of the hole. A four inch orifice could get away with much thicker material than a quarter inch orifice. Flaring the departure side is one way to achieve this with the smaller hole sizes, but why not just use thinner material ?

While the upstream edge is supposed to be sharp, I don't think it needs to be finished like a cutting tool to work. The edge should be just a normal machined (as in a lathe) edge, just broken with very fine emery so you don't cut yourself, or end up with a rough burred edge.

By far the biggest factor influencing the flow coefficient will be the quality of the air upstream of that hole.

Ideally the air should be at complete rest and un-turbulent, and be able flow towards the hole with increasing acceleration from all directions around the hole. That can only be achieved if the hole is mounted flush in a very large flat surface, with nothing upstream to disturb the flow or generate a turbulent wake as it moves towards the hole.

If you can do that, the flow will be entirely consistent.

As soon as you mount an orifice in turbulent fast moving air, or in a pipe there is going to be big trouble, and the results become unpredictable and unrepeatable.

Trying to mount a measurement orifice directly below a cylinder head is a futile waste of time. As the valve is opened further the flow pattern changes, and the readings will be all over the place. Move anything even a fraction out of line, and everything changes.

There absolutely MUST be a very large settling chamber before the measurement orifice, and the orifice must be out of direct line of the test setup.

Sorry guys, end of psychotic rant.

Posted: **Sat Apr 30, 2005 12:18 am**

The drawing is of a typical orifice bench setup with the flow disk residing on the bottom plate of the top plenum. The intake and control flow valves setting of to their respective sides [this is how mine are anyway when you're standing in front of the bench.

The red line represents the top of the bench with 1, 2 & 3 being discharge holes.

Now, I trust the formulas found in the forum for calculating potential CFM at a given pressure drop. I have no reason not to, they correlate nicely with published figures of Superflow units by altering the Ce number just a tad bit.

What the air is really doing is the question I'm posing and hopefully, the answers will help to resolve some of the questions about calibrating. In reality, it will probably raise more but......

Hole 2, should flow very directly into the orifice hole. It will probably be turbulent due to the direct blast of air on the orifice as many seem to suggest.

Hole 1 or 3 should help to reduce the turbulence and therefore indicate a calmer, more accurate pressure drop measurment.

But, in this design, [and i'm asking, not stating it as fact] what's really going on with the air? Nobody seems to be considering or at least mentioning the location of the intake or exhaust control valves as possible influences on the flow or the nature of the flow pattern which would seemingly influence the viewed measurments.

Using any of the 3 discharge holes, would it not stand to reason that the air flow is being favored off to one side of the hole because of the flow control valve locations? Hole 2 should be the least affected and 1 and especially 3, if testing intake flow, the most affected.

Consider hole 3. Wold the air not be bent sharply around the orifice and be favoring the right side of the hole? [intake flow]

Consider hole 1. Would the air not tend to also approach the orifice from an angle rather than flowing evenly around the circumferance of the hole?

In this design, I would think hole 3 would have the most contorted flow pattern.

Would the flow through the orifice be evently distributed? To what degree would it influence the actual flow?

Is it even remotely logical to presume that the air would exit neatly past the intake control valve and that all air following it would move down from hole 1 or 3 and slide over in an orderly fashion that would promote and evenly distributed pattern across the orifice hole itself?

The two brown lines that are perpendicular to the orifice disk line are theoretical baffels that may or may not straighten the flow a bit and help minimize the air pattern favoring any particular side of the hole.

My thoughts are that the entire flow path should be taken into consideration. Such details as the distance from the centerline of the discharge hole to the centerline of the orifice to the location of the flow control valve will influence flow and therefore the actual calibration of each builders bench.

I can easily buy into the concept of not having the discharge hole and orifice in a direct path with one another to avoid turbulent flow but I would think that the best configuration for the plenium beneath the orifice discharge side would be one in which the air is a bit more guided toward the vacum source. Then again...maybe I'm completely wrong. That's just the nature of this game. What you think the air should do, what your eye tells you it should do, often has little bearing as to what the air want's to do.

All things considered, I'm in total agreement with Tony's statement about a large settling chamber being a must have. If you don't, then I think everything I've tried to point out in this rather long post can haunt your efforts.

Posted: **Sat Apr 30, 2005 1:36 am**

Here is a quick diagram of my bench. I kept the orifice plate directly inline with the test cylinder hole. I used a couple of diverter discs above and below the orifice to block any high velocity directional flow at the orifice plate. The air must go around the disc and come at the orifice from all directions. This also forces the air to slow down before it can approach the orifice. Each disc is about 6-8" diameter and held by 4 all-threads from above or below. I think this eliminates the problems that would occur, especially at high flow rates, if the air were allowed to shoot directly into the measuring orifice.

The lower section is where the pressure control (bleed-off) valve is located and the suction and exhaust doors which open or close (air cylinder operated) to the back motor section of the bench.

One thing that is not shown is a second orifice plate on a slide on the same level as the plate, that adds ~250cfm to any other range on my plate, for ranges up to 600cfm.

Rick

Posted: **Sat Apr 30, 2005 2:32 am**

One thought I have previously had, but have not yet tried, is to have a large area fine mesh screen, or a plate with lots of small holes directly across the flow path inside the settling volume. A slight additional pressure drop is of no real concern at that point because we are measuring the differential pressure across the actual orifice plate.

The idea here is to break up any very large unstable swirls and eddies that may develop coming out of the test piece, and force the air through a sort of strainer with lots of small evenly spaced holes. A slight overall extra pressure drop would tend to even out the flow over the whole flow area of the strainer plate, and create lots of very much smaller swirls and eddies. These might all break up and dissipate in a much smaller distance beyond the strainer plate, and may produce something more resembling low velocity laminar flow going into one side of a large settling volume.

That may be more effective than just the circular baffle plate that Rick has suggested. I have never tried either idea, so don't know what the results might be. But I feel Rick's idea is probably a step in the right direction.

It is quite common for automotive hot wire airflow meters to use built in fine mesh screens to break up vortexes and eddies that can sometimes form upstream of the airflow meter. People that have removed those screens have sometimes encountered trouble, but not always. A sheet of honeycomb flow straightener might work even better, but a slight measurable pressure drop might not be a bad thing to spread out the flow more evenly over the entire cross sectional area.

You can buy commercial punched steel perforated plate with various standard sized holes. Typically the combined area of the holes are arranged to be half the total area, so it would not be terribly restrictive, and would be fairly easy to try.

Posted: **Sat Apr 30, 2005 9:13 am**

I agree with and like the screen idea. I too, have thought of doing the same thing. I will build it that way if/when I build another one. I have been using mine like the drawing for about 15years, and it seems to work very well. Very repeatable and orifice sizes that are comparable to a SF bench.

Orifice plate is .125" square edged for dual direction flow.

Rick

Posted: **Sat Apr 30, 2005 11:00 am**

Ricks bench is the way I would do it, if I were to build an orifice bench.

I like Terry's drawer idea for an orifice too. To expand on it, you could use just a thick metal/plastic plate with a bevel on one side to form the sharp edge, then you could simply flip it over for intake/exhaust....no need to have two Cds for one orifice. The thickness of the plate would allow you to have plenty of clearance around the orifice of supporting structure that would disrupt the orifice flow.

Maybe someday I will draw up my idea for a reversable air power module that fits onto the bottom of a bench like Ricks.

John

Posted: **Sat Apr 30, 2005 12:15 pm**

Thanks Mouse. I like your idea of flipping the orifice plate over for reversing the flow, if you're using a beveled orifice.

I've seen a lot of notes here and there about the bevel on the orifice plates. From all the reading I've done, the reason that a bevel is added, is to reduce the amount of non-recoverable pressure drop from the orifice. When you have high pressure drops across an orifice like in an idustrial application, then you need to increase the thickness of the orifice, in order to prevent the plate from flexing. But the thicker orifice plate results in larger pressure drops. Hence the bevel to reduce the amount of non-recoverable loss. For the low pressure drops we seen in these bench applications, I would think that a standard square-edge orifice that's 1/8" thick would be fine.

Has anyone done testing with a square edge orifice and one with a bevel for comparison?

Thanks Guys,

Terry Terezakis

Posted: **Sat Apr 30, 2005 7:14 pm**

I still prefer to keep the flow through the whole bench in one direction, and have two test holes in the bench top. One that sucks and one that blows. Not having to reverse anything simplifies the whole bench design in so many different ways.

I suppose it could be argued that blow testing does not really require smooth laminar air to enter what is being tested. Exhaust components work in highly turbulent flow anyway, and what comes out of the bench blower should be uniformly turbulent, as the pipework never changes.

For testing things like intercoolers, turbocharger pipework, and exhaust mufflers, high velocity turbulent blown air may be more realistic than trying to draw ambient air through the same part. Usually getting the air into the entry side of something like a muffler can be rather difficult, as fitting cones or flares will effect the result. Entry loss can sometimes make up a significant part of the total measured pressure drop.

In my bench I use one large diameter centrifugal blower with a 2.75" round exit plumbed to a 3.5" test hole through about three feet of 3.5" pipe.

The flow measurement orifice is located before the blower in calm air, and blow test pressure is measured with a pitot static tube just below the blow test hole.

I would value peoples opinions on the use of deliberately turbulent air for blow testing, as opposed to reversing the whole bench and trying to keep the exit air as calm as possible with a large settling chamber.

Just trying to kick around a few ideas, and get some opinions.

Posted: **Tue May 17, 2005 1:30 am**

Guys,

Could the "settle chamber" be as easy as adding another 2" tall chamber
on top of the MSD design bench with two large slots at the sides
of the bottom forcing the air to find the orifice disc from the sides?

Don,t know if that was clear, .. my drawing might be worse, ..

Isn't this how Superflow handles the problem?

Tractorsport Flowbench Forum Archive

more mods and i think ive gone insane