Orifice in tube

[maineSS]

I've been looking for design data that would apply to an "orifice in a tube" configuration. I've seen some comments about orifice dia vs tube dia factors, but I'm looking for something that quantifies the relationship. I see Audie uses this type of sensor with a flow straightener, and the idea is used in industrial flowmetering, so some sort of design proceedure must exist. I tried searching with "orifice in tube", but didn't find much- is there a thread I'm missing?

[Tony]

I'm looking for something that quantifies the relationship.

There isn't anything, because the up stream tube FORCES the orifice to flow in a very unnatural way.

You can always fit an orifice into a tube, and use the resulting pressure drop to measure flow.
It is done all the time, simply because it is easy and very convenient.
But there is no way to know for sure what pressure corresponds to what flow, unless it is first calibrated against a known flow reference standard.

The beauty of a flat sharp edged orifice plate mounted on a large flat surface, it that it flows in a very predictable way, as long as the up stream air is mostly undisturbed.
The air accelerates and approaches the hole radially from all around, and falls over the edge. Very little air rushes straight towards the hole from directly above.

As soon as you start blasting a high velocity jet of air straight towards the hole down a pipe from directly above, it totally screws up the flow coefficient. What you end up with is highly dependent on the relationship of all the parts in a very non linear and uncertain way. It is the sort of thing best tested and measured rather than predicted.

It can be made to work, if you are planning to mass produce identical parts for commercial use, and do some very expensive flow calibration. For home use it is just not practical or economic for a one off.

[jfholm]

I used to do an orifice in the tube. I found that as long as you keep the orifice smaller than 40% of the tube id it seems ok. The closer you get in size to the tube id the more it changes the CD of the orifice. Keep the tube as large as possible. I was eventually using 8" pvc tubing for mine. I found though it is much easier and accurate to just build a PTS bench like talked about here on this forum. I chased my tail for 3 years trying to go the other route and spent more time keeping the flow device calibrated than flowing heads.

John

[Tony]

Have to agree with John.

Usually orifice in a tube is used only for convenience, say four inch pipe with two inch orifice. That will not flow anything like a two inch flat orifice reference plate.

Once you start getting something like a twelve inch pipe with a one inch orifice, the results will become very predictable, because the pipe diameter is so large, the air velocity approaching the orifice will be very low, and it begins to act just like a flat orifice plate in a large volume settling chamber.

Your best bet would be to get a set of the Forum flow bench plans from Bruce, and construct a very well proven design.
It solves all of the problems, it will definitely work very well, and you only have to build it once.

[maineSS]

If the plans are like the schematic Gif posted in Flowbench 101, I have certain reservations about accuracy and overall flow efficiency. It seems to me the configuration will generate unecessary internal turbulence- especially directly upstream of the orifice. I'm not being nitpicky hypercritical here, as most other bench internal configurations seem to place internal airflow second to packaging, and do much worse from an internal turbulence standpoint. If I've got things all wrong, then by all means correct my assumptions!

The "settling box" shown in the Gif is going to have considerable internal turbulence- if you make one side of plexiglass, light it internally, then flow smoke thru, you'll see it's more like a mixing box. From here, the flow has to orient itself to the tube connecting the settling box to the orifice, in order to allow the flow to expand to pipe dia, a rather long tube will be required. Once the flow arrives at the orifice chamber, it will start becoming turbulent as it leaves the confines of the tube. Is this a good thing directly upstream of your orifice? It would seem better to get control of your airstream as early as possible, and keep it under control with minimum turbulence until the measurement process is complete. A nice side benefit to minimum turbulence is increased flow capacity, especially at higher depressions where blower efficiencies start dropping off rapidly. The Performance Trends design looks like a start in the right direction....

[Brucepts]

The Flowbench 101 is simply showing how a flowbench works not a design criteria . . . and now on to my thoughts on this;

I would say it's good to question things . . . but, given the fact that the PTS Design was developed out of the forum post discussions from 3 prior Flowbench Forums I have run, it's a pretty good idea we got something right I'd say

Not to mention the latest 33+ page thread comparing it to a well known commercially produced Big Blue flowbench on the market.

I know a lot of people don't want to spend the $$'s that I ask for my plans but in the end it's well worth the initial cost over the savings and headaches one goes through to design and build a flowbench. It also helps support the Flowbench Forum of which by the way I ask nobody to pay for.

By all means feel free to pursue an "orifice in a pipe" design and report back how it works, we always are open for discussion seems lots of people have design ideas but never come back with any results of what they found, happens a lot with the floating depression style flowbench designs also???

Personally, I feel I do not have to defend my design as it speaks for itself with all the discussion it generates on other forums from people who are more than satisfied with the accuracy and reputability of the design. My website did not get to the Top of the Search engines because I spend my money on Search Engine Optimization (SEO) as I spend ZERO dollars on SEO. It got there by forum members who freely discuss this on other forums/websites (thanks everyone)

Am I conceded?? Maybe I am . . . but I offer good parts at good prices so the average guy can build their dream of owning a top quality flowbench.

Keep in mind I have more than one or two PTS style flowbenches around the world working in shops.

Course none of my rambling on above answers your question sorry to say . . .

[1960FL]

MaineSS.

Yes you can do an orifice in a pipe and the information you are looking for is called beta factor it is a calculation that is applied to the CD that modifies the affective CD with respect to this ratio ( orifice to pipe diameter).

The one thing i will note is that at the current time the PTS DM does NOT support Beta factors though maybe it should?

Sounds like a good project and learning experience here is a link to Wikipedia if you follow the math you will see beta ratio
this is a good page http://en.wikipedia.org/wiki/Orifice_plate

In defending the PTS bench over the years i have studied a lot about flow measurement and orifice in a pipe is a good cheep solution but just as in a Pito Element it has limitations to flow and the affect of the Beta ratio causes it to have a sweet spot where the error in "low flow" and "high flow" tend to grow causing one to develop a correction factor at both ends. If you have access to Vizard's new book he tests Audie's product, gives it thumbs up but then goes into the above description on error.

Last of my ramblings but to maybe shed some light on this subject. Orifice in a pipe works it is an industry standard for many types of gas flow measurement!. That said it is not used in industry to measure large ranges of flow and varying pressures they are used in production environments where most factors are kept almost constant and only minimal adjustment to flow is required to meet demand.

In my thoughts you could build a nice bench in this design but it would require and array of 3 or 4 elements to keep flow in each range within an acceptable error and if you use a square edge orifice change in flow direction in not a big problem.

I hope this helps

Rick

[jfholm]

The Performance Trends design looks like a start in the right direction....

In my previous post where I talked about using different sizes of tubing was actually when I was helping Kevin at Performance Trends develope his system. That is where I got my Black Box that I am using on my PTS bench now. Kevin is a great guy and I can not say enough good about him and his software and electronics.

When I was doing the orifice in a tube we also found that it was beneficial to have at least 24" before and after your measuring orifice. We started with a 2" tube and quickly found it had short comings. I then built a 4" tube setup and then we found you need to go to an 8" tube to make it work best.

I have a small work area and found the packaging for this to be a little big for what I wanted to do. That is when I found this forum and bought the plans and built my first bench like the PTS. I threw it together in one weekend as I just did the chamber side and plumbed in two shop vacuums. I can pull 10" H2O through a Small Block Chevy with 270 cfm flow. With the electronics from Performance Trends it is a wonderful bench.

As to my history I also have a Flowquik from Audi. So I have done most of the flow devices and will build the PTS bench first over anything else. Sometimes when you have a flow bench you have to do some porting on heads and not just play with the bench making it work.

John

[maineSS]

One of the things which always surprises me is the effort people make to improve port flow is seldom seen in flowbench internal flow. It's perfectly OK to have multiple 90 deg bends, dramatic volume changes, sharp edge corners, etc- as long as it fits in a 2 x 4' floor footprint! It would be interesting to see how much fan Hp this philosophy uses up compared to applying a "conservation of momentum" design strategey.

Tony made a good point about a sharp edge orifice having a predictable CD if the upstream air was largely undisturbed. This would be the case if you were flowing it on the outside of the bench in a quiet room. But, how big does your orifice box have to be to absorb an incoming jet? Imagine you're standing in front of someone with an electric leafblower. He turns it on, you hold your palm in front of the jet and back up until you no longer feel it. How far would that be- 4ft, 6ft, 8ft? Whatever the distance, that's the minimum size of the orifice enclosure to assure an undisturbed flow with an incoming jet from an electric leafblower- and a multimotor bench at full power will likely be at least comparable.

I appreciate the responses, will be back later with more- Thanks!

[Brucepts]

Some discussion based on a pitot flow element but it carries over to the orifice in a pipe design in regards to "beta ratio".

http://www.flowbenchtech.com/forum/view ... beta+ratio

Beta ratio on a PTS Style Flowbench is: ~.211 to .071

The orifice plate does not "see" a stream/column of air . . . tis the beauty of the design

So do you go for "conservation of momentum" or flowbench accuracy repeatability? Assuming the PTS Flowbench is an Airhog? Actually, it's more efficient than the Big Blue Flowbench in the fact it will pull more CFM's at a higher depression using the same motors. I think this last statement might apply for comparison of any other flowbench design on the market?

I have not seen anyone do this study on any designs?