Flow question?

I'd have thought it was a pretty clever trick to inject at 45psi...into a line at 100 psi.

UNLESS you are relying on venturi action, in which case ...don't mess with it if it works.

Please feel free to shoot me down on this one guys!

no gunfire here.

Hmm, with a venturi, the pressure drop at the restriction would have to be less than 45psi for it to work. I think the neck would need to be very narrow for it to work. I prefer to use an ejector.

How can you inject a fluid with just 45 psi into a line presurised to 100 psi. There is something missing. Perhaps there is some baffle or something in whose wake the pressure is less than 45 psi, so that you can get this chemical mix in, so as to difuse into the main stream.

Is the Tee connector an original part of the installation? If it is, I would not suggest replacing it with any wye or other connector. This could be a venturi injector. It looks like a tee but the internal construction is much different. That is the only way to inject a 45 psi stream into a 100 psi system.

This isn't another bong plumbing question, is it?

What's a bong?

45 PSI, WILL NOT PUSH ANYHTING INTO 100 PSI.

THE ONLY REASON IT DOESNT BACK FLOW INTO THE FEED SOURCE IS BECAUSE THE FEED SOURCE IS OF HIGHER PRESSURE THAN 100PSI BEHIND THE FEED SOURCE REGULATOR.

You may be surprised what venturi action can do.

"THE ONLY REASON IT DOESNT BACK FLOW INTO THE FEED SOURCE IS BECAUSE THE FEED SOURCE IS OF HIGHER PRESSURE THAN 100PSI BEHIND THE FEED SOURCE REGULATOR."

Nope. with the venturi tube or an orifice plate, you can even create a vacuum within the flow stream in a pipe. They are also called, aspirators, ejectors, eductors. They can be used for steam, compressed air, or liquid (limited by vapor pressure and temp).

Two words:

Bernoulli's Principal

Two more.

Architrave, banana.

Thank you for the replies. The chemical does actually get into the 100psi line with no problems and the tee is not a venturi. It wa spart of the engineers original design, which I built the prototype in our lab. I am trying to get a little better flow of the chemical because I want to reduce the amount of chemical and injection pressure. The chemical is premixed in a manifold and distributed to the main line @ 45psi. I figured that maybe a wye instead of a tee would give it a slight venturi effect and reduce the backpressure in the chemical line, but like I said, the chemical gets into the 100psi line just fine. I am not an engineer obviously, just looking for ways to improve the current design. Our problem is that we seem to get enough backpressure to make priming the pumps for the chemical line difficult, even with checkvalves throughout the lines. Cost as usual is always a thing to consider, but this is a prototype build and we need to get it operating as efficiently as possible at a low cost. Again, thank you for the replies.

I still want to know how a 45 psi liquid can be injected into a 100 psi line without the use of venturi action.... A 90 degree tee shouldn't provide venturi action unless it's installed in a reduced area throat.

This process can't be occurred.

This process does occur. Yes, the line is @ 100psi, verified by multiple gauges throughout. It doesn't seem like it should work but we can visibly verify chemical at the outputs. Again, thanks for the replies. I was thinking about the wyes for the laminar flow spoken of in the previous post. I will try anyways and note the results.

How .... How .... How?

There is probably a slight venturi action, otherwise, it would not work. 'Something like a spray bottle effect:

The question is: will changing the connection from a tee to a wye make a difference in performance. My answer? I don't know.

I guess what happened is that someone just connected the two pipes together and crossed his fingers. It happened to work so all's well. Which brings me to Del's comment, "if it ain't broke, why fix it?"

I know venturi action very well. It is happens ONLY where there is a very high speed of the fluid, in which its pressure shall be reduced to satisfy Bernoulli or Energy Equation. So the venture action can't be established without reduction in pressure, i.e. the pressure inside the main run shall be lower than the pressure inside the branch.

In such a case like that, how can I trust that the pressure at run is 100 PSI where at branch is 45 PSI and the fluid with low pressure transported towards the fluid with of high pressure. The fluid transfer is only due to the difference in pressure, and transferred from high pressure portions to low pressure portions (function of a compressor or pump), other wise the fluid shall not transferred.

With respect to changing the connection from a tee to wye (lateral), it is only required in case of fluids with low pressures like as main flare header and its terminals downstream of safety valves, the connection between run and branch must be lateral towards the direction of fluid flow to avoid fluid disturbance at gathering connections.

For pressurized fluids, no need to change the connection from tee to wye, where the fluid is forced to flow.

Venturi action is sometimes there without a classic venturi restriction, the flowing fluid 'drags' the other fluid (at the T) into the flow. The velocity of the flow generates a reduced pressure?

Again...shoot me down if I'm wrong, but Ive seen pipework underneath a swimming pool where there is obvoiusly a fair head of pressure, and yet the flow will draw in additives if a cock is opened.

The other answer maybe that the 100 psi isn't at 100 psi !

The tee and wye will both act as venturis just due to the "pulling effect" of the main line as it passes by, although one will be better at it than the other. The wyes' advantage would be that it allows for a more laminar mix as it enters the main stream, but probably wouldn't address the back pressure problem.

If you haven't already done so, I would suggest redundant valves on the 45psi line mounted in opposite directions. Most valves are designed to seal the higher pressure on the input side, but since you have the higher pressure on the output side, the pressures will try to equalize (bernoulli's equation) and the valve will be too weak on the output side to prevent it. By putting another valve in the opposite direction, when both are closed, both side will be sealed, preventing the equalizing of the pressures.

Example: When testing pressure sensors over a given pressure range in production, it was much quicker to build a test station with a tank farm of set pressures ran into a manifold for switching pressures vs. continuously ramping a pressure controller. The one drawback was that each time you put a higher pressure on the output side of the manifold, it would create back pressure issues with the lower pressure lines. The problem was solved by using the 2 valves, with the 2nd possibly being a 3-way for venting so pressure between the 2 valves is released.

If the first injection point works...

Add another one the same ! Twice the input... ( at a point a little further up stream or down straem where they will not interfere with eachother)

If it ain't bust ..don't fix it !

Can you explain us how is the design such 45psi liquid gets into 100psi tubing without ejector or venturi action involved? Super high velocity through the 100psi pipeline (crating a vacuum at the injection point)? Just guessing.

That will be an interesting found and subject to discuss on a new forum.

SaC.