Orifice sizing

[wfeldmann]

Does anyone know who to size an orifice with the following parameters. What I am trying to find is the diameter of the orifice I need to have to drilled in a orifice plate in a swagelok fitting to allow 2 gpm flow with a .5 psig drop and the fluid is DI water. The value I came up with was 0.153 inches, but I am not very confident in this value.

flow rate - 2 gpm
tube size - 3/8"
Specific Gravity - 1
Beta - 0.4 to 0.6
Pressure drop across orifice - 0.5 psig

 

[katmar]

Is the 0.5 psi pressure drop the permanent pressure drop (i.e. are you designing a restriction orifice) or is this a flow measuring orifice with close tapping points?

What is the actual ID of the tube? It will have an impact on the orifice sizing.

Just a note - pressure differentials are neither absolute nor gauge. You should specify your pressure drop as 0.5 psi and not 0.5 psig.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[wfeldmann]

Thank you for your response and clarification.

What I am trying to maintain is a 2 gpm flow rate, but it does not have taps. This flow is through a 3/8" SS .065" wall thickness tubing and I was going to use a swagelok union with orifice inside the union to control the flow. That is why I am trying to figure out the diamter of the hole to drill in the orifice. The 0.5 psi was a value that I selected. I was trying to keep the pressure drop bewteen 0.5 and 1 psi, but the main variable I am trying to main tain is the beta around 0.4 to 0.6. To answwer you question I believe the ID of the tubing is 0.375" - 0.065" = 0.31".

The equation I used was the following.

d= diameter of orifice hole (?)
Q= flow rate (2gpm)
K= constant variable for flow through orifice (0.62)
H= Head across the orifice (1.155)

Pressure (psi) = (Head * Specific Gravity)/2.31
0.5 = (Head * 1)/2.31
Head = 1.155

d= (Q/19.63(K)(Hsqrt))sqrt
d= (2/19.63(0.62)(1.155 sqrt))sqrt
d= 0.153"


ID of tubing is 0.375" - .065" = 0.31"
Beta = 0.153/0.31 = 0.493

 

[Latexman]

Better check your math. I got d = 0.391", which is bigger than your tubing!

Good luck,
Latexman

 

[wfeldmann]

You were correct I forgot to take the square root of the entire equation. Basically if I increased the pressure drop across the orifice I will lower my diameter. If I increased my pressure drop from 0.5 psig to 3 psig I would get a diameter of approx. 0.25".

Pressure (psi) = (Head * Specific Gravity)/2.31
3 = (Head * 1)/2.31
Head = 6.93

d= (Q/19.63(K)(Hsqrt))sqrt
d= (2/19.63(0.62)(6.93 sqrt))sqrt
d= 0.249"

My Beta would now be 0.806.

 

[Latexman]

The math checks out now. Congratulations.

However, I have some doubt, because you really did not answer or elaborate on katmar's question on permanent pressure drop in a way that makes me think you know enough about this subject to make it work right the first time. Plus, you use a K = 0.62 with a beta of about 0.8, and you never mention checking the Reynold's number. All these put together creates doubt in my mind. Does anyone else have this feeling?

Good luck,
Latexman

 

[wfeldmann]

Pertaining to katmar's question.
Is the 0.5 psi pressure drop the permanent pressure drop (i.e. are you designing a restriction orifice) or is this a flow measuring orifice with close tapping points?

Yes, it will be a permanent pressure drop with a constant pressure supplied from a header that should be a consistent 45 psig from a pump. I am trying to design this as a restriction orifice and it will not be monitored for flow. There are nine other taps off the main header that are cooling other equipment that also have restriction orifices in them. I have a total of 34 restriction orifices in this system. There are 4 in a 3 inch line, 2 in a 1-1/2 inch line, 20 in a 1 inch line, and 8 in a 3/8 inch line. This tubing and piping are cooling bus bars and a chamber that is being heated by induction coil and they keep the coils and bus bar and associated power units and load matching stations at a desirable temperature. I hope this helps explain what I am trying to do.

I have all the other restriction orifices completed, but I was having a problem with these 3/8" tubing lines, because my software program does not go below 1/2" lines, so I needed to do it by hand.

Thanks for your help

 

[Latexman]

Thanks, that helps explain my doubt. Btw, a 3/8" SS .065" wall thickness tube has an ID = 0.375-2x0.065 = 0.245".

Good luck,
Latexman

 

[wfeldmann]

Latexman,

You have been very helpful and I appreciate it.

 

[Latexman]

No problem. Are you adding this one 3/8" run to an existing facility, or is everything new?

Good luck,
Latexman

 

[dcasto]

use a slinted bronze filter with a Cv of 2.83 from swagelok.

 

[wfeldmann]

All runs are new and there are a total of 8 - 3/8" tubing runs that cool the bus bar. There are two zones per chamber and each zone has a positive and negative bus bar and the tubing is run on boths sides of the bus bar for a total of 8. From our bus bar heat calculation that mechanical engineering performed it stated we needed a flow rate of a minimum of 2 gpm to keep the bus bars within operating parameters. They tee off from a 3/4" pipe and go through the bus bar and tee back into a 3/4" pipe and back to the return of the pump and heat exchangers.

 

[wfeldmann]

What is a slinted bronze filter? How exactly would this work and would you have to worry about contaminants over the long run? I will look it up and thanks for the option.

 

[Latexman]

dcasto - did you mean sintered or slanted? I'm betting on sintered.

wfeldmann - sounds interesting. Are you in the design, construction or start-up phase? How long are the 3/4" supply and return mains? How long is each 3/8" supply and return branch?

Good luck,
Latexman

 

[wfeldmann]

I am in all three phase of the project. We have just finished design and we are currently in fabrication and in 3 months will be testing. Each 3/8" run is 25 feet in length and the 3/4" run is about 15 feet.

 

[Latexman]

Sounds good! I was interested in the size and status of the job.

Good luck,
Latexman

 

[Latexman]

I forgot to ask, did that tubing ID error push you into 1/2" tubing?

Good luck,
Latexman

 

[dcasto]

sinter yes. They are fairly pourous, but if you have trash, then no, they won't work.

Another use is for slowing the air bleed rate on valves. For example, you find that your liquid esd valve is in water hammer and you need to slow the closing, just pop 1 or 2 in and the vent, the valve slows up. We kept several different Cv values of the disks around and used them as needed.

 

[wfeldmann]

Yes and thank you for finding that mistake with the ID for the tubing it caused us to re-evaluate what we can use. We are thinking of taking the restriction orifices out and replacing them with needle valves, because we have so many unknowns and some of the process values we got I don't trust. At least with a needle valve we have some room for adjusting.

What we are building is a plant to destroy stock piled chemical weapons in Richmond, KY. The plant will probably be operation in 3-5 years and operate for another 5 years after that. We expect to have one of our systems finished by this September so that they can start pouring the walls around it.

Thank you for all your help.

 

[Latexman]

Sounds like a noble cause, glad I could help.

Good luck,
Latexman