Restriction Orifice 7

[jamiegla1]

Hello all,
I am trying to size a restriction orifice that will drop the pressure in a line from 1000psi to ~300psi. Flow is 300GPM in a 3" line. Would venturi be better in this case and would the high DP destroy the RO? Any good eqns to size this? Ive been ref. Crane 410.
THanks!

 

[Eltron]

Assuming you're talking about gas flow, you're looking at a choked flow situation.

When the gas velocity reaches sonic velocity further increases in upstream pressure do not cause any further increase in the gas velocity through the orifice. Thus the flow is "choked." However, the only parameter that is choked is the velocity. As the upstream gas pressure increases, the density of the gas also increases; and since the mass flow rate is a function of density, the flowrate increases linearly with pressure.

To calculate the correct restriction size to use in a choked flow system try the following equation:

d = SQRT(Q/(.01749(P1/29.7)SQRT((29/MWx)SQRT(528/T)))

Where:

d = Orifice diameter / mm
Q = Gas flow / mL / min
P1 = Inlet pressure / psia
MWx = Molecular weight of gas x
T = Temperature / °R

Alternatively the coefficient of flow of a critical gas can be used to determine the restriction size with the folowing equation:

Cv = Q((SQRT SGx x T)/(816 x P1))

Where:

Cv = Coefficient of flow
Q = Gas flow rate / ft3 / hr
SGx = Specific gravity of gas x
T = Temperature / °R
P1 = Inlet pressure / psia

The Crane handbook is an excellent suggestion, but don't discount the manufacturers of the restrictions, either.

Hope that helps.

 

[jamiegla1]

Thanks Eltron,
I am dealing with liquid however (35%wt amine in H2O). THe application is for minimum flow around a pump. THe FCV wont be able to take taht DP so im trying to find a good way of knocking it down before it gets there. Im reading things about venturis, anyone have any experience with them?

 

[Longinthetooth]

Would not a venturi give you full pressure recovery? I.E. no advantage?
If your FCV wont take the DP, neither will a single orifice. It will be sacrificial. One way round this is to use several orifices in series each take part of the DP. You have to allow sufficient distance between each to get full flow recovery. 3 or 4 should do it. I've done this and they last forever.


Regards,

 

[jamiegla1]

Thanks geordie87 that sounds like what i was looking for. Are there equations for sizing these in series?

 

[SeanB]

"THe application is for minimum flow around a pump."
Are you talking about a spillback for pump minimum flow protection or a warm up bypass? If it is for minimum flow protection with a RO you will be continuously spilling back which is very inefficient. Also, 300 GPM in a 3" line sounds high to me. Have you checked the velocity?

 

[toothless]

Call this guy from CU Services. He makes exactly what you are looking for - a prefabricated (based on your specs) staged energy diffuser system. But remember, as your flow drops from 300 gpm to zero the DP across any restriction device will drop thus putting the DP back onto the control valve.

Ramsey Cronfel
CU Services LLC
725 Parkview Cir
Elk Grove, IL 60007
Phone 888-394-2303

 

[jamiegla1]

Sean yes its for minimum flow protection. velocity is high but this is the extreme case and will rarely be encountered. Just looking for the best way to knock the pressure down so control valve doesnt see the brunt of it.

 

[ApC2Kp]

jamieg1,
The contact for CU services would be one good resource. One item of information that needs to also be considered is the fluid temperature and how close it might be to flash or boiling point.
If the restriction orifice or diffuser is down stream of the control valve, then it could keep the fluid from caviting inside the valve. It is probably less cost replacing restriction orifice / diffuser than replacing the control valve / trim.

 

[jamiegla1]

According to simulation, the fluid will not flash (140F) as it goes from 1000psig to 280psig. ANybody know anything about pressure let-down valves?

 

[Assumptions]

Have a look at the Yarway valves. They were made for these application.

If you use an RO on its own for min flow, you will have to add this flow rate into your requirements for your process design flow rate.

http://www.yarway.com/control_valves.asp

 

[TD2K]

Yarway makes a minimum flow valve as assumption suggests that is used in BFW circulation applications which will be more dP than you are taking.

http://www.yarway.com/control_valves.asp

However, they aren't intended for continuous use. I wanted to use one where most of the time I had no forward flow and would be recycling back to my tank, they did not recommend using it.

Conversely, you could look at a pressure reduction valve. Cavitrol trim or a similar type trim will be able to handle this type of a pressure drop. The major problem with these trims is that they have small openings which makes for a very good filter if you have solids.

Your simulation might show that the liquid won't flash based on the downstream pressure but with that drop, I bet it will be cavitating within the valve. I don't have the necessary spreadsheet here but the Fisher catalogue has a procedure to check for cavitation and this with dP, my guess is you will get a cavitation warning. However, most valve sizing programs will check and let you know if you do or don't have a problem.

 

[JimCasey]

You did not specify temperature, but I'm betting that with this kind of pressure drop, you will get cavitation and will be destructive. The reason I mentioned temperature is that if the vapor pressure is higher than the discharge pressure you will get flashing. Flashing can be damaging, but it is not as destructive as cavitation. I don't think you are looking at flashing because you hint that you expect liquid downstream of the orifice.

Others have pointed out (correctly) that you will need a multistage device. Fixed devices, by their nature, have lousy turndown and as someone pointed out as you reduce the flow the control valve will be taking the DP at low flows.

So you choice is to use your valve for on-off with the multistage breakdown orifice, or to use an engineered valve with internal cavitation-control technologies to modulate the flow. Life could also be simpler if you can cut down your pump impeller to decrease the developed head.

One real interesting treat is that amines can be aggressive toward stellite, and severe service valves frequently have stellite insides.

My recommendation is that you get with your favorite valve company's factory application engineers for a multistage anticavitation valve. Valtek, Leslie, Masoneilan, Fisher all have ways of avoiding cavitation damage and you might want to compare their proposals, and relocate your existing valve to a different application.

 

[Mche]

jamiej1,
You can provide two RO in series such that the flashing/caviation/noise can be avoided. This will be cheaper way than providing valves.
Regards,

 

[jamiegla1]

Thank you all for your help and advice

 

[sailoday28]

SeanB has asked a good question. Does vendor know if min flow from discharge is recirculated to pump suction? If so, during this mode, is pump energy into fluid a cause for concern?

 

[jamiegla1]

During normal operation, 300 GPM discharged from pump. ~25 will be recycled via control valve or ARC to the pump suction. How will pump energy into fluid effect the scenario if the ARC knocks down the P to the suction P? Im a new engineer so this is a little hazy to me.

 

[JimCasey]

All the kilowatts that go into driving the pump go into the fluid. If the pump is 70% efficient, the energy goes into potential(pressure) and kinetic (velocity) energy, with 30% of the energy directly heating the fluid. However, if you just recirculate the fluid, any heat not dissipated though the walls of the system goes to heating the fluid in the system, and ALL of the energy consumed by the pump motor eventually goes into heat. If you just deadhead a pump, the fluid inside can boil in seconds. On a short-cycle recirc such as you describe here, the last time I calculated one I got a temp rise of 1.6 degrees F per minute, but that was all very system-specific.

MCHE: 2 orifice stages are better than one, but whether 2 stages are enough depends a lot on how much pressure has to be dropped. Much like calculating how many steps it takes to get from your front porch to your driveway.

 

[Latexman]

This has been an issue of great debate before in my long career, but here goes . . .

JimCasey,

Re: "However, if you just recirculate the fluid, any heat not dissipated though the walls of the system goes to heating the fluid in the system, and ALL of the energy consumed by the pump motor eventually goes into heat."

Not true. Remember "lost work" from thermo class? I quote from Balzhiser, Samuels, and Eliassen, "Lost work is defined as work that could have been performed but was not because of dissipative effects ot irreversibilities.
Whenever an irreversible change within a system leads to the lowering of an energy potential (such as pressure, temperature, electrical potential, etc.), without transferring as much energy to the surroundings, in the form of work, as possible, lost work results and entropy production occurs."

For example, instead of two orifices in series, what if jamieg1 installed the exact length of 1" line and fittings to drop the pressure from 1000 psi to 300 psi? Would ALL of the energy consumed by the pump motor eventually go into heat, or just 30%? Look at it another way. Assume you have the same flow of fluid in an identical 1" pipe and fittings going from point A to point B. Would ALL of the energy consumed by the pump motor eventually go into heat, or just 30%, between points A and B? 30%, right? In this context, it's a no-brainer. The potential energy gets dissipated by friction with no appreciable temperature gain. The same holds for the previous 1" pipe and fittings and two orifices in series. Only 30% is lost to heat. The other 70% is lost to "lost work".


Good luck,
Latexman

 

[djack77494]

Latexman,
I'm sorry, but I can't agree with you on this. If you put 100 kilowatts of electrical energy into the pump's motor, you'll get somewhere near 90 kW out as mechanical energy. The other 10 kW is lost (as heat emanating from the motor) due to the inherent inefficiencies of converting electrical energy to mechanical energy. A bit more energy is lost "on the way" to the pump. I'm speaking of the energy that would be lost in the bearings, for example. Let's say that ultimately 85 kW of mechanical energy is available inside the pump. If the HYDRAULIC efficiency of the pump is (say) 30% per earlier postings (incidentally, that is a very low number), then 25.5 kW of useful work is done on the liquid. The remaining 59.5 kW is dissipated into the pump (as heat). When you take a pressure drop across an orifice and return to suction pressure, you dissipate the useful work (25.5 kW) that you had gotten into the liquid. (Again, as heat or temperature rise.) So, I would contend that all of the energy "consumed" by the pump is ultimately converted to heat.
Let me know if you think I've gotten it wrong.
Doug