Restriction Orifice Plate Sizing

[DeuNucleus]

Hi everyone,
I am currently working on a cooling system. I need to choke the piping network with restriction orifice plates in order to assure the desired flow in each branch and to make sure that the pump will deliver what it should. The thing is, I don't know how to size those plates, I would greatly appreciate if someone could help me with that.

Thank you,
Alex

 

[LittleInch]

err, did you try searching this site?

This pooped up straight away

http://www.eng-tips.com/viewthread.cfm?qid=60695

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[BigInch]

Trying to control flows to multiple individual branches using orifice plates in each one is not a very good method to assure successful control. It is often difficult enough in one branch. Remember that orifice plates only control to one particular flow rate at any one pressure differential, or vice versa. Should any one branch suffer any form of upset, or perhaps need to be shutdown, ALL other branches will be subject to flow redistributions and pressure changes that could severely disrupt normal performance. For example in the extreme, the pump, pipe and equipment may have to be sized for a head and power that allows full flow though the smallest diameter branch, if all other branches were shut down at the same time. Higher flows through fewer branches may overheat, or overcool those branches in an HVAC system... for example.

 

[DeuNucleus]

Good morning LittleInch,
I wanted to make sure I didn't make mistakes in my reasoning. Here is how I proceeded.
1- Calculate the head loss of the system with the desired flow in each branch.
2- Determine the additional pressure drop required to assure the desired flow in each branch and to match the pump.

After that I chose where to put my restriction orifice plates I determined the pressure drop needed for each plate.
Once done I used Bernoulli (This is where I am not sure).

dp = (RHO/2)*(V1^2-V2^2)
then
dp = (8*RHO*Q^2)/PI^2)*(1/D^4 - 1/d^4)

dp : Pressure drop needed [Pa]
RHO : Density of the fluid (P.glycol 30%) [kg/m^3]
Q : Flow [m^3/s]
D : Pipe diameter [m]
d : RO diameter [m]

I tried to verify my answer with the equation given in the topic you sent me (14.6mm vs 16.4mm). But I think there is something I didn't understand... Is the constant ''C'' is dimensionless? Because I did a dimension analysis on the given equation and I find myself with something that didn't match...

Thank you for your time.

 

[bimr]

Here is an online calculator to size orifices:

http://www.tlv.com/global/TI/calculator/water-flow-rate-through-orifice.html

Have to agree with BigInch, don't see why you are using orifices.

Why not use pressure reducing valves and rate setting valves? Orifices are not adjustable.

 

[MJCronin]

I agree with bimr. A flow or pressure controlling device is an acceptable solution....

MJCronin
Sr. Process Engineer

 

[LittleInch]

I think it all depends on how much your RO is as a percent of the overall pressure drop.

Normally ROs are used to provide a physical limit to gas or liquid flow in the event of systems not working correctly.

alternatively they can be used to distribute flow from a header (think fire sprinklers).

~IMO they are only valid for what you're talking about if the percent pressure drop across the RO is over 50% of the total for any branch. It then becomes much less sensitive to any changes in flow, pressure etc, but implies large wasting of energy in your system.

If, as I and others here suspect, you're trying to use these as cheap "trimming" flow control devices, then you're doomed to failure and you really need to consider something else.

in terms of your question about C - I didn't write it, I just gave that post as an example of what is available on this site to guide you.

your best long term bet is some sort of simple flow control valve. This sort of thing is very common in HVAC cooling systems and many valves are available, this is just an example

http://heating.danfoss.com/PCMPDF/VDC6Q602_AB-QM_DN10-250.pdf

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[DeuNucleus]

Thank you all for your answers, but I think I wasn't clear enough about the system I am working on. It is not a HVAC, but a cooling system for an alternator. None of the branches will be shutdown because if so, a component of the alternator will overheat.

And yes the RO will provide more than 50% of the pressure drop for each branch. The reason why we chose to put RO is because we don't want the flow in each branch to be re-adjusted after the installation since all the heat exchange calculation have been made with a given flow.

Mechanical Eng. Student

 

[BigInch]

Well then WHY do you need ANY orifices? Size each line properly to deliver the flow you want in the respective branch.

 

[katmar]

As long as the supply pressure remains fairly constant using ROs is a reasonable way to solve the problem. If the flow rates and pressures vary then you need something a bit more intelligent and the flow or pressure controllers recommended by others would be needed. I agree with BigInch that the system hydraulics (basically the line sizing) should be done to match the available pressure but the reality of the situation is that pipes come in discrete sizes and it is much cheaper to add in a bit of pressure drop using an RO than to add in extra length of pipe.

If you are having to use ROs on all the circuits then a better way might be to lower the supply pressure by having the pump impeller resized.

Katmar Software - AioFlo Pipe Hydraulics

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[Softedge]

if you are locked into using orifices, consider multiple port orifices to minimize turbulence within the system.