8 replies to this topic

[go-fish]

Hi,

I am working on closed cooling water system which has an expansion drum to adjust for volume expansion. My question is that since its a closed system is there a need to provide a pump min. flow line from supply pump discharge to the expansion drum?

Thanks

[ElSid]

No. the expansion tank works on pressure and is there for thermal expansion (pipe stress relief) and surge protector. Many engineers try to use it as a reservoir ... think of the volume in the tank and let me know how it works as a reservior
ASPE May 2011 Breaking the Paradigms of 21st Century Pressure Booster System Design David P. Carrier

"Tanks don’t store much water! As a result, the pump downtime is not very long. The actual amount stored can be calculated by Boyle’s law: P1/V1 = P2/V2. Simply put, the pressure in an enclosed vessel rises on both sides of the bladder proportionately. If the precharge is 80 pounds per square inch (psi) and you add 90 psi, the pre-charge must be increased to 90 psi, which means only a finite amount of water will be stored based on the pre-charge and the shutoff pressure.

Here’s a typical rule of thumb: The average storage is about 10 percent of the total volume of the vessel. Thus, a 120-gallon tank will store 12 gallons or less of water for downtime. Sounds more like a drip-load, don’t you think? When you apply variablespeed drives to a typical system, there is no pressure change at the tank; therefore, a bladder tank mounted next to the variablespeed booster does nothing except waste your client’s money."

He forgot surge (assumes VFD's can handle everything) and thermal expansion (which we deal with in processes)

[fallah]

go-fish,

It (min flow line) is needed either going to the suction line or to the expansion tank; because there would be the possibility of discharge line blockage. Of course, to keep the expansion tank level it is needed to provide the make-up water entring the tank.

Edited by fallah, 08 December 2012 - 02:29 AM.

[go-fish]

I am a bit confused. According to Elsid's response, expansion drum is not a reservior so min. flow line is not required. But Fallah mentions that a min. flow line is to provided for blocked discharge. But then don't we need to size the drum for certain holding time + expansion volume?

My question was about whether do we require pump min. flow line for pump protection in this case or not.

Also, I am trying to figure out the hydraulics and pump differential in the system from the plot plan. I am modeling the system with the supply piping network and calculating the line losses. In order to account for the return piping network, I am adding a pressure drop element with delta P = supply piping network line losses + pressure drop across each HE at the users. Is the approach correct or am I missing something?

[fallah]

go-fish,

Minimum flow line for pump protection is required either in closed loop or nonclosed loop, because being in a closed loop doesn't remove the possibility of the pump shut off while is in operation.

For calculating differential head of a pump in a closed loop you should calculate frictional head loss in the loop from the pump discharge to the pump suction. In fact, static head wouldn't be included in the differential head of a pump is operating in a closed loop. Obviously, to calculate differential head for a pump in a network with several parallel loops (due to various users) you should consider the loop with highest head loss.

[go-fish]

@ Fallah

I am considering a control valve located at supply lead of each user in order to maintain a constant pressure at the battery limits of each user. Is the approach correct or should I remove all control valves and restrict the battery limit supply pressure only on the loop with highest head losses and let the other users get whatever pressure they are getting?

However, my understanding is that a control valve will be required at each user in order to limit the flow to the HE. Please advise.

Edited by go-fish, 10 December 2012 - 02:49 PM.

[fallah]

go-fish,

The hydraulics of the cooling water network are too complex and it is so difficult to design the system such that the hydraulics balance to get the right flows to all the users. A pressure control valve at each user battery limit cannot provide required flow for the user and flow control valve might to do so.

Anyway, cooling water pump should be sized to handle a specified rate of the water based on summation of all users requirement with a head such that can cover farthest and highest user.

Normally actual flow is set at each user by manually adjusting a manual valve, located at user battery limit, until the desired process temperature is achieved.

Edited by fallah, 11 December 2012 - 01:57 AM.

[go-fish]

Is there a good reference literature for designing closed cooling system?

[ElSid]

Tons of resources. Let me go back a bit. Yes, a minimum flow is required for some systems (i.e. cooling tower) and pumps (motor limits). Ask the manufacturer. Closed loop or open loop, the difference is NPSH. There are ways of self balancing branches for any type of flow In commenrcial applications, they are called PICV (Pressure Independent Control Valves) http://www.oeashrae....trol_Valves.pdf the industrial "name" which has been used for decades eludes me at the time. As for references, Bell & Gossett has some. Look at Mechanical Engineering forums/books as we do these every day

I wish this thing would send me an e-mail when someone responds to a thread I've posted to! Sorry for the delay.

UPDATE: Valves are called Rate of Flow Control Valve or Flow Control Valve. Example http://www.flomatic....es&r=95&pid=241

Edited by ElSid, 11 December 2012 - 06:12 PM.