Pipe draining back to tank 1

[Sparky4598]

I am struggling to figure out if a closed loop pipe system could drain all or even some of the fluid out of the pipe back into the reservoir after the pump shuts off.

Fluid is 50/50 propolene glycol / water. The system is a chiller fluid system. The reservoir is a vented tank. The pump pulls fluid from the tank, up 40ft vertically and into the building, horizontally about 150ft, down 35ft to a sealed heat exchanger, and back up and out to the tank outside the building. So the inlet and outlet of the pipe system is at the same elevation.

I originally thought there is no way it could drain out as long as no air is in the pipe or enters the pipe some way. But then I thought could the weight of the fluid cause enough of a pressure drop at the top of the pipe to cause the fluid to boil and allow some of it to drain back and overflow the tank after the pump is turned off. EDIT: There is 1 butterfly valve and 1 check valve that closes when the pump turns off.

If I left out any details necessary, I Apologize. Just let me know and I can add any info necessary. Thanks!

 

[georgeverghese]

My suggestion to relocate the return line at the tank to a lower position is going to encourage steam hammer on restart, since downcomer vertical drop is approx 100ft > 35ft of PPEG/water static head.
Any sequence of operations of this butterfly valve for startup or shutdown isnt going to help - during low PPEG flow, the downcomer liquid holdup is going to drain out into the tank anyway.
You need a bigger tank to accommodate the liquid holdup in this 200ft downcomer that is going to almost completely dribble out during low flow. Otherwise install another tank in parallel with this one with a level equalising leg and vent line. No need for automated sequencing of butterfly valve on return line. Max level with 200ft downcomer completely drained out with combination tanks to be kept below the entrance nozzle of the PPEG return line at the current tank.
If the pressure drop in the 200ft downcomer with a vertical drop of 120ft at max flow is > (120/34) * 14.7 = 53psi, the downcomer will be completely liquid full. If dp is less than 53psi at max PPEG flow, max liquid holdup will be less. This amounts to a very high 26.5psi/100ft over the 200ft of of straight pipe length in this downcomer.

 

[1503-44]

George, what 200 ft?

With a backpressure at the valve of 18 psia, the downcomer will not drain. The flow rate with that backpressure will depend on the pump curve and pipe diameter (unknowns to us) and it must become the minimum system flow rate to avoid hammering during operation and at startup.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[katmar]

George, you have misread a dimension somewhere. The overall height is 40 ft. The hammering has been reported as occurring only at start up against a closed return valve. It is the combination of the vapor bubble at the highest elevation collapsing on start up and the shock wave reflecting off the closed return valve that is the source of the problem.

LittleInch has proposed eliminating the vapor bubble by trapping pressurized liquid in the circuit at shut down. Because the liquid is incompressible only the slightest leak of liquid through one of the valves would be required to lose this pressure while the pump is off and the vapor bubble may still form. Only implementing the proposal will prove it one way or the other but I believe it would be worth trying.

My proposal was to accept that the vapor bubble would form and to eliminate its rapid collapse and the bouncing of the shock wave by starting up at a very low flow rate and with the return valve open. The vapor in the bubble is the equilibrium vapor from boiling under vacuum and not air, so a high liquid velocity is not required to remove the vapor bubble - as would be the case if it was trapped air. This would be a more complex and expensive solution and it would make sense to start with LittleInch's plan.

The best solution of all would be to eliminate the unnecessary vertical loop by re-routing the piping at grade level. This would solve the hammering problem and also the possibility of the lines draining back into the tank and overflowing.

Katmar Software - AioFlo Pipe Hydraulics

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[1503-44]

Or just find out what the minimum flow is to keep the line packed and see if that is compatible with the chiller operation. There may also be some opportunity for On/Off operation when only minimal flow is required, if lower flow rates are required at all. Especially since a constant speed pump is there now and the system itself is not going to be suited to vfd. There apparently was not much concern at operation over a wide flow rate range to begin with. So let's don't just automatically assume that lower flow rates are necessary.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[georgeverghese]

@sparky has drawn a vertical leg and marked it as "80ft horizontal", so which is it ?
I've interpreted it as an 80ft vertical drop. Total straight length of the downcomer ( verticals + horizontals) = 200ft.

@katmar, sequencing the automated butterfly valve during startup/shutdown as suggested is of limited use. The valve is open during normal operation, and at low PPEG flow, liquid holdup in the downcomer will be next to nothing, if indeed this entire downcomer is self draining.

The sketch shows 2 sections of "horizontal" runs, one of 35ft and the other 40ft right now. If either of these dont have a healthy say 1:50 downward slope, air from the tank entrance nozzle is going to struggle to get up (in the reverse direction) to the top sections of this downcomer to replace air entrained in the PPEG ( flowing in the forward direction) to break transient partial vacuum. What is it like now on these horizontal runs? If a healthy downward slope is non existent, I'd suspect this downcomer is rattling with steam hammer in normal operation also. There is also a low point liquid leg of unspecified length ( which contributes to additional liquid holdup and (worse still) doesnt help to allow air to travel in the reverse direction) at the automated butterfly valve.

That low point leg on which this butterfly valve sits is a symptom of rookies setting up the piping on the downcomer. And the process engineer on the engg team probably didnt have a clue either. At the least, the entire downcomer run should have a healthy continuous downward slope to allow for free countercurrent flow of liquid and air phases.

 

[Sparky4598]

Wow I have missed a lot of responses. I'm sorry, things got busy and two other projects took primary focus. I will do the best I can to respond and catch up,

So we did try manually closing the butterfly valve before the pump was turned off and the water hammer issue stopped. Tested several times to be sure. The plan is to modify the control logic to close the valves before the pumps turn off. The manufacturer of the chillers is sending someone out this week to make the mods.

To reply to some key comments I picked out:

The real root of the problem is the unnecessary 35 ft rise

the vertical rise is necessary as the pipe has to cross several walkways and forklift traffic areas and going under slab is certainly not an option.

Exactly - and so did whoever did the routing - was it you?

the routing was sort of set by the building design and the location of the HPU and the chiller, but yes I had the heaviest hand in it. The chiller manufacturer elected to supply the check and butterfly valve.

@sparky has drawn a vertical leg and marked it as "80ft horizontal", so which is it ?

it is 80ft horizontal if it is labeled 80ft horizontal. I apologize if my hand sketch was not clear enough. That is why I put labels. The horizontals have an approximately 1:20 slope towards the chiller in each horizontal run.

Flow Rate = Q = Unknown

flow rate is 55gpm. Friction loss is unknown, but pipe is 2-1/2" Sch40 with approx 29 total 90deg bends. The sketch I shared was a slight simplification so not to confuse people any more than necessary. There were some bends to go around building columns, etc.

Thank you all for your responses and insight here, very much appreciated!

 

[LittleInch]

So that is the valve on the entry into the tank?

Glad it worked.

Maybe next time make it a proper closed loop with a pressurisation / expansion tank....

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

 

[katmar]

@Sparky4598 - thanks for the report back. These reports are rare, but very useful to those who were involved in the discussion and to those who find the thread years later when researching similar problems.

Katmar Software - AioFlo Pipe Hydraulics

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[1503-44]

Is there a minimum flow rate, or just the 55gpm.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[MedMilou]

The main concern in this scenario is whether the fluid in the closed-loop pipe system can drain back into the reservoir after the pump is turned off. This could happen due to the weight of the fluid causing a pressure drop at the top of the pipe, leading to boiling and overflow.

Factors affecting drainage

Several factors can influence whether or not the fluid will drain back into the reservoir:

Fluid properties: The viscosity and density of the fluid mixture (50/50 propylene glycol/water) play a role in determining the pressure drop and boiling point.

System geometry: The elevation changes and pipe lengths affect the pressure distribution and potential for cavitation (vapor formation due to pressure drops).

Air ingress: If air enters the system, it can create pockets that disrupt the fluid flow and lower the overall pressure, increasing the likelihood of boiling.

Pump operation: The pump's shut-off characteristics and any residual pressure in the system can influence the initial flow direction after the pump stops.

Preventing drainage

To prevent drainage, consider the following measures:

Maintain proper system pressure: Ensure the system operates at a pressure that exceeds the vapor pressure of the fluid at the highest point in the loop. This can be achieved by adjusting the pump settings or using a pressure relief valve.

Eliminate air pockets: Regularly vent the system to remove any trapped air that could contribute to pressure drops and cavitation.

Install a check valve: A check valve at the lowest point of the loop can prevent the fluid from flowing back into the reservoir.

Consider a surge tank: A surge tank at the top of the loop can absorb pressure fluctuations and prevent boiling.

Use a vacuum pump: If necessary, a vacuum pump can be used to lower the pressure in the system below the atmospheric pressure, ensuring that the fluid remains in the pipes.

Additional considerations

Material selection: Choose pipes and components made from materials compatible with the fluid mixture and capable of withstanding the system pressure and temperature.

Insulation: Insulate exposed pipes to minimize heat loss and prevent the fluid from reaching its boiling point.

Regular maintenance: Regularly inspect and maintain the system to identify potential issues such as leaks, air ingress, or component wear.

By implementing these measures and carefully considering the system design, the risk of fluid drainage can be effectively mitigated, ensuring the proper operation and safety of the closed-loop pipe system.

 

[Sparky4598]

So that is the valve on the entry into the tank?

yes it is. And yes, ideally it should have been a closed pressurized tank. There was not nearly enough coordination between the chiller manufacturer, the HPU manufacturer, and myself. Lessons learned.

@Sparky4598 - thanks for the report back.

absolutely no problem! I always do my best to post a resolution. I always hate when I come across a thread that would answer a question I have and there was never a resolution or answer posted.

Is there a minimum flow rate, or just the 55gpm.

the minimum flow I suppose would be 0 when the pump turns off, but there is no variable speed pump or throttling valve installed in the system. It is intended to be constant 55gpm during operation.

 

[LittleInch]

Medmilou - chat gpt??

Sparky, glad it worked, it's always good to know.

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

 

[1503-44]

0 min flow rate confirms my suspicion of 15 May 2nd post. So I'd just shut down with and maintain a pressure above 18psia, as I mentioned on 15 May first post. That confirms that keeping 18 psia and above at shut down can probably solve this problem. When you start up again, there should not be any hammering. If there is, your pump start should probably be slowed down.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."