Continous Flow Loop

[imaqtno]

Has anyone here designed a continuous flow loop before? The difference is - imagine a sphere inside a 10" pipe that needs to be controlled between 2m/s and 4 m/s. The sphere cannot (and must not) pass through the pump (therefore the flow and discharge will be introduced using a 'wye'. Essentially I am trying to run this slug continously in a loop. I do not want to use any flow control valves if possible.

I have thought about 2 pumps at 180degrees from each other but I am concerned that the pressurized flow from pump 1 would enter into the inlet of pump2 and making the circuit ineffective.

Thoughts????

 

[PUMPDESIGNER]

No experience on this, but the following seems correct.
Install a "pit row" pipe, like on circle race tracks. The pipe wyes off the loop, runs straight, then through a circulator type pump controlled by VFD, then the pipe wyes back into the circle.


PUMPDESIGNER

 

[imaqtno]

I may not have the luxery of a VFD. In reality, I am circulating a pig (electronic measurement tool for inside a pipeline). It measures approx 10" x 120"L. It will offer a lot of resistance to flow.

The concern I have is that the pig gets stuck and the fluid just recirculates in the short circuit (racetrack "pit row").

I was toying with the idea of 2 pumps, with the discharge of one near the inlet of the other, but 180 degrees apart in the flow circuit (like 2 pit rows but short circuiting.

The other idea I have been toying with is splitting the flow into 4 - essentially on each leg of the cirucuit (but keeping the suction in one spot). I may use 2 pumps again (therefore 8 "pressure" points).

Any thoughts about a high pressure (say 2000 psi) on the inlet of a pump?

BTW, thanks for your reply.

I have been launched into fluids for this project and haven't touched this subject for 15 years. My specialty is more heat transfer, shock and vibration.

 

[StoneCold]

imaqtno
If you have one pump or two you end up with a spot where the pig is approaching the discharge of one of the pumps and the downstream pressure will be higher than the upsteam pressure and the pig will have to stop. But if you have one pump and move the suction and discharge points using automated valves and signals from where the pig is located you could keep the pig moving. For instance if the loop were a giant circle like a clock. And the discharge of the pump was located at 12 O'clock and the suction was located at 10 O'clock when the pig moved to the 9 O'clock Position you would discharge of the pump to the 8 O'clock position and the suction to the 6 O'clock position. Once the pig was past the 12 O'clock position you would switch the flows back. This assumes that the momentum imparted by pumping into the ring using directional flow is enough to keep the flow in the ring going clockwise regardless of where the Pig is and friction.

Another idea that might work is to set it up like a hot wheels race car set where the car coasted around the track until you hit the little spinning wheels that pushed the car along. Do the same with the pump only pump through the loop for 80% of the time then pump through a bypass while the pig goes by then switch back to pumping again.

I admit that both ideas are a little crazy but then again not many people want to pump a pig around in a circle either. Which begs the question where do you put in the corn?

StoneCold