Positive Displacement Pump Flow While Off

Depending on the pump, and the head of piping run on the suction side, maybe, maybe not.

A back flow preventer is in order, regardless.

Excess flow valve prevents "excess" flow, not back flow.

I think the way to go is run a test...then take appropriate action...

I think this is what emergency shutoff valves are for....You could maybe put a differential switch near the end of the line that shut a valve on the suction line and locked out the pump with manual reset...that way if pressure was lost in the line the control would shut the valve and pump off until it was reset... A differential type switch that would not be operable until the pressure reached the minimum cutout then would have to maintain the pressure within a range of say for instance 100 psi max and 80 psi min, then once it left that range the switch would activate...

http://teecoproducts.com/wp-content/uploads/2015/08/NH3-Anhydrous-Ammonia-Products.pdf

If the break occurs in the discharge line of the pump, one would assume that the pump is running at the time, so the leakage would be the pump capacity. If the pump is stopped there will be some resistance to the flow. You should know the clearance of the vane in your pump so you can calculate an equivalent orifice size and get a rough idea of the flow.

However, there is no way that I would allow the excess flow valve to be removed under any circumstances. A failure on the suction side of the pump would result in a 100 psi jet of NH3.

It would be interesting to know what the pump discharge pressures are when running and when off. There are a few possible scenarios.

If the pump discharge pressure drops to zero, the upstream ammonia pressure could drive the pump and motor. This would give some, but limited, resistance to flow.

Another item of concern is if the excess-flow valve closes, then the pump will run dry, with damage likely.

Were the <...positive displacement pump...> to not be mechanically secured at stop, then it is quite possible for flow to occur, as the <...100psi...> would be there to drive the <...rotary vane...> round.

<...If there was a break in the piping on the discharge side of the pump, would the liquid NH3 flow through the pump?...would there be any significant pressure/head losses through this pump?...> Yes - up to <...100psi...>, as this driving force is available to drive the <...pump...vane...> round were the pump mechanically not secured at stop.

A good solution would be to install a manual positive shut-off valve upstream of the pump instead of relying on the <...excess flow valve...> and the <...positive displacement pump...>.

If the motor is directly coupled to the pump, there will be significant flow if the discharge line is open, until the pressure in the storage tank is just a little above atmospheric. A worm type gear reducer between motor and pump will effectively stop pump shaft rotation, but that will slow down the flow, not stop it.

In the case of a locked shaft, eventually the fluid will equalize, due to the 'slip' characteristic of the pump. Few pump designs have zero slip when not driven, perhaps certain reciprocating/piston & valve designs. You can determine the slip characteristic of the pump by comparing actual flow rate at a given speed to the volumetric flow rate at that speed and pressure, with no internal leakage (pump curve). The internal leakage through a vane pump will increase as the pressure drops, due to less vane seating force available.

Rotary vane pumps are used as Hydraulic and pneumatic motors so I think a break in the piping on the pump discharge could just turn the pump into a motor if the prime mover was not active.

I believe that you need to set the pump up so the flow is detected by s suitable sensing arrangement. When the prime mover is off then shut off valves on the supply and discharge need to be actively closed. With the correct sensing and control by a PLC then both underflow and excessive flow can be programmed into the protective control.

Underflow can be a closed valve, blockage or pump failure while excessive flow could be a broken pipe.

Couple of things... first, you need to ensure that the workplace safety rules enforced where you live permit you to even consider the removal of the shut off valve.

Second... forget the idea completely.

If you have ever been involved in an active liquid NH3 leak event, you would know how fool hardy this effort is.

I have been involved in such an event... when a blade on a scrapped surface rotary cooler managed to penetrate the wall of the cooler and liquid NH3 was released to the atmosphere. Myself and another team member went into the process area wearing SCBA equipment to look for a missing worker and it was quite the experience.

In a nut shell... this is a bad idea.. period!

The leakage through the pump will depend on the design. If the pump is reciprocating piston with spring check valves, then the flow rate could be considerable as the valves may not restrict flow at all. If the valves are cam operated, the flow could be zero. For a gear, scroll or rotor type pump, the flow rate would be determined by the rated internal leakage of the internal clearances of the pump. Some pumps I have seen have a ratchet and pawl device to prevent the pump from freewheeling, but these are rare.

NH3 is nasty stuff, potentially lethal, and I would not only have the excess flow valve, but a positive shutoff that can be triggered with appropriate pump recirculation to allow safe shutdown of the system in the event of a line failure.

I'm reminded of a local story a few years ago where a metal thief attempted to steal copper piping from the roof of a grocery store in Vancouver, WA. Unfortunately the line he chose to cut was in an ammonia refrigeration system filled with anhydrous NH3. It killed and flash mummified him on the spot.

1. The OP stipulated a rotary vane pump, so the other types are irrelevant.

2. If the store's piping was copper, it did not have ammonia in it.

So the observation that flow rate would depend on internal leakage in the pump and potential freewheeling of the pump would still apply.

Revise copper to copper based alloy and you still get a dead metal thief.

<...NH3 is nasty stuff, potentially lethal...>

So is OH₂ - one can drown in it if one isn't able to swim...

One of Einstein's early development was a magnetically driven sealed compressor pump for a residential refrigerator. It was inspired by a newspaper report Einstein read about a family that was killed by a leaky pump seal in an ammonia based home refrigerator.

PS: The metal thief remains dead.

That point is not understood here.

<Unsubscribes>