First Principles Fluid Pumping Question

[SwinnyGG]

Howdy gents,

I'm a mechanical engineer but pumps and fluid power systems are not my forte.

I've got a problem I'm trying to solve and I've done some calculation, and think I know the answer, but I'm looking for a sanity check.

I have two tanks - tank A is roughly 800,000 gallons, and is roughly 30 feet tall. Tank B is roughly 3 million gallons, and is roughly 55 feet tall. Both tanks have nozzles in the lid, and a nozzle 3 feet above the tank bottom. Both tank foundations are the same elevation, and the tanks are roughly 20 feet apart (20 feet of pipe run between tanks, so when I do math I don't worry about the losses in the pipe too much).

I need to fill tank B from tank A as quickly as possible. Assume tank A is full, and I open valves between the two and allow gravity to drive water to equilibrium (tanks are self-venting and will equalize).

Question is- once I start having to pump water into tank B, which of the two options is faster:

1) Run the pump output up the side of the tank and dump in from the top. Pump will be continuously pumping against approximately 55 feet of head.

2) Connect the pump to a nozzle 3 feet above the bottom of the tank. Pump head will start out near zero downstream head, which will increase in a linear fashion as the water level rises in tank B.

Question is... is the simple 'downstream head is all that matters' principle actually correct? Does the volume of water downstream of the pump play a factor at all in the load on the pump, or is it simply the difference between height of the pump and the height of the water level downstream?

 

[LittleInch]

TBH this forum is usually about high pressure hydraulic oil ands would be better in the pipelines, piing and fluid mechnics forum next time.

But anyway, your question.

The inital part of your fill will be a transient flow as the head difference is constantly reducing form your start point if you are just doing this under gravity.

So flow will start to fall probably in an exponential curve.

So you need to decide at what point does your pump take over.

Then you're just into practicality.

If you want to use a large volume low head centrifugal pump, you are almost certainly better off doing the feed into the top of the tank and optimisising your pump for efficiency and flow.
But you could flow into the lower nozzle and let the pump run with a bigger flow but lower head. The problem is that there is a limit before the pump is running at too high a flow (referred to as end of curve. So you might need to insert a control valve to slow the flow down to some pre set maximum so you don't damage the pump.

Having said that, your larger tank is only going to fill to 15 foot high assuming you start empty.
So just buy or use a high flow low head pump and pump into the lower nozzle. Just make sure your pump is at the smaller tank end and do the maths to make sure you don't flow too fast when the big tank is pretty empty.

I think that's a long way of saying the issue is head difference between the level of the water in the first tank and level in the second tank. You might get a big swing and if the friction losses in te pipe are low then you are possibly better off with an axial flow pump - high flow low head.

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

 

[SwinnyGG]

I appreciate the response.

Sorry about the wrong forum, I'm not in this neck of the woods much.

That all makes sense. The pump is the same in either case (there is a real physical pump which is already on the site) and I had planned to pump from ground level, but went down a weird rabbit hole of calculating how much work my pump is doing by raising X gallons of water already in the tank by X number of feet in so many seconds etc. Then the realization hit, maybe I'm overthinking here and the simple answer is the right one.

 

[itsmoked]

Other points to consider that actually matter though not hydraulically.

A pipe to the top needs no valve as it is a fixed amount of water that can come back if that line is opened whereas one is definitely required to the lower nozzle. If that lower valve acquires a leak the fix of opening it up or replacing it could present some difficult issues. Multiple valves should be considered.

Every valve can be closed causing confusion or other operational issues. That could be avoided with over-the-top delivery.

Pumping into the lower nozzle will likely stir up settled debris in the bottom of the tank severely impacting the water quality for a significant period of time. Top-fill with any water already in the tank avoids this.

Keith Cress
kcress -

http://www.flaminsystems.com