It doesn't affect the pump curve per as a you have a flow of usually 20 to 30%, but that flow then can't be used to go forward so all the energy for that flow is wasted.

An ARV is much better as it allows recirculation flow when there is no or low forward flow, but closes when you need all the flow.

How does the system sense when you need more or less flow? Is that based on tank level? Or?

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

Does the tank have level indication? You can also add air pressure to the tank to maintain ~50% level without needing a bladder. An operator will need to periodically adjust the air pressure in the tank to maintain level. You will need to add a pressure relief valve to the tank.

Only if the tank is a pressure vessel.....

We could though do with a sketch of the system.

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

You would need to size the orifice to be at a flow which is less than 65 psi on your pump curve for one unit.

No idea what this is unless you post the pump curve but might be quite a lot of flow.

How did the system turn the second pump off? What pressure is it set to revert to one pump?

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

I assume the lead pump will be continuously running so the orifice would be sized for the manufacturer's recommended minimum flow at corresponding discharge pressure of pump at the pump curve differential head corresponding to the recommended minimum flow. The lag pump could still cut in at 50 psi I assume but it would cut out significantly before minimum flow is reached so that both pumps would not operate at minimum flow simultaneously. With a recyce line like this you are going to loose some capacity to the end users per what flow is being recycled but with two pumps available capacity you still may have more than enough flow.

Quote (LittleInch)

Only if the tank is a pressure vessel....

In this case I believe it is a pressure vessel.

Shipboard potable water systems use bladderless hydropneumatic tanks for pressure dampening. They're quite simple though the watch may have to adjust the level every few days.

LInch seems to be talking about the supply tank. Tug seems to be talking about the busted bladder tank.

I can't see why you would want to run pumps constantly. Can you explain? Just turn them off when flow is below minimum. Usually a high pressure switch does that job. Where is it?

Apparently you are blowing out the bladder, possibly by the pressure spike of one or the second pump quickly turning on, or by the pumps discharging at too high a pressure. Why not just solve that problem? Again, the high pressure switch could be used for that too.

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

It appears that you have a bladder tank with bladder that is not designed for the dead head pressure of the pump or just on the borderline so it keeps on busting. This appears to be the case since the pump start is at 65 psig so the maximum pressure output of pump must be close to 100 psig at zero flow (when bladder is fully compressed).

Ideally the best solution would be to replace tank with a new tank of higher design pressure. The attached has tanks available in pressures up to 300 psig rated pressure. The tank should be sized to give a minimum run time of pump once it cuts in. I believe that most pump motor manufacturer's recommend about 5+ minutes minimum run time and no more than 5 starts/stops per hour. So when the pump cuts in the vessel bladder size should be such that it will take 5 minutes to compress from where it is at at cut in pressure to the discharge pressure. Without a recycle line a high pressure switch shutdown should shut off the pump when it reaches recommended minimum flow.

The orifice recycle line is a cheaper option to replacing the bladder tank but your pumps will run continuously causing them to have more maintenance issues and reduce overall life, and use up a lot more energy and cost to opearate.

• https://files.engineering.com/getfile.aspx?folder=0bf55ee1-bed3-47da-9760-5

We don't know the duty cycle or the pumps. They may run 90% of the time in which case s recirculation line is acceptable. They may run 10% of the time in which case the recirculation line is not acceptable.

I'm asking why recirculation is even needed. A recirculation line only reduces operating pressure and useful flow and increases energy and maintenance costs. Nothing acceptable about that. Its just an emergency hack to keep it running. Recirculation is for mixing, temperature control, or a temporary means to get to operating condition. Without further explanation from OP, it seems to be the wrong way to attack this problem. 5 billion systems just like this run just fine without recirculation, using only low pressure start and high pressure cut out.



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

[highlight #EF2929]@LittleInch
How does the system sense when you need more or less flow? Is that based on tank level? Or? The demand for flow will fluctuate throughout the day and the PIT downstream of the pumps will regulate the pump status.

How did the system turn the second pump off? What pressure is it set to revert to one pump? Second pump will turn off once system pressure rises above its cut-in pressure so 50PSIG.

@Snickster
I don't think design pressure is an issue. There may be sediments or rust in the bladder tank that's puncturing it. But that's just a theory.

@TugboatEng
You can also add air pressure to the tank to maintain ~50% level without needing a bladder / bladderless hydropneumatic tanks - This is a good suggestion but does not seem operator-friendly.


@1503-44
You are blowing out the bladder, possibly by the pressure spike of one or the second pump quickly turning on - looking at the trends, we do not surpass the 125psig design pressure
Emergency hack to keep it running - exactly what we need. Any suggestions?

There are multiple operating curves for Q vs head. Which one is yours?
What is supply tank operating pressure? And max normal level?
Lead pump cuts in at 65psig, and cuts out at ??

If the bladder is capable of withstanding the max cutout pressure from lead pump, bladder rupture may due to some corrosive element in this water, or microbial growth. Most bladders are made of some cheap rubber (Buna N or similar), so see if there is a better material on offer.

I say a quick fix is as someone previously suggested which is to operate the tank without bladder with just an air cushion. The tank is an ASME tank rated for 125 psig. If you empty the tank and fill with air at atmospheric pressure with the system shut down, as the tank fills it will compress the air space. At 65 psig the space is compressed about 5 to 1 to provide same functioning as tank with bladder. Only difference is that over a period of time air will dissolve in the water so will need to recharge tank. Sight glass on side of tank can be use monitor air charge. Air dissolves in water by Henry's law I believe which basically states that water has a certain saturation point of air in water for a given temperature/pressure. Since you are taking the water from an atmospheric tank the water may already be close to saturation with air so once in the system the air in the tank may not dissolve into the water any further to any great extent so it may keep it's air charge for a while without needing to add any more - worth a try. I recall air cushioned expansion tanks were used commonly in water pumping systems before bladder tanks became popular when I was a very young engineer.

https://www.eng-tips.com/viewthread.cfm?qid=501898

If you prefer not fixing what you've already got, you could try the air cushion method, but only if inlet and outlets are in the right places. I don't see where those are.

Several people have asked what turns the pumps off and at what pressure.

What is your theory as to why the bladder is failing?

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

Mounthorn,

There are still some answered questions though.

1) What actual flow is there over the space of 24 hours? i.e. flow rate vs time of day chart
2) How often and for how long does pump 1 just stop?
3) Do both pump just start and stop within +/- 1psi of their set point? - So no start pressure / shutoff pressure difference? That's very strange.
I can't work out how the second pump works if that's the case as as soon as it starts the output flow from pump 1 will halve and the pressure rises by 5-7 psi instantly and then the second pump stops as soon as it starts? Doesn't make sense, but if it is doing that your second pump and motor are not going to last long....
4) My guess from the data so far is that your pump is the -10 series on the curve above - can you confirm?
5) You do realise wrt the bladder that those tanks work on the bladder being filled with water and the air is the cushion on the outside? The instructions says to pump up the tank to about 2-3psi below the pump start set point - i.e. about 62 psi. Is that what you have done? If not then what you have done is inflate the bladder well beyond it's design size as the initial pre charge from the factory is only 12 psi. Did anyone actually read the installation manual?

Here for anyone interested in how these tanks work http://www.johnwood.com/wp-content/uploads/2014/03...
note item 2 under operating / maintenance instructions.

I don't think the idea of filling the tank with air is a good one here as there is a risk of the air getting into the supply line and then creating all sorts of issues, Especially if you don't have a level guage. The water volume will then be used to supplement the first pump flow for a while, but if the pump stops at 65 psi then it won't be able to pump much water into it.... In either case, the amount of water stored will depend on tank size ( you haven't told us), but will probably be only 10-15% of the tank volume.

But in answer to your first question, yes it could work, but I think the flow would need to be quite high through the orifice plate to get the pressure below the cut off point for the pump assuming that at times you have no or very little flow and your pump is the -10 or even the -9 version. So maybe 40-50% of your max flow is going no where, but you're paying for it. Also your motor and pump are then running continuously so wearing out faster and your second pump will probably start operating a lot more as a lot of pump 1s flow is going straight back to the tank.

If you're serious about doing this long term, then look into ARV (Automatic recirculation valves) which vary the return flow depending on outlet flow and shut off automatically. But if there is low or no flow for long periods you are basically just throwing energy (money) away.

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

PS. LInch gives you some good advice there.)

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

Bladders in bladder style pressure tanks go bad from the bladder going up and down with each pump cycle, which is like bending a wire back and forth until it breaks. Bad tank bladders are a sign of too much pump cycling. Controlling the pumps with Constant Pressure Valves would be an easy fix regardless of how much the flow rate varies. In this way there is no recirculation.

Temperature and being stretched way beyond what they were designed for I will give you, but if they don't actually move at all there is no point in having them.

There may be some sort of hysteresis and max no of cycles. Constant pressure valves wouldn't solve the issue of lack of throughput though...

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

As long as it is pumping cool water there is plenty of throughput in a constant pressure valve. Just doesn't work with hot water.

A pity mounthorn never got back to us....

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