Bladder Tank

Additional information is required.

How much water is being used per hour? (And thus how many times are the pumps currently cycling?)

What is the desired (maximum) number of pump starts per hour? (And thus from the above and pump capability will give you indication of how much storage volume is necessary?)

We use bladder tanks in a differnt mode, since "constant pressure" is important. A pump is continuous running from a VSD and the bladder tank absorbs the instantaneous changes in demand. From what you are describing, the customers/downstream process will be receiving the full scale pressure variation. (Like I have on my domestic water supply from tanks.)

Is sizing the tank done in the same manner(calculation) for both pressure boosting water pumps for domestic use(which operates very often) as well as for fire pumps which may not operate for years?.

1 - I don't know exactly how much water flow per hour but it's between 20 & 60 cu.m per hour . this different between night & day hours

2 - pump currently cycling is only 9 second per one cycle it means 400 cycle per hour

3 - the mentioned booster pump used for domestic water to feed a big hospital

4 - exist bladder tanks 2 x 500 liter

Did you read my #5?

1 - it's the pump flow that matters in finding pump start frequency. Max start frequency is when the demand = 1/2 pump flow.

2 - is that 9 sec start-start (to give 400 starts/hour)? If so it's much too high.

4 - as the min/max pressures are 3.5 and 5.5 bara (assuming your figures are barg), if tanks are precharged to 3.5 bara (no water in tank), when pressure rises to 5.5 bara the air volume reduces to 3.5/5.5 = 7/11, so water into tank = 4/11 x tank volume. With bladder tanks 2 x 500 liter, in/out water volume ~ 360 litre. If pump flow = 40 m³/h, max start frequency is ~ 28/hour.

If you're getting 400/hour, there's something wrong. Have you checked the precharge pressure? Too high or too low will increase start frequency.

Above assumes constant pump flow, but in practice this will vary with the pressure, depending on the pump curve. Need to estimate the average flow to get a better answer, but if you use a timer to limit start frequency it's not too important.

to Codemaster

- that 9 second is the actual frequency ( start to start time ), I know it's too much . Also we can't reduce because pump controlled by pressure switch ,not by timer , pressure fast failed due to high consumption.

- our max/min pressure is 4.5/2.5 bar gauge pressure

- average flow ~ 4o Cu.m /hour

- our main problem is water hammer which occurred after each pump stop , accordingly main water pipes break from time to time.

- We consulted a specialist engineer , he asks to replace bladder tanks by 2*1000 lit. instead of 2*500.

- So my question, are bladder tanks is the main reason for our problem . and what formula can be used in such case , that's to assure that water hammer will stop or at least absorbed by bladder

.

You can control by pressure switch and a minimum-run timer. When the pressure switch closes on low pressure, the pump starts and the timer starts. If the pressure switch opens before the timer has expired, the pump continues to run until it does. If the timer has already expired when the pressure switch opens, the pump stops immediately. So if you can stand say 10 starts/hour, the timer is set at 1/10 hour = 6 minutes. Starting frequency then cannot exceed 10/hour. You would have to buy a timer and fit it if there isn't one already.

I still think there is something wrong with your system. Have you checked precharge pressure? Best to drain the pipework, then check it is ~ 2.5 barg. This also checks for broken bladder - if the pressure doesn't build up there's a leak.

If you replace the tanks as per your specialist engineer, according to my calcs you will get max 14 starts/hour. That's worst case, at high and low flow start frequency is less, so it might be OK, depending on the pump. But it would be a lot cheaper to fit a timer and fix your existing tanks.

With a timer, you would probably be OK with 1 x 500 litre tank, if it's properly set up.

- if the pressure rises up from 2.5 to 4.5 within some few seconds , what you expect to happen if the pump continue run for the remaining 6 minutes. pressure will be more and more increase until it reaches for maximum pump pressure which is 12 bar.

- pressure tank was adjusted before for 2.5 bar (air pressure ) while it was empty from water. also bladder is not Punctured ( not leak ).

The timer method only works if it's a centrifugal pump - see my #5. Is it? I wouldn't expect to see a PD pump for this duty. If it is a centrif, 12 bar closed-valve pressure for a working pressure 4 - bar is highly unusual. Can you give more details - motor rating, connection sizes, speed, Q/H curve etc?

I assume you can follow my calcs in #7. If the precharge is correct, can you see any reason for start-start time 9 sec vs calculated > 2 minutes? All I can think of is the flow is much higher than 40 m³/h.

Yes it's multi stages centrifugal pump

- I know that 4 bar working pressure for a 12 bar closed pressure pump is highly unusual and unsuitable selection for pump size. but it's the exist one and it's costly to change. So it rapidly rises up pressure with high flow within few seconds and suddenly stop which creates water hammer due to back flow of water.

- Now I'm looking for how to absorb that hammer while using the same pump,

- I don't have water flowmeter to measure pump flow ,Also when i looked for (H-Q)curve for the pump ,I found that we are out-of curve ( less than min pressure )

Should have asked before - is this a new installation, or has it worked OK in the past but has developed this problem?

If the pump is rated for 40m³/h at about 12bar, it's likely to be delivering far more off the end of the curve. If you double the tank volume as per your specialist engineer you'll still have 200 starts/h, and it won't do much for the water hammer problem. Whatever you do the pump must stop at some point, so the water hammer is a separate problem. Maybe a different type of pump non-return valve would improve it.

Presumably the pump stops quickly when 4.5barg is reached, so doesn't get to 12 bar. As an experiment, you could bypass the pressure switch and let it run for a minute or 2, see what pressure is reached, then stop it manually and check water hammer. Also see how hot the casing becomes. That's if your bladder tanks are OK for ~ 12bar.

You might consider changing to a lower-speed motor, which wouldn't cost the earth. If it's currently 4-pole, going to 6-pole would drop the pressure to ~ 5bar, just about right. But you'd need to check the pump curve to see if you still have enough flow.

Hello khairy Aish - any more developments on this?

HI Codemaster.
Yes I Have some good news.
1 - we reduced motor speed by activating frequency converter which is already available in control panel but we didn't know before about it , this reduced pump discharge and increased cycling frequency to 25 second instead of the previous 9 second , water hammer is noticeably reduced.
2 - We find that the two 500 L pressure tanks are connected to pump line through
a header of 1 1/4 inch which is too small.
3- Now we are in way to add additional two pressure tanks with the same capacity 500L , in same time we will increase tanks header to 3 inch, this to reduce flow speed in header to 6 ft/sec
After we finish I will inform you with the final result.
Thank you for your interest

Thanks khairy Aish. Be good to hear what happens when the changes are made.

Take a look at this link and see if you can use any information from it.

sizing is only important as to the cycling of the water pump as the pressure rises and falls. municiple water has steady pressure of coarse, so a modest size tank should be satisfactory.

If you size it so that the volume in/out of the tank is enough to give start frequency in usual range (about 5 - 15 starts/hour, bit depends on supplier) it's likely to be big. Incidentally, max start frequency occurs when the demand = 1/2 pump flow. E.g. for 5 starts/hour, tank in/out volume comes to 2 m³, total volume ~ 2 x that. To avoid this, it's usual to have a minimum run timer, set at 12 mins for 5 starts/hour. If there is no demand, the pump runs against a closed valve for that time. This is for a centrif, but unlikely to be anything else for flow/head given.

Not sure about tank sizing basis with a timer, but I estimate pump discharge 50mm dia, and from knowledge of booster sets I'd expect a tank something like 300mm dia x 500mm.