Surge tanks and damping 2

[dbday]

Can anyone comment on whether a bladder type surge tank on a water pumping main needs to be evaluated for different temperature conditions.

Say from just above freezing to plus 40 deg C.

The surge analysis reports that I have seen always seem to assume constant temperature, but on large tanks would solar heating not be an issue and cause a change in the damping characteristic of the tank ?

Any comments would be appreciated

 

[bimr]

The purpose of surge tanks is to reduce pressure fluctuations and water hammers.

A different type of tank is a thermal expansion tank that is commonly used to account for the expansion of water with temperature.

Surge tanks are not usually designed to account for the expansion of water with temperature. However, unless you have extremely long lengths of piping and a small surge tank, it should not be a big issue.

 

[dbday]

I was thinking in terms of the air inside the surge tank being heated / cooled by ambient conditions and so changing the air volume from the design point, and if this would have a significant effect, or at least one that should be considered.

I understand that the purpose is to reduce waterhammer and that the system may be likened to a spring / mass system, the air being the spring and the water being the mass.

I would also guess that that the water would stay more or less the same temperature as the main pipe would be being continually supplied from a very large source whose temperature probably would not change much.

 

[BigInch]

The air volume will vary slightly with temperature, but its not a significant amount in relation to the typical water pressure being applied. Using V2 = V1 * P1/P2 * T2 /T1, you can see that if temperatures vary full range from 0 to 100F, using Rankin degrees amounts to only a (560-460)/560 = 20% variation in volume, whereas only a 10 psi variation in pressure from 30 to 40 psig (~20%), using absolute pressures 45 to 55 psia, corresponds to the same 20%+ variation in bladder volume. Typical waterhammer pressures might reach 60 psig on a system normally operating at 40 psig, which would be a variation in bladder volume of almost 40%.




**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[dbday]

Thanks BigInch

I guess the question I have now is;

Will a surge tank still do the right job if the volume is off by 20% ?

 

[BigInch]

Yes, since the water temperature doesn't change during a surge event, there will be no change in bladder volume due to temperature during the surge event.

Let's talk about a surge event when temp is 1ºF, when your bladder volume is at minimum, say you have 40 psig on it from normal operating pressure (bladder inside pressure also = 40 psig) and the pump kicks in increasing pressure rapidly to 60 psig creating a surge event. Initial bladder volume is 1 ft3 and T1 = T2 during the surge event time (probably just a 1/4 second or so).

V2 = 1 ft3 * P1/P2 , 1 ft3 x (40+15)/(60+15) = 0.73 ft3, bladder inside pressure is also now 60 psig.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[dbday]

Thanks BigInch, point taken about double posting.

The surge event that I am concerned about is the unexpected shutdown of the pump.

Taking a fixed speed pump, the tank would be working at a fixed pressure with temperature being the only variable.

As the level in bladder tanks is not controlled, if the tank is initially pressurised on a cold day, when the temperature rises I will get about 12% more volume and vice versa for setting up on a hot day.

Is this significant when I see that the total volumes of some surge tanks can be as big at 150 cubic metres with maybe 50 to 70 actual cubic metres of air.

This is why I was interested in spring rate, amplitude and frequency.

 

[BigInch]

Very very small volume, so any change would more than likely be in the tenths of seconds. It would also have to occur within that time to mitigate the effect of any surge as well. As long as you have some air in the bladder, it should be somewhat effective.

A temperature change may affect bladder air volume during the day or night, but at any given time air pressure would still have to be equal to normal operating water pressure and surge pressures added/subtracted from that, so a 50% increase in pressure at any time, whether the bladder volume was 1 m3 or 10 ml would reduce the volume by apx 1/2 in each case (w/o spring constant response time being considered).

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[dbday]

Dont understand the small volume comment.
150 cubic metres is just under 5300 ft3, maybe a tank 12ft in diameter and over 40 feet long.

The air volume just before surge event could be 2600 ft3, half the physical volume of the tank, so a 10% variation might mean +/- 130 actual ft3 of air.

If the air in the bladder is heated or cooled, its physical volume will change, if the system pressure is constant then surely the bladder will just displace fluid from the tank if it expands, or allow additional liquid in if it contracts.

 

[BigInch]

Opps, my little mind thought m3 was too big a surge volume and converted to cm3.
<reset>
So, is this a rather large flowrate pipeline that has a valve that gets closed fast? Can you give some more details about what kind of system this is diameter, length, and how you expect surges to be generated, pump trip, start, valve close, relief valve opening, etc.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[bimr]

Rather than go back and forth with people guessing the parameters of your pump system, why not give a short summary of what you have? You would get a better answer.

You have a 150 cu. meter water pressure tank.

Is this a municipal water distribution system?

Is the tank in the sun? is the piping in the sun?

What is the tank turnover?

What is the climate if outdoors?

150 cu meters sounds unusually large for a surge tank.

 

[BigInch]

Then again, if its a 1000 psi 24" diameter pipeline, 450 miles long, it could be rather tiny.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[dbday]

Details I have as as follows;

Pipeline is 900mm nominal diameter and 95 kilometers long
Static head at pump is 120m
Flow is 2250 cubic metres per hour, giving a nominal velocity in the main of 1 metre per second.
Liquid is raw water and it is being pumped to a reservoir prior to treatment to turn it into drinking water.
Pipeline material is FRP
Vessel is to be situated outdoors, sun shade is not envisaged at this moment in time.

My concern is that as the system is specified to control to a fixed volume of air in the tank that as the air in the tank heats / cools that the actual mass of air in the tank changes.

Equally if I control the tank with a fixed mass of air in the tank then the actual volume will alter, the head in the tank will vary from the analytical solution.

I am concerned because the analytical solution, as reported to me, shows a downsurge approaching vapour pressure at some points and a return surge nearing the surge pressure limitation of the pipework.

From my days of learning (some time ago) I do rememeber that if you change the spring rate in a spring / mass system, like we have here, then the frequency and amplitude of the resulting vibration change.

My worry revolves around control of the air volume or mass and whether a subtle change in the air volume could cause a problem.

I can't run multiple analyses to solve this issue as I do not have that skill, but I need to have something upon which to hang my hat as I will be asking someone to spend money to do so.

 

[BigInch]

There isn't a spring rate per say. Compression and expansion of the air will be controlled by the water inlet and outlet rate. First because it is the water entering the tank that will compress the air, and the water's flowrate is limited by the tank entrance nozzle connection (Cd coefficient). So, if that's a spring constant, it isn't one in the classical sense, and its nonlinear at best since its doubtful that water entrance and exit rate will be constant for long. Air can't be compressed faster than the air volume in the tank can change by the water volume entering. As air pressure builds, the water inlet rate will decrease until pressures equalize. As the surge pressures in the water decrease, once the surge wave is reflected, the water will begin exiting, as the air pressure in the tank is above the lessened surge pressure. That exit rate of the water is once again limited by the tank discharge nozzle Cd.

What you describe above is exactly what I would expect for a 30" pipeline, although you didn't mention the normal operating pressure at the tank, such a pipeline when surging would be expected to reach or exceed (within code allowables) the normal operating pressure when properly designed. I would guess that the surge tank volume is pretty much correct.

As for what you're asking about the air volumes in the tank, the critical point will be whether the tank is larger than the surge volume of water expected to enter or not. That is what will increase the pressure of the air and consequently the internal pressure of the tank and surrounding piping. If the predicted surge volume is more than expected, the pressures will rise higher than expected. That's what you need to avoid. As long as there is enough air in the tank to accomodate all surge water entering without reaching a pressure higher than the pipe and tank allowable pressure, you're OK. That would indicate that this system should be checked at the coldest operating temperature it is likely to have, since that would also be when you would have the minimum volume of air in the surge tank.

Let me know when you want to take it to detailed analysis.




**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[dbday]

Thanks for the reply.

The solution reported to me is supposed to allow for a 20% margin at full air expansion - ie minimum water in the tank. Assuming that at the hottest condition that we are starting with 10% too much air then I guess we will be ok at full expansion.

As for the scenario with minimum air volume I will ask for this to be examined and let you know the result if it changes the reccomended surge vessel size.

Regarding the inflow and outflow, the tank and branch connection have to be designed in alignment with the analytical solution as far as resistances and inlet / outlet coefficients are concerned - we have been "told" what we are to provide.

Again if these change as a result of the extra work on the air volume I will let you know.

Thanks for the discussion.

 

[BigInch]

Right, a larger diameter inlet/outlet will increase the rate at which volume can be transferred from pipe to tank, supposedly resulting in less of a buildup of pressure in the pipeline. The size of that connection is quite important.
----------------
You're welcome.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[bimr]

Hydropneumatic tanks have been used to some extent to control water hammer in small and medium-size pump stations. The tank is connected to the main header and the pumps are equipped with swing check valves, preferably with dashpots. On pump shutdown or power failure, the tank supplies water to the main to limit the downsurge in pressure and prevent water-column separation. After the check valves close the return-pressure wave forces the water to flow into the tank, where it is effectively cushioned.

The suppliers of these tanks use computer programs for proper sizing of the tanks and other components.

Obviously you are going to have to maintain the pressure in the hydropneumatic tank using an air compressor to pressurize the opposite side of the bladder. In which case, the pressure in the hydropneumatic tank would be set to remain as the same operating pressure as the pipeline.

If you are concerned with the expansion of the water in the pipeline due to temperature, you should calculate the expected thermal expansion of water and the corresponding increase in water pressure the pipeline.

One would not normally expect a significant increase in pressure with a water pipeline as water pipelines are commonly buried and not exposed to the sun.

However, should you calculate a significant increase in water pressure in the system from temperature expansion, then consider the use of a thermal relief valve on the pipeline.

 

[stanier]

Suggest you consider non slam type check valves rather than swing or tilt check with dashpots.

http://www.noreva.de

In these days of diminishing maintenance resources is it hard to expect anyone filling those dashpots to make sure they work, let alone ensuring the crap is kept from them.

Bladder tpe accumulators generally use nitrogen with a Cp/Cv ratio more like 1.4 than using compressed air. this has an impact on the behaviour of the system.

I do not understand the comment that you cannot do more analysis. Ask the waterhammer analyst to do more runs and see what the outcome will be. It is far less risky than guessing and easier on the stress level than wondering.

Sensitivity analysis under all conditions with varying orifice diameters and volumes will give you a better idea.

 

[dbday]

Stanier - Its not that I can't have more runs done, its just that it has to be paid for and as I have already been given "the solution" I needed something to back up my questions.

Thanks for the advice on the valves, as you say maintenance schedules are for the file.....