Pipe riser emptying rate

[JohnWeal]

Assume pump discharge is at zero metres elevation.
horizontal discharge for 8m length then turns vertical for 18m to an open-ended free discharge.
(Diameter from pump is 700mm then imediately increased to an 800mm pipeline)
There is no non return valve in th line or isolation valves. (previous pump installation for this sewage duty found it beneficial to have the flow come back and back spin the cetrifugal pump which help clear rags etc)
Upstream of the pump assume 1m above pump centre-line which gives a 17m static head. (The upstream head will not vary due to backflow as the volume is large within the sewer)

I need to check that the pipe forces produced due to the backflow (pump trip perhaps being the worst case rather than controlled turn-down) when acting upon the reducer immediately fitted to the pump discharge flange. (and hence the load on the pipe support to prevent loads being transmitted to the pump)


Any advice on this type of calculation?

I was thinking Bernoulli's equation but think as the riser starts to empty and the flow returns to the channel upstream of the pump, the head is reducing but also the velocity is higher as momentum of the fluid gathers.



Regards
John

 

[bimr]

Yes, don't do the calculation. It is not good practice to spin a centrifugal pump backward as it may damage the pump.

Suggest you install a check valve to prevent the backflow from ever occuring.

 

[JohnWeal]

Thanks for the reply bimr.

Yes I wonder why this was accepted on the previous set up. The pumps were Weir 'Swallowglides' and now installing Hidrostal with a 'screw type' impeller which is better for sewage with larger clearances.
I still need to question Hidrostal on the effects of backflow even though the impeller is slightly different from the multi-vane impellers we had on the Weir pumps.


Regards
John

 

[Artisi]

Run-back is unlikely to ever cause any "pump" damage but in some cases can result in mech. seal problems, this you need to check. However you need to ensure that a pump running in reverse can not be asked to re-start.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[JohnWeal]

To give a touch more info on the discharge riser, the 8m horizontal section includes a flow meter (magflo) and a 45 degree bend.
Bearing mind the pipework is 800mm with a 700/800 reducer immediately on the pump flange, (both pump flanges are 700mm) it becomes evident that the number of pipe diameters prior to the magflo is not in accordance with accurate flow measuring tolerances.

The existing setup is not how we would have done things from scratch and with a 35m diameter drywell constructed in the late 80's (of which the 8 pumps are located 6m from the centre - radially), we have no room whatsoever to install a check valve without further disrupting the flowmeter accuracy.

Personally, I would have put the magflo in the vertical riser but this is something that also introduces an increased cost.

Perhaps it may be a necessity if the reverse operation of the pumps is likely to affect its long term operation costs.

We are already 25% over budget but I hate to do things when I know there is a better engineering solution!! I'm sure we can all say that on many projects we have all worked on....

Regards
John

 

[MortenA]

@bimr: Why would it damage the pump? (im serious because im just a chemical engineer. I assume you are not worried about teh pump itself but maybe the coupling, gear or motor?
Besr regards Morten

PS: I read Artisis comments and tend to agree with those but then again im just a chem eng.

 

[bimr]

It is good practice to avoid running a pump in reverse direction. Pump shafts have been broken this way especially if the pump is started while running backwards. The simplest solution is to install a check valve on the discharge line.

The reason is that when a pump starts, there is torque to get the fluid moving and then you have to add the torque caused by the reversing fluid. This may also cause prolonged startup with high motor currents.

There may also be situations where you are pumping against extremely high dishcarge pressures or have accesory items such as oil lubrication pumps that cause additional concerns.

 

[Artisi]

John,

"Yes I wonder why this was accepted on the previous set up" I read this as NO check valves on the Weir pumps - correct?
Again, I see no problem with reverse rotation but why not check with Hidrostal to make sure.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[bimr]

It will be difficult to calculate or predict the reverse flow through a pump from gravity head. However, the reverse flow (and pipe loading forces) should be less than the normal forward flow through the piping that the system has been experiencing.

Is this a CSO application?