Jump to content



Featured Articles

Check out the latest featured articles.

File Library

Check out the latest downloads available in the File Library.

New Article

Product Viscosity vs. Shear

Featured File

Vertical Tank Selection

New Blog Entry

Low Flow in Pipes- posted in Ankur's blog

Gravity Flow


This topic has been archived. This means that you cannot reply to this topic.
7 replies to this topic
Share this topic:
| More

#1 kensheeran

kensheeran

    Brand New Member

  • Members
  • 3 posts

Posted 27 January 2015 - 06:54 PM

I am troubleshooting a water treatment skid.

 

This forum was very useful for confirming the Froude number design guide and source. Thank you very much for previous contributions.

 

Briefly, the skid discharges 15-30m3/h water into a 100mm pipe at 8m elevation. There is no vent in the line. The discharge pipe runs down 4m, across 15m, down 3m, across 2.5m, back up 1m, then across, down etc. (there are about 20 elbows) into an underground tank sump (open to atm). Discharge in the tank is a dip pipe 2.5m below grade, submerged and no holes are drilled into dip pipe. Total pipe length approx. 60m.

The U in the pipe (about 1 meter) is for analysers which have to maintain wet.

 

Yes, the discharge point (inlet of pipe), runs at vacuum (no surprise).

Client needs to acquire a water sample and cannot.

 

I have two questions below:

 

Hill paper suggests Froude number has to be less than 0.3. Chart suggests an 8 inch pipe for free flow.

I have a 170mm ope in the floor.  A 6 inch sched 10 has OD 168mm, ID 162mm yielding Fr 0.32-odd. There is equipment below, adjacent pipes in the floor ope and it is preferable not to have to re-core the floor.

Q1 - how strict is the Froude criterion - will exceeding it by a small margin (even if only for the 0.3meters thickness of the slab + 100mm above and below) make a large difference?

 

My thought is to replace only the pipe section before the analysers with a larger pipe.

The Hill paper and advice on this forum in previous threads suggest that such a tortuous path can be problematic to achieve free draining.

 

Q2 - Will the below pipe route option 1 work or should I use the alternate route option 2?

Route option 1 - horizontal discharge (pipe) drops vertically 2.5m through an ope in first floor, runs horizontal, changes direction 3 times (one 90, two 45 elbows), drops vertically 0.5m, runs horiz again changing direction 3 times (90 deg elbows) and then drops down vertically 3m to the analysers.

The pipe is PVC - I am not sure if standard elbow bend radius r/D is adequate (Hill suggested 5). I can make the 90 degree elbows 2 x 45 degree to control bend. I can also slope the horizontal sections (but that will require extra turns).

 

My route option 2 is to run across at the high level immediately after discharging from the skid before dropping down 'close to' the analysers. Admittedly, this second option also makes the first question moot as a new ope in the floor can be made for 8 inch (it's just money). All the same, I'd be interested in the answer to Q 1.

 

I have no other route options due to clashes, access ways etc. I need the cheapest workable solution, not the best solution. Option 1 is cheaper.

 

Cheers

 

 

 



#2 Bobby Strain

Bobby Strain

    Gold Member

  • Members
  • 3,529 posts

Posted 27 January 2015 - 11:56 PM

Simply calculate the pressure along the various sections starting at the end. You need some accuracy, not guides.

 

Bobby



#3 katmar

katmar

    Gold Member

  • ChE Plus Subscriber
  • 663 posts

Posted 28 January 2015 - 03:07 AM

You have not defined what your problem is, so it is difficult to give specific advice.  All you have said is that the client cannot get a water sample but you have not explained why not.  It is also very difficult to form a mental image of your installation with all the ups, acrosses and downs - a sketch would be much better.

 

An important aspect of designing a pipeline for self venting gravity flow is that only the first vertical downflow section should be designed to be self venting, and even then it only has to be large enough to be self venting from the inlet down to  a point where it is always flooded.  You only want self venting flow where you KNOW you have two phase flow. 

 

For any section of pipe thereafter you actually want to design (if possible) for Fr > 0.65 to ensure that any bubbles that somehow do make it into the pipeline are carried out of the pipe by the water.  If bubbles are getting into the pipe (or even the air in the pipe at start up) and the water velocity is insufficient to flush the air from the system you will get vapor locks at your high points.  Google "slack flow" for more info on this. Problems of this nature often result in cyclical flows and level changes (as described by Hill)  If the velocity is too low to flush the air then you need to have air vents at every high point. Problems with gravity drains are often solved by making sections of the pipe smaller rather than larger. 

 

If you cannot get the velocities correct - as often is the case with variable flows - then you need to look very carefully at slopes, high points and air vents.

 

With regard to your choice of routes - my general experience is that it is better to start with a vertical downleg rather than a long horizontal section.  Two phase flow is problematic at the best of times, but always worse in horizontal pipes.  It is better to quickly get down to a level where you can ensure the line is flooded with liquid.

 

 



#4 kensheeran

kensheeran

    Brand New Member

  • Members
  • 3 posts

Posted 28 January 2015 - 07:48 AM

Bobby

I ran the calc to get Fr of 0.32. Will this self-vent at 6" sched 10 (id 162mm)?

 

Katmar

I can't have entrained air as it will affect the Flow transmitter.

I need to ensure the analysers are always wet and do not syphon out.

Does the large pipe need to go through the floor down to the analyser level?

Attached is a slightly simplified diagram.

 

any advice most appreciated.

Attached File  img-128125652-0001.pdf   226.25KB   51 downloads



#5 katmar

katmar

    Gold Member

  • ChE Plus Subscriber
  • 663 posts

Posted 28 January 2015 - 11:19 AM

Is the outlet from the skid unit pressurized, or is it an overflow? If it is pressurized why do you need vents and air disengaging sections - I would not expect any air to be present. If the outlet is an overflow, how does it flow uphill to the 8" section? You said that the inlet was under vacuum in your first post. You need to explain the system more clearly for us to understand what you are trying to do. From where is the client trying to get a sample but is unable to?



#6 kensheeran

kensheeran

    Brand New Member

  • Members
  • 3 posts

Posted 29 January 2015 - 04:28 AM

Katmar

 

Thanks for the responses - I have resolved the design problem and realise that my sketch (which was an attempt to simplify the pipe routing) will not work because it will entrain air that will affect the flowmeter. I have reverted to a design that 'drops quickly to the right level' as you suggested in the first response, and after reviewing Fig 2c in the Hill paper.

 

However, my more general question remains un-answered concerning the transition from self vented flow to entrained flow as the volumetric flux increases. How strict is the Fr 0.3 criterion for vertical flow? Normally engineering guides allow some deviation or display an 'intermediate region'. Will a flow with Froude number slightly above 0.3 (in this case , 0.32) exhibit vented flow, or is the transition to entrained flow quite abrupt?

Is there any data published on the transition and does pipe size have any influence.

Will the pipe size have any influence (in this case we are discussing and intermediate pipe size of 6").I ask about pipe size because the Hill paper suggests a more lenient sizing for large pipe sizes in horiz flow.

 

FYI - I don't plan on using the 6 inch pipe - I am just asking this question to understand this phenomenon a little better.

 

Cheers



#7 katmar

katmar

    Gold Member

  • ChE Plus Subscriber
  • 663 posts

Posted 29 January 2015 - 03:04 PM

The only experimental data that I have seen reported was in the article by Larry Simpson in Chemical Engineering, June 17, 1968.  This indicated that for Fr < 0.31 no air would be entrained.  For 0.31 < Fr < 1.0 Simpson reported pulsation and vibration.  For Fr > 1.0 a syphon forms rapidly and all gas is entrained with the liquid and flushed from the pipe.  Between 0.31 and 1.0 the small bubbles will be entrained while the larger bubbles will be released and vented and depending on the quantity of air a syphon may eventually form, but the time will be variable.  For design purposes I do not use Fr > 0.31.  In fact the common practice is to size for Fr = 0.3 and then select the next larger standard size pipe.  For example, if your calculation indicated that you needed a 7.2 inch pipe you would select an 8" pipe with an ID of 7.98 inch.

 

Whether you can accept a Fr > 0.31 depends on your process.  You will get some air being carried down with the water and your decision will depend on what impact that air has on your process.



#8 PingPong

PingPong

    Gold Member

  • Members
  • 1,446 posts

Posted 30 January 2015 - 04:39 PM

Briefly, the skid discharges 15-30 m3/h water into a 100mm pipe at 8m elevation.
How do you know that the flowrate is 15-30 m3/h? What determines that flowrate?

 

If the present flowrate is a result of the resistance of the present 100 mm pipe, then increasing the pipesize will increase the flowrate.






Similar Topics