Flow rate from gravity water supply 3

[highwater]

Dear All,

I work in Scotland in private water supplies. We've a new project and I' d appreciate some help.

We've located a spring source some 1000m away from the clients property. It lies about 30m above the existing, large capacity, holding tank. The intervening ground is undulating.

We want to lay in an overland pipe from source to tank and want to achieve a feed of at least 100 litres / 20 gallons per hour to the tank.

The client is on a tight budget and we're tempted to use the smallest available PE pipe (15mm internal dia). This size is also easiest to roll out over the ground!!

I'm no mathemetician so have got bogged down with Darcy Weisbach etc! It also seems that to work, these equations assume you know the flow rate and or the applied pressure. In my case I want to know flow rate and have no input pressure!(other than the "vaccum" /syphon effect of my net 30 metre fall.

This work is quite common for us (perhaps we should know the answer to this one already!!)but it would be useful if there was a simplish formula we could use when presented with a variant of the the above in future.

Note: we have a very good fall (20m over 200m) immediately on exiting the spring towards a river - before climbing again on the other side, then dropping again steadily to the distant holding tank.

 

[25362]

From an old Hydraulic Institute Pipe Friction Manual, for a steel pipe with ID=0.622 inches, running full with 0.7 (US)GPM of clean water, and a surface rugosity [ε]/D=0.00289, a friction drop = 0.74 ft per 100 ft of pipe equivalent length, should be expected. Could this be of help ?

 

[katmar]

highwater,

Plugging your data into my software gives a flowrate of 250 litre/hour if the difference in height is 30 m, and 200 liter/hour for a 20 m difference. I have assumed an inside diameter of 15 mm and a pipe roughnes on 0.2 mm. You pipe may actually be smoother than this but flexible pipe is often less smooth than rigid sections.

The undulations are not important, provided that they never go higher than the spring of course! I have used a length of 1100 m to allow for these undulations.

Theoretically you could use an 11 mm ID pipe to get 100 litre/hour, but there are likely to be kinks etc in the pipe that will give you less flow than that calculated and I would stick with the 15 mm pipe even if there is an 11 or 12 mm pipe available.

You are correct that the use of Darcy-Weisbach requires that you know the flowrate in advance. This makes a trial-and-error solution necessary. That's what computers are for!



Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[chicopee]

Is it going to be a straight line pipe layout? How many feet of pipe are you anticipating to cover the 1 km distance? How many elbows (90 & 45 degree types)?, how many valves? and is the storage tank head at the same elevation as the water supply head?

 

[highwater]

F.a.o Chicopee

Thanks for your interest.

Storage tank is approx 30m lower than source, and 1000m distant. Pipe run is largely direct made from 100m lengths of 20mm (external) MDPE connected using straight couplers. The 1000m above allows for the undulations. The only valve fitted will be to slow the flow (if required).

I do not need exact values as long as I exceed the approximate minimum requirement of 20 gallons (imperial / british) or 100 litres per hour. Source yield is in excess of 1000 litres per hour.

Hope this helps and thanks again.

highwater

 

[BigInch]

Scotty,

You don't have to be a math wiz anymore, but I do recommend you have some idea of what your doing. See this link,

http://www.processassociates.com/process/fluid/dp_1.htm

I entered Flowrate:
select volume basis = 20 gph (US gallons)
Fluid's Physical properties:
Density = 62.4 lbs/ft3
Viscosity = 1 cP
Piping:
Length = 1 Km
Diameter select Inside /equivalent = 15 mm
Condition:
Select Fouled, absolute roughness = 0.005 mm

Report:
Select your pressure drop units and Head loss units
kPa and m
[COLOR=black yellow]CALCULATE[/color]

Read the report

Results
Pressure Drop 16.925 kPa
Head Loss 1.727 m (Fluid)
Reynolds Number 1784.3 Laminar Flow
Fanning Friction Factor 0.0089672
Velocity 0.119 m/s
Percent of Erosion Velocity 3.1
Input
Flowrate (Volume Basis) 20 US gal/h
Density 62.4 lb/ft³
Viscosity 1 cP
Pipe Length 1 km
Pipe Inside/Equivalent diameter 15 mm
Pipe Condition Fouled
Absolute Roughness .005 mm

I wouldn't really recommend building a pipeline 1 km long using 15mm diameter pipe. I would use a minimum of 2in, but I understand costs sometimes control the work. Just expect to have some problems keeping it clean. At least its plastic pipe, so you can easily make cutouts to snake it if you have to.

The problem there is that its flow rate will be low for keeping the inside clean. It may be better to run a much faster flow and find some productive way to use the water arriving at the tank.

You will need a good valve with some pressure or flow control capability, [balls not especially good at that), because you will have more pressure than you will need if you chose to run at 20 gph.

If you lay this pipe on the ground, how will you winterize it? Don't you expect it will freeze?
Will you be able to drain it for the winter?

If you lay plastic pipe on the surface, I doubt it will last very long, even if it is UV stabilized, somebody just might decide to borrow some of it.

If it is to be a surface run, leave some "snake" in it for temperature expansion and contraction allowances.




Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[highwater]

F.a.o. Big Inch

Thanks for your valued input. I take your various points.

The area is a remote open hillside,owned by client."Borrowing" of the pipeline is possible but unlikely. Supply is a backup for an existing spring. We're experiencing a "relative" drought here and the existing supply is drying up.

I imagine that come autumn all will be well again so this is likely to be a temporary short-term fix, possibly used each summer only. Freezing will thus only be an issue insofar as it damages the pipe if left full of water. (dont know whether under full, unrestricted, flow conditions the 8 deg C source temperature will be sufficient to prevent freezing in the exposed pipe (sounds like another thread!!). If not, draining would be straightforward.

It may be that the client will ultimately develop this source into a permanent solution in which case a larger diameter buried pipe is probably the answer.

Another tempting benefit of the reduced pipe size is that there should be no danger of draining the source and airlocks.

Thanks again

 

[chicopee]

Applying Bernouilli's equationand making certain assumption such as pipe roughness of .000005 leading to an about friction factor of .015, gate valve K value of 4.5 (1/2 opened), rounded pipe entry and exit K values of .04,height difference of 98.43' and flow rate of 20 gph, pipe dia is between 3/8" and 3/4", therefore , 15mm ID is acceptable. Note such calculations are for ideal condition which is much unlike actual conditions.

By the way continuously running water should not freeze which I can attest when I went to college in Montana and winters were in the -30 to -50 deg. F range.

 

[BigInch]

Strange, 1" pipes freeze in Texas. A few leaves or a small rock, sand accumulation at a low spot, tank freezing up can all stop a 15 mm pipe flowing dead cold in its tracks.

Right. I recommended a larger diameter for a number of reasons, non-ideality being only one, but if he thinks 15 mm will work, not much I can do about it.

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[djack77494]

I would also recommend a larger diameter pipe for the same reasons you've heard. In addition, I would try very hard to avoid any low spots between your source and destination. They will fill with silt/sand/etc. and lead to problems. My preference would be for a continuous slope of (say) 1/4" per foot or better. Understanding reality, I would accept pipe segments that had no slope at all, but I would be extremely adverse to any uphill flow.

Another point I would make concerns your sizing calcs. You are dealing with a gravity flow system. Your pipe may very well be only partially filled. Thus, many of the typical references used in "normal" hydraulics calculations will NOT apply. Educate yourself in how to calculate gravity flow hydraulics before proceeding.
Doug

 

[BronYrAur]

Just to drive the point home, a 15mm pipe is only a little larger than a 1/2" pipe for those of us who don't use SI units. Just thinking about a 1Km long line with only 30m of static head to provide flow is concerning. You will definitely want a very large collection apparatus consisting of a large area of very fine strainer or mesh to keep all sediment out of the pipe. This will probably require routine maintenance to keep the strainers clear of debris. I would encourage you to use the largest diameter line possible. I would also encourage that you have the means available to flush the line with pressurized water in case a blockage occurs.

 

[BigInch]

Ya. Its a half-mile garden hose.

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[katmar]

Wow! Is highwater building a water supply to a nuclear reactor, or a temporary backup for domestic water??? You don't need to apply high-end industrial standards for a simple application like this. A simple sand-trap and an inline strainer at the start of the line, and a union at each low point for draining, will keep it running reliably enough for the intended duty. There is a storage tank at the downstream end that would probably keep the house supplied with water for a good few days while a bit of cleaning is done.

If the entrance to the pipeline is flooded, and the throttling valve is at the downstream end, the pipe will always be flooded and Darcy-Weisbach or Hazen-Williams analysis is correctly applicable.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[BigInch]

Katmar, take it easy. Is there a problem? I was just showing him he could find the head loss all by himself, which is probably a good thing since he's "in the business" and now won't have to ask our advice all the time now.

IMO no matter what the application, its still a gravity feed garden hose a half mile long, and I don't like to see anybody learn the hard way, no matter if its not a nuc. Theoretically yes it works and Darcy-Weisbach gives a very conservative answer, if it does indeed give a 30 m head loss as you say, albiet DW should only be used for large diameter pipelines with water. For example, if I use 250 lph and the Process Associates Hl online calculator, I get 19.5 m lost, not 30, which agrees with my personal spreadsheets using Colebrook and Churchill. And OK, maybe I just have this inate adversion to using a garden hose, given my large [Ø] pipeline experience, but again no problem, its his garden hose. I get impatient when I have to wait for the water to come out of my 30 ft hose.

Sand filters or strainers clog up and then... freeze.

PS. You don't have to assume 250 l/h; he gave the flowrate at 20 gph, although he didn't say if they were US gallons or Imperial.

Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[highwater]

It seems I've stirred things up quite a bit!!!

Thanks for all your help and occassional doom laden portent. The good news is It works! Only connected today so it's early days for confidence.

20mm (15mm internal) MDPE, "only" 850m from source to tank. We got a big low spot (-30m) at 150m from source, then steeply back up about 20m then climbs for a further 10m over 250m or so then it's all downhill from there to holding tank.

Spring catchment chamber is gravel packed inside and out to hold back silt. Dynamic level in chamber is some 100mm above coarse outlet strainer, so entrained air is unlikely.

We've installed a "T" and scour valve on lowest point.

Pipeline self-started, and was still purging air when we left tonight. Flow rate not measured but estimated at over 50 gallons us or imp per hour. May yet rise as air continues to purge. We made need to valve this with a restrictor of ball float as excess overflow may be problematic.

A happy client and a happy highwater, thanks for all your help and interest.

I'll let you know if the system stalls / silts / freezes / or otherwise dies.

 

[katmar]

highwater - thanks for the feedback. It is always satisfying to know when a problem is solved.

From your latest description it sounds as though the highest point is very close to the height of the source. It may take quite a while to syphon all the air out, but it seems you are getting the required flowrate anyway.

BigInch - bigger is not always better (and I am not having a go at your name!). A 50 mm pipe would cost roughly 4 times the price of the 15 mm pipe and would be much more difficult to install. It would have a cross sectional area of 11 times the 15 mm pipe and therefore a velocity of 1/11th if it was throttled to the same flowrate. This would encourage silting out. Horses for courses.

I am absolutely with you that it is better to give a reference that enables the OP to solve the problem by himself. It's the old proverb of "better to teach a man to fish..." I was too lazy and just gave the answer.

I did not assume 250 l/h. It was calculated from the assumptions of available head and pipe size. Your calculation of only 19.5 m head required is correct for normal roughness assumptions - I had assumed a rougher pipe to give a bit of a safety factor for kinking, joints, Murphy etc.

Peace
Harvey

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[BigInch]

Harvy, Really. No worries. I guess they don't call me Big Inch for nothing.

Highwater. Good Work. We all wish you well.



Going the Big Inch!

http://virtualpipeline.spaces.msn.com

 

[ggrindle]

I do not believe this system will work.

You say you undulating hills.

A siphon cannot be sustained beyond a certain point.

Atmospheric pressure is 14.7 lbs/in2. Convert that to feet of head and its about 33 feet. If any of your undulations are over say 30 feet you can foret about a siphon working.

Now if the spring is higher than any of the hills over which this piping must climb or is no more than 30' lower then the positive pressure pushing the water will make this so that it is not really a siphon, but a positive pressure system pushing the water over the hills.

I don't believe you've mentioned if any of the hills were higher than the spring, however, springs are not usually located at the tops of hills! So I suspect you will be going over some hills higher than the spring.

Good luck.

GG