Previous in Forum: Selection criteria between pipe & tube   Next in Forum: Pneumatic Wizardry!
Close
Close
Close
10 comments
Rate Comments: Nested
Participant

Join Date: May 2008
Posts: 3

Selecting Pipe Size

05/09/2008 9:52 AM

For a given pressure and a desired flow rate, how can you determine the mininum diameter pipe to allow for desired flow?

Register to Reply
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.

Good Answers:

These comments received enough positive votes to make them "good answers".

"Almost" Good Answers:

Check out these comments that don't yet have enough votes to be "official" good answers and, if you agree with them, vote them!
Guru
Popular Science - Biology - New Member Hobbies - Musician - New Member APIX Pilot Plant Design Project - Member - New Member Hobbies - CNC - New Member Fans of Old Computers - ZX-81 - New Member

Join Date: Jan 2007
Location: Centurion, South Africa
Posts: 3921
Good Answers: 97
#1

Re: Selecting Pipe Size

05/09/2008 11:06 AM

A lot of factors will influence the choice of pipe diameter. (Source of pressure, distance, elevation difference, type and class of pipe etc)

Assuming you have municipal water at a almost constant pressure, (The use by other people in the are will affect the pressure).

Then you must know what pressure is needed at the end of the pipe. This pressure required must firs be deducted from the original pressure to get the maximum allowable total friction value.

Assuming that you don't have access to the true ID, Reynolds number etc of the pipes the best would be to obtain friction curves from the supplier. (or his supplier).

This is what you should do

PLossmax = Pavailable - Pend

max friction = PLmax * 100 / (distance pipe + equivalent length of fittings) == max allowable friction rate per 100 unit lengths.

Convert to units on friction graph.

Any pipe size equal or bigger than the crossing point of Q and F can be used,

A bigger size would be better because it will allow for fictions in supply pressure.

If this is not what you are looking for please supply more quantified information.

The pipe type preference would help.

__________________
Never do today what you can put of until tomorrow - Student motto
Register to Reply
Participant

Join Date: May 2008
Posts: 3
#2

Re: Selecting Pipe Size

05/09/2008 11:35 AM

I should have given more information to eliminate assumptions. This is in a papermill and the water is more of a paper slurry/water mix. The pressure is from a pump down stream with another up stream (both discharge and suction pressures present). The pressure present is 50 psi. The desired flow between two 24" headers is currently 14,000 gpm and I am going to connect the two headers with as small a pipe (about a foot apart) as possible (Economy reasons) to bypass some cleaners. There is also a gate valve between the two headers. According to Cameron's friction charts, I was thinking somewhere in the neighborhood of 18" diameter pipe should work, all pipe is stainless steel schedule 40. I was, however, wondering if there was a formula out there that would answer these types of questions.

Register to Reply
Guru
Hobbies - RC Aircraft - New Member Hobbies - Automotive Performance - New Member Hobbies - DIY Welding - New Member

Join Date: Jul 2007
Location: Fort Lauderdale Florida
Posts: 5708
Good Answers: 123
#4
In reply to #2

Re: Selecting Pipe Size

05/10/2008 12:31 AM

The formula for flow rate I find is;

GPM=29.71 times diameter squared times square root of nozzle pressure. Hope this helps.

__________________
Bob
Register to Reply
Anonymous Poster
#3

Re: Selecting Pipe Size

05/09/2008 2:05 PM

visit hydraulicsupermarket.com there you can get lot of infirmation

Register to Reply
Guru
Engineering Fields - Piping Design Engineering - New Member Egypt - Member - Member since 02/18/2007

Join Date: Feb 2007
Location: Cairo, Egypt
Posts: 1733
Good Answers: 248
#5

Re: Selecting Pipe Size

05/10/2008 3:19 AM

From handbook "Applied Process Design for Chemical & Petrochemical Plants" Chapter 2- Fluid Flow :Quote Flow can be determined for a fixed head system by equation: GPM liquid = 19.65 d2 (hL/ K)1/2 Unquote

Therefore, you have to be aware with the conditions of applying of each equation, i.e. to be familiar with loss of pressure in a pipe line which is caused by:

1. Resisting friction of the interior surface of the pipe.

2. Internal resistance of one particle or layer of the liquid sliding over another, known as viscosity .

3. Specific gravity or density.

4. Special losses:

a , Entrance to the pipe.

b. Sudden enlargement or contraction of the cross section area. i.e., change in diameter of the pipe.

c. Obstructions due to fittings and valves, sharp bends meters, scrapers, or sediment in the line.

So, to compute the rate of flow through a pipe line the following data must be known or assumed:

1. Inside diameter of the pipe

2. Length of the pipe

3. Difference in elevation

4. Operating pressure

5. Viscosity of the liquid

6. Gravity of the liquid

7. Roughness of the inside of the pipe

__________________
It is better to be defeated on principles, than to win on lies!
Register to Reply Score 1 for Good Answer
2
Guru
Engineering Fields - Piping Design Engineering - New Member Egypt - Member - Member since 02/18/2007

Join Date: Feb 2007
Location: Cairo, Egypt
Posts: 1733
Good Answers: 248
#6
In reply to #5

Re: Selecting Pipe Size

05/10/2008 6:31 AM

The following is an example explaining all the hydraulic parameters and how to use these parameters to calculate the optimum pipe size for certain liquid flow.

Example: Determine pipe diameter to allow 2000 GPM of water to flow between two atmospheric storage tanks, max. allowable velocity is 12 ft/sec. Piping system consists of 60 ft pipe length, 6 elbows LR 90o and one globe valve. The flow is by gravity from a the bottom of the 1st storage tank to the open top of the 2nd storage tank with an elevation of 8 ft.

Solution: Pipe cross-sectional area (min.), A = Q/V = (2000 x 0.133681 ft3 )(1/60 sec)/(12 ft/sec) = 0.37 ft2 = 0.37x122 in.2 = 53.48 in2.

A = (π/4)d2 and dmin. = √(4A/π) = √(4 (53.48 in.2/ π) = 8.25 in.

NPS 8" Sch. 40, ID = 7.981" < 8.25, therefore that size is not permitted because it will increase the velocity. So we have to use the larger NPS 10" Sch. 40, ID = 10.02" > 8.25 O.K

Vnew = Q/A = [(2000x0.133681/60) ft3/sec] / [(π/4)(10.2/12)2 ft2] = 8.138 ft/sec

Energy Balance: (P1/W) + (V12/2g) + Z1 + HP = (P2/W) + (V22/2g) + Z2 + ∑HL

P1 = Patmosphere = 0 ............. V1 = 0............................. Z1 = 8'

P2 = Patmosphere= 0 ................. V2 = ?..............................Z2 = 0

HP = 0……………………….HL = ?

HL = Z1 - (V22/2g) ……………(1), where HL = f(L/d)(V22/2g) + K (V22/2g) …………….(2)

From equations 1 & 2: f(L/d)(V22/2g) + K (V22/2g) = Z1 - (V22/2g)

(V22/2g)[ f(L/d)+ K+1] = Z1 ........and V2 or V = √[(2g Z1) / (f L/d +K +1)]

• Entrance coefficient, K = 0.5

• L/d : For 60 ft of pipe L/d = 60/(10/12) = 72

For 6 LR elbows, 90O, L/d = 6 x 20 = 120 & For globe valve, L/d =340

Total L/d = 72+120+340 = 532

To calculate friction coefficient, f :

Re = V D / Υ = (8.14)(10/12)/(1.4 x 10-5) = 4.86 x 10-5 ……….(Flow is turbulent)

And ε/d = [0.0005 / (10/12)] = 0.0006

From Moody graph, f = 0.018 & f (L/ d) = (0.018)(532) = 9.575

V = √[(2g Z1) / (f L/d +K +1)] = √[(2x32.2)(8) / (9.575 +0.5 +1)] = 6.82 ft/sec.

Due to friction loss, the velocity decreased, 6.82 < 8.14 Not Good where 8.14 ft/sec = min. flow velocity in 10" pipe for 2000 GPM.

So, try NPS 12" sch. 40 pipe, d = 11.938"

Vmin = Q/A = [(2000x0.133681/60) ft3/sec] / [(π/4)(11.9382/12)2 ft2] = 5.73 ft/sec

Re = V D / Υ = (5.73)(11.938/12)/(1.4 x 10-5) = 4.07 x 10-5

and ε /d = [0.0005 / (11.938/12)] = 0.0005

• From Moody graph, f = 0.0175

• L/ d = 60/ (11.938/12) + 120 + 340 = 520 & f (L/ d) = (0.0175)(520) = 9.1

V = √[(2g Z1) / (f L/d +K +1)] = √[(2x32.2)(8) / (9.1+0.5 +1)] = 6.97 ft/sec.

(and 6.97 > 5.73 is OK, where 5.73 ft/sec = min. flow velocity in 12" pipe for 2000 GPM)

Check Re & f for new velocity:

Re = V D / γ = (6.97)(11.938/12)/(1.4 x 10-5) = 5.17 x 10-5

ε/d = [0.0005 / (11.938/12)] = 0.0005

From graph, f = 0.0175 & f (L/ d) = (0.0175)(520) = 9.1

V = √[(2g Z1) / (f L/d +K +1)] = √[(2x32.2)(8) / (9.1+0.5 +1)] = 6.97 ft/sec. (Solution has converged)

Actual flow : Q = V A = (6.97 ft/sec.) (p/4)(11.983/12)2 ft2 = 325.0247 ft3/sec. = 325.0247/0.133681 = 2431 GPM > 2000 GPM OK

So we recommend using of NPS 12" sch. 40 pipe, d = 11.938", where the velocity is V = 6.97 ft/sec and Q = 2431 GPM

Important Note : One could install a pump into the system and reduce pipe size to meet the requirement, in this case the head of pump HP shall be added into the equation in which the velocity shall become : V = √{[2g (Z1+HP)] / [f (L/d) + K+1]}

Notes: 1. There is a recommended velocity (or velocity range) per each fluid (liquid or gas) to eliminate the friction loss and minimizing the erosion. 2. American gallon = 0.133681 ft3

Exercise : What will be happen if the 1st tank is not atmospheric (P1 ≠ 0) and has an internal pressure P1?

__________________
It is better to be defeated on principles, than to win on lies!
Register to Reply Good Answer (Score 2)
Anonymous Poster
#9
In reply to #6

Re: Selecting Pipe Size

03/29/2009 9:17 PM

Free book about this at http://www.drbratland.com/

Register to Reply
Power-User

Join Date: Jan 2007
Location: Houston,Texas
Posts: 378
Good Answers: 24
#7

Re: Selecting Pipe Size simply

05/10/2008 9:15 PM

For such a short and simple installation, good 'rules of thumb' work extremely well. A very good 'rule' is to keep the liquid velocity above 5'/sec and below 10'/sec. Using a 24" ID pipe results in +/- 10'/sec @14,000GPM. An 18" ID is much too small and would likely have signification cavitation erosion problems leading to rapid thinning and leaks.

__________________
Keith E Bowers, PMP
Register to Reply Score 1 for Good Answer
Power-User

Join Date: Feb 2008
Location: Northeast Ohio, USA
Posts: 267
Good Answers: 9
#8

Re: Selecting Pipe Size

05/12/2008 8:13 AM

The size of the pipe you must use will be determined mainly by the device that is pushing the fluid through the pipe. You have indicated that this is a pump. The first thing you must do is develope a "SYSTEM HEAD CURVE" for the various sizes of pipe you are considering. This is done by taking a chart (available in most pump manufacturer's catalogs) showing the friction loss per 100 feet of pipe of various sizes as determined by the Hazen & Williams formula. This will negate the need to calculate these values over and over. From the tables select a pipe size that you believe to be closest to your needs. Then take the flow starting at just below your present flow and calculate the loss per hundred feet times the actual length of pipe in the system for several values going up to somewhere near 1.5 times your present flow. Add the static head (the actual elevation difference from the delivery point to the discharge point) to each value you get. Plot these on the pump curve.

Do this for the next two pipe sizes larger than you think you may need as well as at least one size smaller than you think you will need. Now, take each of these values and (since you have indicated that you are pumping this slurry), create a curve using the characteristic curve of your pump. Your "zero" flow will be right at the static head mark. The "HEAD" will be listed on the left side of the pump curve from bottom to top. The flow will be listed across the bottom from left to right. Connect the points you have marked ont he pump curve to create a curve based on the information you have developed. Where the characteristic curve of the pump crosses the developed system head curve is the delivery point. The flow you should expect and the head it will take to deliver that flow using your pump is that delivery point. Select the one that you can accomplish in the smallest pipe size.

The friction loss per hundred feet will also depend on the coefficient of friction of the pipe you are using. I would guess that that number will be about 140. The tables you are using will probably have that value as an option. If not, there is a ratio of whatever friction loss coefficient the tables represent to that of the pipe you ar using. Apply that ratio ( direct proportion) to your values.

Let me know how you fare.

__________________
All that is necessary for the triumph of evil is for good men to do nothing. - Edmund Burke
Register to Reply
Guru
Engineering Fields - Piping Design Engineering - New Member Egypt - Member - Member since 02/18/2007

Join Date: Feb 2007
Location: Cairo, Egypt
Posts: 1733
Good Answers: 248
#10

Re: Selecting Pipe Size

08/20/2009 12:48 PM

The following CR4 Thread Arabian Heavy Crude and Flow Rate is very interesting in sizing of a pipeline by calculating the pressure loss using ready made charts for such a fluid for a given flow rate.

For sizing of a pipeline, note that there are many scenarios of different diameters where each scenario satisfies the flowrate and hydraulic requirements. Every scenario characterized by such a pipeline diameter and certain pressure drop, taking into consideration that there is a recommended flow velocity for each fluid.

• For pipeline with larger diameter (higher cost), there is a smaller pressure drop which needs a smaller pumping station (low initial and annular cost).

• For pipeline with smaller diameter (lower cost), there is a high pressure drop which needs a higher pumping station (high initial and annular cost).

And there must be an economical study to facilitate selection of the optimum scenario.

__________________
It is better to be defeated on principles, than to win on lies!
Register to Reply
Register to Reply 10 comments

Good Answers:

These comments received enough positive votes to make them "good answers".

"Almost" Good Answers:

Check out these comments that don't yet have enough votes to be "official" good answers and, if you agree with them, vote them!
Copy to Clipboard

Users who posted comments:

Abdel Halim Galala (3); Anonymous Poster (2); bob c (1); Hendrik (1); Keith E Bowers (1); Natterre (1); The Commoner (1)

Previous in Forum: Selection criteria between pipe & tube   Next in Forum: Pneumatic Wizardry!

Advertisement