The way you answer this question is to use a reference like "The Crane book" from the Crane company.  They are in Chicago, Illinois.  It is called "Flow of Fluids through valves, fittings, and pipe".
It gives formulas and tables for estimating delta P through pipe of various fluids.  If you are interested in water or air specifically through carbon steel pipe, there are tables to use, you don't have to calculate anything.
Now, to give you some guidance, the maximum flow you can get through a pipe is based on delta P, errosion, cost of energy, etc.  As you try to push more fluid through a pipe, the delta P goes up.  Pressure drop costs energy.  In addition, as you increase the velocity, the erosion rates increase.
In general, I have used the values of 5-10 feet per second as the range of velocities to design for in a piping system handling liquids.  Below 5 and you spend too much on the pipe size, above 10 and the energy costs become excessive.

I hope that helps you.  If you need more, contact me at:

Robert.Sander@Solvay.com


Bob Sander
robert.sander@solvay.com

The maximum permissible velocity of fluid in a pipe  depends on the amount of solids being carried by it (erosion criteria/ solid deposition criteria), nature of fluid and the possibility of flashing in the pipe (due to pressure drop in which case the single phase may become multi-phase),corrossive nature of fluid and pipe type and the maximum permissible pressure drop criteria. For simple water pipes, many of these factors may not be significant and can be excluded and maximum permisssible pressure drop criteria may be governing. However for many other fluid types (e.g. hydrocarbons), any one or more of these factors may become governing.      

The answers that the others have given you are very good ones, in other words there is not a singular value for the velocity of a fluid inside a pipe.  Most rules have been derived from both cost and operating experience.  Thus, I will take liberty in using the previous answers and order them as follows:

a)  What are the process requirements (pressure and flow) downstream.
b)  What is the service (regular flow line, intermittent, blow-down, depressurizing)
c)  Is noise an issue?
d)  What are the variations in the flow and pressure with time
e)  Fluid characteristics (composition, solids, Newtonian or non-Newtonian)
f)  Phase behavior (single phase, two phase, does it have water?)
g)  Flow regime (slug, annular, mist, bubble, etc)  This is particularly important when designing pipe leading to separators because when the flow is in the mist regime gravity separators do not perform as desired.
h)  Cost of energy, equipment and construction.  Many rules of thumb were developed in the 1950´s and 1960´s, since then the ratio between the cost of energy, equipment and constructions costs have varied significantly and are very widespread throught the world.
i)  Start-up and commissioning considerations (During construction there is an abnormally high amount of debree and other contaminants present) and the pipe must be designed to allow for this phase of the project.

If you're looking for tables and charts that give recommended velocities and diameters, or methods for calculating,Cranes "Flow of Fluids" is good bet. So is, Ingersoll/Dressers "Cameron hydraulic Data"; so is Navco's "Piping Datalog". All three are used extensivley in the design of Process Plants.

franky,

All of the comments above are good, valid answers to the question "What is the recommended pipe size for a given service ?"

The answer to your question - "What is the maximum velocity for water service ?" is "How long do you want the pipe to last ?" With even modest sediments and a velocity of 20+fps, you will not get much life oft of the stsrem and you will be paying high energy costs to move a liquid from point "A" to point "B"

The USArmy has a guideline on piping design available on the web:

Try:www.usace.army.mil/inet/usace-docs/eng-manuals/em1110-1-4008/c-3.pdf and look at page #8

Tell us more about you particular situation and we will be able to make recommendations....

Good Luck

MJC

Use guidelines with caution and an open mind.  Typically 5 to 7 fps are used for pressurized lines and 3 to 5 fps for drain lines.  Don't get bent out of shape if it goes above 7.  All is correct that you use more energy as the pressure goes up, but it is also dependant on the run length.  The best example of this that we have all seen, but few have studied is under our own kitchen sink.  Typically a kitchen sink faucet can deliver 3 to 5 gpm.  Typically these faucets are connected with 1/2" pipe.  This guideline gives you a velocity of 3.2 to 5.3 fps.  Funny how that works out on the low side of the 3 to 5 fps number, but in older homes that a fitted with galvanized pipe there is a much smaller usable pipe bore due to calcification.  Also note that most kitchen sinks has 1.5" tube drain lines.  That works out to a velocity far lower than the 3 to 5 fps number, but don't forget about the hair/grease/chicken bone obstruction.  

Then as far as the maximum velocity is concerned, closer inspection under the sink shows the 1/2" line connected to the faucet through a shutoff valve that has a 1/4" bore and a 1/4" id tube (3/8" od).  The 3 gpm is going through this tube at almost 20 fps and 5 gpm screams by at over 32 fps.  But it is at the end of the run and only about a foot.

In a perfect world you would pick a flow that resulted in a system pressure drop that fell on the highest point on the pump efficiency curve.  Stuff like this, is why guidelines should only be used as a 1st approximation.  Temper it with experience, and a more rigorous calculation where appropriate.

Hi Franky,

It is just for additional. Maybe, if you need something like "design practice" you can ask to EPC (Engineering Procurement Construction) Company people. In my experiences, many companies have own spec., they have own prediction and consideration. You can choose one of them which similar with your current job (e.g. : NGL or Fertilizer) as your guideline.
Please tell me if you need further informations.

Bye

Franky (if you're still out there)

In the oil and gas business we use API 14E "Design and Installation of Offshore Production Platform Piping Systems" as a guide.  14E gives the following equation to help designers select acceptable erosional flow velocities:

V, ft/s = C1/(fluid density, lbm/ft^3)^0.5

they recommend setting C1 = 100 when better C1's are unavailable. This relationship is used for gas and liquid. This equation gives a V of 12.7 ft/s for water.

M. Salama presented a paper at the 1993 OTC recommending the following C1's

135 for carbon steel
186 for stainless steel
236 for duplex stainless steels

These velocities are higher than many industrial designers are comfortable with.  They are commonly used offshore, where the pressure drops are deemed acceptable.

Regards,

Gunnar

How come this thread , which was orginally started in 2000, still continuing ?

gkandy, Probably because it deals with engineering fundamentals that are needed and utilized on a daily basis by engineers/technicians living and working in the trench's.

saxon

For clean hydrocarbons there is also a limitation on linear liquid velocities to avoid charge separation and static electricity discharges thereof.