Tabulated Friction Head Loss for Non-Water Services 2

[ghamsa]

Cameron hydraulic data book has tabulated friction head losses for water in different pipes sizes at varying flow rates.
At specific flow rate (GPM) and pipe size & schedule the tabulated data shows velocity (ft/sec), velocity head (ft) and head loss (ft) per 100 feet.

My question is:
Can I use this tabulated data for other liquids i.e. hydrocarbon, say SG 0.5? If yes I can; then what correction factors I should consider?

Thanks for usual help

 

[katmar]

This is an interesting question. The answer is in the Darcy equation. For the flow of incompressible fluids in pipe this is
[Δ]P = ( [ƒ] x L x V² ) / ( 2 x D x g )
where
[Δ]P is the head loss in feet of the flowing fluid
[ƒ] is the Darcy friction factor (dimensionless)
L is the pipe length in feet
V is the fluid velocity in feet/second
D is the pipe ID in feet
g is the acceleration of the earth's gravity in ft/s²

From this you can see that if you change the fluid's SG and you keep the flowrate the same in volumetric terms the only factor that can change is [ƒ]. Fortunately, in turbulent flow the friction factor is a fairly weak function of Reynolds Number. While halving the SG will halve the Reynolds No it will have much less effect on the friction factor. So, as a first estimate, you could say the head loss is unchanged.

However, the viscosity will also have an impact on the Reynolds No and therefore on the friction factor and head loss.

It is very important to note that the head loss only remains constant when measured in feet of the flowing fluid. If you halve the SG, and measure the pressure loss in PSI you will find that the lighter fluid has a significantly lower pressure drop.

The effort in applying a correction factor to your tables will be as much work as doing the calculation properly in the first place. So I would say that for rough estimates you can assume that the head loss is unchanged (remembering that it must be measured in feet of the flowing fluid), but for serious work you should do the calculation properly.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[25362]

It appears that the error following katmar's advice wouldn't be too great for Reynolds number in the range 30,000-100,000.

 

[hanon]

I use the Chen equation which give explicitally the friction factor. I quite useful because it gives directly the friction factor for a given Reynold number without having to iterate as happen in other correlations which are implicit.
Regards, Hanon

 

[Montemayor]

Harvey, as usual, is exactly correct. As a trivial exercise, some years back, I took the Data in the Cameron Data Book (pages 3-12 to 3-33) - which are the same as those found in the Hydraulic Institute Engineering Data Book - and tested it with the Darcy equation. It was exactly the same resultant pressure drops. What I did was I set up my Excel Spreadsheet with the basic input data and then applied the Darcy equation to all the input flow rates. The results were the identical Data Book results.

So, guess what was used to generate the Data Book(s) table?

If I had need for a specific liquid -other than water - I would create the same type of table using the same equation: the Darcy. It's quite very simple to do on an Excel spreadsheet.

 

[25362]

Others indicate the friction drop, psi/100 ft of pipe length, for turbulent flow, is proportional to [ρ]^0.85[×][μ]^0.15 for medium rough pipe surfaces. The exponents change a bit as function of surface rugosity.

[ρ] = density of the fluid
[μ] = absolute or dynamic viscosity of the fluid

Thus the friction losses for a liquid hydrocarbon (HC) compared with water (w) would be:

[Δ]p_HC=[Δ]p_w[×]([ρ]^0.85[×][μ]^0.15)_HC[÷]([ρ]^0.85[×][μ]^0.15)_w​