e/D - field data versus Moody

Reviewing Moody's 1944 paper.  Moody, L. F. (1944). "Friction factors for pipe flow." Trans. ASME, 66:671-678.

His Fig. 2 is a look up for e/D values. I plotted the Winter/Summer friction factor numbers on the figure and the e/D values range as thus:

or

RC: 0.00015 to 0.0006
CC: 0.0005 to 0.00112

In comparison, previous Master's work found e/D values for

RC: 0.000075 to 0.00045
CC: 0.000075 to 0.00095

These are significantly smaller values for the friction factors we see on the pipelines.These values were based on Colebrook/White equation.

I'm seeing Reynolds numbers in the 10^4 (1L) to 10^5 region. I think C/W is only good in the transition region. I'll need to verify.

One of the things I do like that Moody did was to plot all the different equations (Von Karman/ Pandlt/Nikraduse, Colebrook, Rouse) against each other.

Had an "aha" moment.

We pump in the transition zone (purple box below). I've known this for a while.

Moody also discussed how difficult it is to come up with a real absolute roughness value. The values the industry uses are the ones based on the sand grain sizes from Nikraduse. Moody indicates the engineer should rely instead on the hydraulic roughness, or e/D.

From Moody:

When the thickness of the laminar layer, which decreases with increasing Reynolds number, becomes so small compared to the surface irregularities that the laminar flow is broken up into turbulence, the flow conditions pass over into the zone of "rough pipes" with complete turbulence established practically throughout the flow. Viscous forces then become negligible compared to inertia forces, and f ceceses to be a function of the Reynolds number and depends only upon the relative roughness, giving horizontal lines of constant f in the chart.

Which means the evidence I see with the cyclical f versus Reynolds numbers values DO have significant viscous forces at work. So can this mean that we could have differing flow regimes based on water properties of the water (density/temperature/etc)?

Moody also mentioned the results of Colebrook plots that produced results from an Ontario large diameter penstock. The f v. Re values put the results close to very smooth pipe.

Another item to vet out is the absolute roughness/boundary layer values. Appearently, Moody based his work on a constant value 6.08, which is considered somewhat high.

Also, Rouse showed the e/D values Colebrook found, fall within experimental scatter.

So now I'm starting to think about how this relates to biofilm. These values are all based on static roughness measurements. I think I may want to consider abandoning the pipe roughness v. biofilm roughness in favor for a dynamic hydraulic roughness of the pipe.

I do see cycical variation. Do I tie it to temperature? Viscocities? I lean toward viscosity since previous work showed temperature was somewhat independent of other water properties (the linear regression class work).

Could this be significant work?

Literature Review.

Friction factor is independent of the Reynolds number (Darcy, H. : Recherches experimentales relatives au mouvement de 1'eau dans les tuyaux. Memoires a l'Academie d. Sciences de 1'Institute 
imperial de France, Bd. 15, 1858, p. 141.)

Relative roughness (k/r) where k is absolute roughness. (v. Mises, R.: Elemente der technischen Hydrodynamik. Leipzig, B. G. Teubner 1914.)

Since kinematic viscosity impacts the Reynolds number, I would like to understand what kinematic viscosity really means.

Kinematics: Geometric in nature, defines motion without regard to forces that cause motion or result from motion.

Kinematic viscosity is traditionally measured by noting the time taken for a fluid sample to travel through an orifice in a capillary under the force of gravity.

Every type of fluid possesses differing amounts of resistances against deformation. The measure of that resistance is called viscosity. Viscosity expresses the fluid’s resistance against either tensional stress, or shear stress.

Fluid Dynamics

So I got my Experimental Fluid Mechanics book and started on page 1.

Ugh.

Thought I didn't have to deal with thermodynamics ever again, but it totally makes sense that it is a big part of it.

First thing off the bat: total internal energy is a function of entropy and density.

e = e(s, rho)

Also I watched this course

Problem Statement - first draft

This is a first attempt at a problem statement.

Standard design parameters used by engineers for pipeline design uses static hydraulic conditions between new pipe and old pipe by varying friction factors. Previous work has found hydraulic conditions cyclically vary during one calendar year to where energy loss ranges change from values equal to either new pipe (winter conditions) or aged pipe (summer conditions) friction factor values. Thus, there are two different types of roughness: pipe roughness and biofilm roughness. Pipe roughness is based on actual rugosity of the pipe wall. Biofilm roughness is based on the seasonal variations of friction factor, which directly quantifies energy loss, that can be attributed to biofilm growth. 

Previous work on pipe roughness for large diameter pipe, or pipe that is 60-inches in diameter or larger, found that when experimental results were plotted on the traditional Moody diagram, the ratio of the roughness of the pipe to the inertial forces captured with Reynolds number does not follow the classic model used with the Moody diagram for large diameter pipe. 

Biofilm is considered a viscoelastic medium in that the thickness can vary. Previous work has found velocity of the water in the conduit has an inverse relationship with the thickness of biofilm. The higher the velocity, the thinner the biofilm and visa versa. Biofilm thickness is also impacted by environmental conditions of the water within the pipe. An environmental process which can directly impact pipe roughness over time may be due to the presence of biofilm in the pipeline as evidence shows that nitrification occurs within the pipeline. This process lowers the pH of the water in the line which then can put any cementitious lining in jeopardy of increasing pipe roughness since the Langlier Saturation Index (LSI) will be negative, which indicates calcium carbonate (CaCO₃) dissolves in water. In the system studied, the limestone aggregate as well as the sand contain high amounts of CaCO₃ which have been eroded over time increasing the rugosity of the inner pipe wall.

This paper will try to 

Establish the difference between pipe hydraulic roughness and biofilm roughness in long large diameter raw water transmission pipeline.

Establish there is a relationship between algal growth and increased friction factor values and in turn, determine the change in the hydraulic roughness of the pipe. 

Establish if there is a spacial and time relationship between LSI, pipe rugosity, and biofilm growth.

More basics

Reynolds number is the ratio between inertial and viscous forces of a fluid.

Here's a link to a lecture that goes over more mechanic basics. I've been reviewing this one.

I like how it goes over Reynolds numbers pictorially.

Came across this in reading the BR Tech memo

What experimentation indicates that frictional resistance varies at the first power of the velocity and the second power of the pipe diameter?

Back to shear wall stress basics

I'm pretty exhausted tonight, but I did dust off some basic review.

Headloss can be found using the following equation.

 where:

Combining that equation with the Darcy-Weisbach yields the following for the wall shear stress:

Thus the wall shear stress is located at the wall of the pipe, where the velocities will be smaller than what is found towards the center.

So biofouling makes the wall shear stress increase when the film gets thicker. This would effect flow rates as it will impact the velocity distribution profile. So I can now use the wall shear equation to relate it to velocity.  Right now, I get flow rates based on differential pressures in a venturi tube.

 These are based on set diameters and distances within the venturi tube. I need to review how these go back into SCADA with the Pressure Indicating Transducers (PIT). The idea is the flow rate is a true flow rate measurement and I can use the venturi meter to come up with a true velocity based on a fixed diameter. The thought is that biofilm decreases the area of the pipe. Could I get a correlation of biofilm thickness to the pipe wall with this analysis? If I used the pressure transducers I have out on the pipeline, could I come up with an aggregated thickness (like a mean thickness value)?

Non-neutonian fluids are fluids like toothpaste and cornstarch mixtures - i.e. sludges. Sludges will tend to move in the laminar flow range. (Refer to email from colleague from the Day 1 post). While algal fouled water may have a higher viscosity value, it's not moving the regime into laminar territory. I do believe viscosity is at play, so this does have merit.

Nitrification and Carbonation of Pipe

3. Biogenic nitric acid The biogenic ammonia present in water environments stimulate the concrete corrosion caused by the biogenic nitric acid. The ammonization bacteria (urobacteria) decompose urea and cause the ammonia formation. Bacteria belonging to Nitrosomonas, Nitrosococcus, Nitrosospira and Nitrosolobus genera oxidize the ammonia to nitrous acid subsequently oxidized to nitric acid by Nitrobacter and Nitrococcusbacteria. This acid reacts with calcium compounds in the hardened cement paste to form the water-soluble calcium nitrate, easily rinsed from the concrete structure[17]. It has been shown that the biogenic nitric acid produced by the nitrifying bacteria causes severe corrosion of the mineral building materials. High cells numbers of the nitrifying bacteria were present on both historical sandstone buildings as well as on modern concrete buildings [8, 9].

We know we have Nitrification going on in the pipeline. Looks like we're generating nitric acid which is why we're getting high levels of concrete degradation.

Is there a way to determine this in the concrete samples of the pipe with the petrographic analysis? Can he test for calcium nitrate?