Abstract
A copper pipe apparatus was used to compare the differences between the experimental and theoretical friction losses in pipes with different geometries. The difference in the head and the flow rate were measured using the apparatus. From this it was possible to calculate the equivalent sand roughness, friction factor and relative roughness. These results enabled comparisons to be made. It was found that the diameter had a large effect on friction factor which was the cause of head loss. The greater the diameter of a pipe the lower the friction factor. (sean)

-
-

Aim(s)
-

The aim of this experiment is to calculate the difference between the experimental values and theoretical values of frictional losses in different pipes so a comparison can be made. It also aimed at giving participants a firsthand experience of friction loss to help understand head loss in pipes. (sean)
-
-Method
-

1. Turn the pump on and ensure it operates at its maximum supply pressure, checking that at least one pipe is fully open to flow. Leave the pump running to remove all air pockets possibly within the pipes.

2. All parameters in the experiment were measured. Use measuring tape to measure pipe lengths and variations in elevation. To measure the pipe diameters use the vemier callipers on the cut sections of each pipe and finally use the electric thermometer to measure the temperature.

3. Using the ball valves the flow was directed through pipes A, B and C and using the gate valve with the three ball valves shut allowed for flow through pipe D. using the manometer connections and attaching them to the appropriate pipes on the apparatus the piezometric heads was obtained in meters of water head. Great care was to be used to prevent overflow and underflow which dramatically affect the readings.

Three readings of head entry pressure and head exit pressure were recorded per pipe. With the manometer connections swapped from reading entry pressure to exit pressures each measurement to reduce error.


4. The flow rate was also measured three times per pipe by timing, using a stopwatch, how long it took to pump five litres into the sump.
6. Across the experiment consistency needed to be maintained so all the parameters that were measured at the beginning of the experiment needed to be measured at the end. (sean)

-


Results
- Shayne from the results you put into the table a believe the relative roughness for pipe A should be

0.0035
0.0030
0.0036
respectively. I think you looked on the wrong side of 3 for the friction factor when reading off the moody diagram. (sean)
-Also in the pipe parameters in the change in elevation row it is in (mm) and the change in elevation for pipe D is given as 41.2. I believe it is corretly converted later as a agree with the rest of the results.(sean)

-I was pretty ambiguous with regards to the pipe parameters for pipe D, sorry about that. What I meant was, shouldn’t it be 412 instead of 41.2. As it is in (mm) and from the photo’s it can be seen that it is much larger than this.

-(Kate) can you double check the moody diagram (the one in your course notes is easier to read and larger) with the corrected lab results that I have just put up.
-I doubled checked, make the 0.0028 into 0.0036. You can have a quick look if you like to confirm. (sean)


Discussion
-It should be noted that pipe C has half the flow rate of that of pipe A and this is obviously because the pipe C has the flow rate of the pipe A split between two identical pipes(the paralell set up) but we only measure the head loss in one of the paralell pipes. This indicates that the difference in head loss between pipe A and Pipe C should be exactly half which is confirmed by the results.(sean)
-
The calculations are extremely sensitive to variations in the pipe diameter. The actual velocity, Reynolds’ Number and equivalent sand roughness are inversely proportional to diameter, whilst the friction factor is proportional to diameter. The diameter can affect the results in that a large diameter will relate to a relatively large friction factor which may not correspond to a relative roughness line on the Moody Diagram. This means that there is no way to tell the relative roughness. (sean)
-I am unsure as to how the difference in elevation effects the friction factor or head loss.(sean)
-If a reynolds number is above 2x10^3 the flow is possibly partly laminar and partly turbulent or wholly turbulent, for each pipe the reynolds number was was above this number with the lowest going to pipe C. This was because the it had a flow rate approx. half that of the others due to it being in paralell. The next lowest was pipe B which had a diamter approx. twice the size of the other pipes which meant it had a lower flow velocity. The reason the pipe D had a slightly lower reynolds number than pipe A was becuase they ad slightly different flow rates. This could be contributed to the larger friction losses in pipe D due to the rust.(sean)

-It should be noted that the friction factor is inversely proportional to the veloctity squared. As this is the dominate factor in the friction factor it can be seen how the diamter reduces the flow velocity so much as flow velocity is equal to the fow rate divided by the area of the pipe. This is why the friction factor is so much smaler for the larger diamter.(sean)
- Since Pipe A and D have similar flow velocity the reason why their friction factors differ so greatly is because friction factor is also proportional to the head loss and inversely propportional to the lenght. As pipe D is shorter friction factor increases. As head loss increases friction factor increases. (sean)

Error Analysis
Sources of error

-manometer pressure difference (can't remember why it could induce an error; can someone help me with this one)
I sent an email to shayne on my oppinion to the manometer pressure error. (sean) (KATE) Yes I received that, Shayne is too busy today and tomorrow to contribute much more. Please CC me on any further emails to save Shayne from needing to forward on to me.

-(sean)- roger, also another contributing factor to error in the manometer pressure is if when they are being calibrated to move the range in which the manometers up and down on the two measuring manometers it is possible to get air bubbles or some minor discrepancy between them.

-manometer fluctuations
-paralax error
-diameter
-


Conclusion
-The results of this experiment show that teoretical frictional losses and experimental frictional losses are closely linked and the difference could mainly be contributed towards exerimental errors that were reproduced in the results. It also shows that the diamter has the largest effect on friction factor. Finally the degration of a pipe over time leading to rust also has a dramatic affect on friction loss. (sean)


Head losses occur during the flow of a fluid through a pipe from friction, added accessories (taps, couplings) and elevation changes. As a pipe ages, it develops deposits of the internal wall and as such result in losses in kinetic energy and thus are converted into heat. If this situation were magnified to a domectic, commercial or industrial scale, then these losses would result in extra energy consumption to achive a required output.
It is necessary to allow a pipe flow to become fully developed before analysis can be conducted. Since a larger diameter pipe required and longer entrance length, the length of pipe provided in the experiment was not sufficient for analysis.
A minor variation in measurement of pipe diameter has a significant effect on the friction factor and volume flow rate calculation and an error at this point will result in an incorrect pipe flow analysis. It was for this reason that the use of precision measuring equipment was necessary.
Employing the use of the Bernoulli equation is useful in determining the energy balance in a pipe flow problem but it is flawed when pipe losses need to be accounted for. A modified version to the Bernoulli equation detailing head losses provides a useful tool in the analysis of real pipe flow problems.





-
-


Recommendations
-
-
-


References
- http://www.lightmypump.com/help14.html (sean)
-
-