Increase in EGL?

danago

I recently attended a lab at uni where we were required to examine the conservation of energy for a fluid steadily flowing through a pipe with a series of obstructions.

The actual setup was quite simple; A pipe of length ~4m ran horizontally (no change in elevation throughout the whole flow) with a series of 18 manometers evenly positioned along the tube. Along the pipe were two obstructions:

- A venturi meter
- An orifice plate

Based on the pressure measurements along the pipe, along with the known mass flow rate through the tube, we were required to plot an energy grade line (EGL) and hydraulic grade line (HGL). Now based on energy conservation principles, i know that the EGL should never increase (there were no sources of energy such as a pump), but should instead steadily decrease due to major and minor losses throughout the flow. However, this is not exactly the case that i found in doing my calculation:



Note that at manometers 12 and 13, the total head of the fluid increased very slightly. With regards to the obstructions mentioned above, manometer 12 is about 0.2m after the venturi exit, and manometer 13 is immediately in front of the orifice plate.

We conducted four different trials at different flow rates, and the exact same thing happened each time. Now i do understand that this is not possible, so it must be due to a calculation error. The reason i first proposed for this discrepancy is that because of the flow obstructions and resulting turbulence, the actual "effective" area of fluid flow may not have been the same as the cross section area of the pipe, perhaps leading to a misleading calculation of the velocity head?

The question i am asking is -- What possible reason is there for this misleading jump in the EGL, and is there a simple way to account for it in my calculations?

Thanks very much, any help is greatly appreciated,
Dan.

danago

To help in answering my question, here is a diagram of the setup that was used:



The manometer scale corrosponds to the manometer scale used on the EGL/HGL plot in the first post.

Thanks again.