I am planning to investigate the effect of bend curvature of a pipe on velocity of water. For this, I will need to find the velocity of water through various pipe bends. Would placing an object in the water such that the objects velocity can be measured and equated to the waters velocity do the trick?
I assume you want to see the velocity profile and not just the average velocity?
If so a common method (used for all the photos in Tritton) is to introduce a small particle/dye into the fluid. If you have see-through pipes you can quite easily caluclate the velcity of the particle/dye. You can then change the radius at whoch it gets introduced to see the velocity profile.
In order to stop turbulance I would recommend a slow or viscous fluid (low reynolds number), do you know about dynamical similarity?