Do you pump the overflows and leaks back to the process? If yes, then the discharges from the process will be the products (floats and sinks) and magnetic separator, isn't it?

Yes to both questions.

Do systematic Planning and start discussing how to control magnetite loss with your superior and subordinates and maintenance team. And plan how to reduce above loss. That is how it could be controlled. I also suggest don't put unnecessary Magnetite in fresh feed. Put Specific Gravity of cut under control. Which helps in controlling Magnetite loss. Try this. I have done same to control it.

Most of the answers are contained in your question itself.

1. For "Loss through product" please take samples of the clean coal, reject and middling for laboratory test by thorough washing and determine the .magnetite content. Repeat the test after suitably changing the settings of the rinsing screen, such as spray water rate, screen vibration etc.

2. For "Loss through magnetic separator" similarly take the samples of non-magnetic product and determine magnetic content in the laboratory.

.

3. There should be no "Loss due to overflow of dilute medium tank and bath" if the overflows are pumped back into the dilute media sump according the standard practice.

4."Loss due to leakage from pump and pipe lines" is difficult to assess even with careful material balance computations. Please try to follow the best practices for minimizing the leakages.

The through systematic analysis of each stream suggests is the right way to go. Once you know where the loss occurs, you can modify that process to limit or stop the loss of Magnetite.

Yes sir, first of all I am analysing the possible routs of magnetite loss from a heavy medium bath coal beneficiation plant and after wards try to reduce this by taking possible corrective action.