After the initial experimental investigation of a single-magnet levitation, we gain a better understanding of the control and the detection scheme. In this paper, we will outline a plan for 4x4 magnetic suspension.
Goal
The goal of the second stage experiment is to figure out how to control such a more complicated system with at least 3 degrees of freedom. Once stable suspension is achieved, we need to measure the transfer function of different degrees of freedom to see whether we are able to achieve low-frequency isolation or not.
At this stage, we will still use analogy circuits for control. In the next stage, we will apply digital control which is essential for low-frequency control.
Basic Configuration
The schematic configuration of the proposed setup is shown by the figures above. The left one is the side view and the right corresponds to the top view. The are four control coils four OSEM shadow sensors. Since we only have 3 degrees of freedom (vertical+tile+yaw) to control, in principle, we can spare one of the sensor and one coil. One more sensor gives us some flexibility in defining the control matrix, as long as we can control 3 degrees of freedom independently.
Hardware
Mechanical parts
(i) 2 mounting plates
>> We only have one at the moment, and we need order one more.
(ii) 2 plastic plates for holding the levitated magnet and the sensors.
>> We can easily find them in the mechanical shop.
(iii) 4 clamps
>> We have 2 at the moment, and we need to search whether there are spared ones in 40 m or not.
(iv) 4 poles
>> We already have 4 of them.
(v) 4 Flags
>> We try to find them in the cabinet.
Electronic parts
(i) 4 Coils (800 to 1000 turns with a wire diameter of 0.5 mm)
>> We can either buy some bobbins to make them by ourself or we can order 4 of them from either Torelco http://www.torelco.com/products.html or APW http://apwcompany.thomaswebs.net/coil-winding.html which can manufacture custom coils.
(ii) 4 OSEMs and associated circuit components for them
>> We already have 4 old spared ones we can use. If we can use ordinary operational amplifier, then we will not need to order anything.
Timeline for the experiment
(i) 2 or 3 days to design the matrix circuit for the control;
(ii) 2 days for testing the circuit and measuring the transfer function;
(iii) 3 days for assembling mechanical parts and making necessary adjustments;
(iv) 2 days for assembling electrical parts and testing the error signal;
(v) Several days for achieving stable levitation (Uncertain);
(vi) If we succeed in obtaining stable leviation, we will measure the transfer function of different degrees of freedom and measure noise performance.