Project
| # | Title | Team Members | TA | Documents | Sponsor |
|---|---|---|---|---|---|
| 42 | Monitoring System for Rotating Turbines |
Alaa Qarooni Asm Khandker |
Zhen Qin | design_document0.pdf design_document0.pdf final_paper0.pdf other0.pdf presentation0.pptx proposal0.pdf |
|
| For our project we are proposing a monitoring system for turbines (in airplanes, various cooling systems, etc.) or other sub systems that have mechanical components that rotate and/or oscillate with fixed RPM/frequencies Implementation and components: -We are going to implement this primarily using a photonic integrated system that employs a fiber coupled laser diode. The output of the fiber will couple on to an appropriate semiconductor photodetector. -We will introduce a gap in the fiber where we can introduce an armature connected to the axis of the fan/turbine. This will allow us to mechanically pulse the laser at the frequency of the turbine. -Alternatively, we can use a doped fiber to introduce gain inside the fiber cladding in order to account for any losses at the gap. The VCSEL in the coupler will now drive the new gain medium. -The detector will have to once again be selected accordingly (to prevent saturation). However, instead of Q-switching a fiber laser we can pulse the output of a fiber ring amplifier should we need a more substantial laser output. This may be easier to implement. -We will use a discriminator circuit to generate alternating electrical signals from the fluctuating detector photocurrent. A microcontroller will be used to calculate the time between zero signals (corresponding to a low Q for the fiber cavity; meaning the laser was clipped by the armature). The inverse of this time will give us the frequency of the turbine. -> An Arduino micro-controller is most suitable to achieve this and following bullet point, as it provides the necessary complexity and flexibility required by these steps when implementing. -We will use the microcontroller to determine if this frequency (which is constantly being calculated at every point in time while the system is operational) is held at a predetermined and programmable optimal value, or if it is falling or rising. -If it is falling below optimal value the microcontroller will turn on a backup power supply and turn off the current one which is now losing power (we can test this by manually disconnecting the primary power source). -In the rare case that the frequency is going above optimal (we can simulate this with a potentiometer which will raise the power supplied to the turbine), the microcontroller will perform the same function. -We will allow for a small tolerance in the optimal RPM so that the backup isn't constantly being turned on and off for very minute and insignificant changes in RPM. Finally, it must be said that we are using a fiber so that the controller and additional circuitry does not have to be placed near the actual turbine. This will be convenient if we are trying to implement this in a machine where the location of any kind of fan will restrict the placement of our board. Furthermore, the usage of a fiber makes the laser more controllable, gives us a stronger beam for turbines operating in inconducive atmospheres and eliminates a lot of alignment issues. Basically, the system will be modular and can be applied to existing machines without having to redesign them. -> We have talked to Prof. Peter Dragic about this project idea and he said that he can provide us with the necessary components we need, in addition to lab space for testing, debugging, etc. |
|||||