Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

500 kg Solar CSP reflector dish mechanism problem

  1. Jan 22, 2017 #1
    I am currently re-fabricating solar concentrator dishes to allow for automatic tracking of the sun during the day and month. The industry term for these dishes is 'CSP' (Concentrated Solar Power).
    Task at hand:
    The CSP parabolic dishes are currently manually adjusted to the position of the sun in the sky everyday. We are assigned the task of automating their movement in-order to save on everyday labour. I already have tracker electronics ready with me and am only left with the main task of re-fabricating the mechanism/structure.
    Automation is achieved using a linear actuator and a prismatic cylinder mechanism placed on either side of the CSP dish swivel bearing lever arm (please see sketch below). The linear actuator pushes and pulls the mechanism and the prismatic cylinder moves accordingly, thereby creating a torque effect on the swivel bearing and rotating the CSP dish as a whole. The problem lies in 'synchronisation'. What if the prismatic cylinder mechanism gets stuck? This would result in breakage of an expensive parabolic dish and can have other disastrous safety consequences.
    My solution:
    In order to add a redundancy mechanism in case of failure, I have come-up with an idea of using a tension wire pulley arrangement which connects both ends of the lever arm together. I achieve this by welding a clevis on the bottom and top of the dish and connecting the tension wire to them and running it via a pulley. This implies the linear actuator pushing/pulling on the dish would also tighten the tension wire and cause the prismatic cylinder to compress.

    I would like to validate my solution further from PF. Please throw out some ideas. I embedded a video also below for a more clear visualisation of the system. The engineers in the video are trying to manually move the dishes by loosening a locknut arrangement both on top and bottom of dish.


    Attached Files:

  2. jcsd
  3. Jan 22, 2017 #2
    Wow ... that's some very serious dishes you have there ... Stainless steel reflecting surface? ...aimed at a central tower ?

    I've been looking at and experimenting with solar for many decades , have my own 6kw PV set up .... I would guess your project was started many years ago when the price of solar panels was relatively high ... PV panels have fallen in price so much over the last 5 years that nothing can compete with them ...

    the size of your dishes makes them very vulnerable to wind damage and any actuator would have to be waterproof , and rain gets in everywhere .
    The type of actuator you need can be purchased , but the price would be too high , and the travel distance too short ... if you make one yourself rain will get in and other problems .... I would do a serious cost analysis ...Sorry I can't be more positive ...
  4. Jan 22, 2017 #3


    User Avatar
    Science Advisor
    Gold Member

    Feedback control system .

    Gives precise positioning and can also detect if drive is behaving abnormally .
  5. Jan 22, 2017 #4
    Hi Oz. We did not manufacture these dishes, so I'm not too sure on the material although I could take a guess. These are not SS, probably silver and copper coating on glass.

    Price of PV unfortunately does not have major effect on CSP. These are just reflectors and have no photovoltaic component in them for energy conversion.

    We use IP65 ingress protection standard for our actuators and sure they can withstand wind and rains for the next 3 years.
  6. Jan 22, 2017 #5


    User Avatar

    I am having trouble identifying the actual pivot point of the dish. The video only shows views of either the top or the bottom. Do you have a photo of the the entire unit? What is the primary reason for the bottom actuator, is it for stability of the dish?
  7. Jan 22, 2017 #6
    CSP fabrication parts (1).jpg
    Hi JBA. I hope the photo above makes it clear to you where the swivel point of the dish exactly is. This is an eccentric rotation.

    I've marked in the photo where I would place the actuator and top-support assembly. The top-support assembly would be a prismatic mechanism with a rod moving inside a pipe. Primary reason for the actuator is to generate a torque and rotate the dish.

    As you can see from the photo, the existing top-support is bent to a small extent. In order for me to avoid this during actuation, I need to make sure I have a redundant mechanism/electronics in place to synchronise movement, which is why I came up with the idea of the tension wire and pulley arrangement (see main post) and a current sensor to sense 'actuator-block'.

    Hope explanation was clear?
    Last edited: Jan 22, 2017
  8. Jan 22, 2017 #7
    Thanks for the reply Nidum.
    Detecting if the drive is behaving abnormally would be possible only if I use a current sensor to sense actuator-block. By actuator-block I mean that some part of the mechanism in the system has malfunctioned/stuck. In this case, the motor driver would see a sudden current surge for a certain period of time and the algorithm in effect should auto-shut off the entire system.

    Was this what you meant?

    Precise positioning is unfortunately not possible because my actuators don't have reed sensors built-in to sense stroke length. The only solution I see is to either use a accelerometer or a rotary encoder at the joint. Positioning and re calibrating would be a challenge though.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted

Similar Threads for Solar reflector dish
Attaching fins behind a solar panel