Category: Lifting Arm

Week 9 – 10 Jul: Actuator Algorithm

Lifting ArmLeave a Comment on Week 9 – 10 Jul: Actuator Algorithm

Before we can code anything to control the actuators in our lifting system, we first need to derive an algorithm for the process. As of now, we are not sure which actuator will take longer to fully extend as we have not completed our real-life motion study, so two key assumptions will be made:

  1.  500mm actuators take longer to extend as they are longer in length.
  2. The first actuator (closer to support tower) of double actuator configuration will take longer to extend as it will likely bear more weight.

 

Here is Seet Ynn’s draft summary for the actuator algorithm (if simplified to up and down motion, and forward and backward motion of arm is ignored):

  1. Press UP button, then all 4 actuators extend
  2. While length limit of 2x 300mm double actuator is reached, 2x 500mm actuator will continue to extend.
  3. When reach lifting arm height of 2.75m, first actuator of double actuator configuration will stop extension and second actuator of double actuator configuration will continue to extend.
  4. When reach lifting arm height of 2.8m, second actuator of double actuator configuration will stop extension.
  5. When the lifting arm stops moving, SMRT technicians can proceed with unit replacement.

 

Here is Seet Ynn’s draft, in full, with thinking process:

  1. Press UP button, then all 4 actuators extend (2x 300mm double actuator in parallel between support tower and first arm, 2x 500mm double actuator).
    1. The 2x 300mm double parallel actuators have to extend together as they both handle the movement of the first arm. If only one of the actuators extend, this will cause the first arm to be bent sideways and cause unintended strain on the first arm.
    2. The 2x 500mm double actuator configuration must extend with the 2x 300 double actuator to increase the rate at which the arm lifts, especially since the 2x 500mm double actuator will take longer to extend.
  2. While length limit of 2x 300mm double actuator is reached, 2x 300mm double actuator will stop extension and 2x 500mm actuator will continue to extend.
    1. 2x 300mm double actuator cannot extend beyond length limit to prevent damage to the actuator and programmable logic controller.
    2. 2x 500mm double actuator will continue to extend to increase height of lifting arm.
  3. When reach lifting arm height of 2.75m, first actuator of double actuator configuration will stop extension and second actuator of double actuator configuration will continue to extend.
    1. Lifting arm height of 2.75m is chosen, as the targeted height of the lifting arm (excluding clamp system) for it to be within the working space is about 2.8m. A vertical clearance of 5cm and slower motion should be sufficient to ensure safety and good condition of the parts.
    2. Movement of a single actuator to ensure greater precision when the unit is inside the train and close to the small working place. This will better ensure the safety of workers who may have to adjust the unit in the working space to better align it to its lock. This greater precision is also important to prevent the unit from being scuffed up against the LED screen behind the working space, and vice versa.
    3. Since first actuator is slower than the second actuator, first actuator will be stopped and locked so the lifting arm will be slowed down but not too slow
  4. When reach lifting arm height of 2.8m, second actuator of double actuator configuration will stop extension.
    1. Any future extension of the double actuator configuration at this point may cause the unit to hit the top of the working space or be too high to slot into lock, even with our highly mobile double ball joint configuration
  5. When the lifting arm stops moving, SMRT technicians can proceed with unit replacement.
    1. Safety first!!

Week 9 – 9 Jul: How to Control Actuators??

Clamping System, Lifting ArmLeave a Comment on Week 9 – 9 Jul: How to Control Actuators??

Ohhh noooooo, Mr Raman is on leave 🙁

This meant that we can’t:

  1. Grind our 2x 15cm rods down to fit into the pillow blocks better. 
  2. Get our 1x 25cm rod 

Here are the two too-big rods:

 

 

The 15cm rod (on right) is unable to fit into a significant length of the pillow block. The diameter of the pillow block probably isn’t consistent or has some (invisible to the naked eye) slight protrusions here and there, which caused the rod to be unable to fit into the pillow block.

One of our 15cm rod (on left) can fit into the pillow block quite well but cannot go past the black screw hole portion of the pillow block. As we can see, there is a slight protrusion from the black screw hole area. This may cause the interference fit between the 15cm rod and pillow block to be too tight.

 

 

Even though the rod can be rotated quite smoothly within the pillow block, even with the poor fit seen in the rod and pillow block on the right, we decided that it is not safe to use it for our lifting arm. Safety first!!!!

 

 

On the bright side, some of our ball joints arrived. These, pictured below, are typically used in cars for stick shift control. They seem to be pretty strong.

 

 

Here is a video of the ball joint rotation:

 

 

 

 

Here are some possible double ball joint configurations which we can try out:

 

 

 

 

 

 

 

We will brainstorm the pros and cons of each configuration before giving yall the update!!

 

Lastly, we need to settle how the Programmable Logic Controllers will work.

 

We bought a wireless controller which has the up, down, left, right button for remote control. However, we haven’t figure out what Programmable Logic Controller to use yet.

 

We had two possible choices for lifting arm control:

  1. Manual control of individual actuators
  2. Manual control of “Up, Down” motion

 

Manual control of individual actuators will mean that the controller will have a set of 3 up and down buttons for all three actuators. SMRT technicians will then have to know which buttons to press to control the actuators to raise the units to different heights. While this will be easier for us to programme, the SMRT technicians will need to be keenly familiar with the stroke length and angle of each actuator. This may be challenging for the SMRT technicians to master. Also, the motion of the lifting arm may be slower as actuators cannot move together, upwards.

 

Manual control of “Up, Down” motion will mean that the controller will have one set of up and down button, which controls all 3 actuators. We will have to do careful motion study in real life, with measurements, translate this into an algorithm with up and down motion and then code it into our Programmable Logic Controller. While this will be a lot more time-consuming and challenging to programme, the SMRT technicians will find the lifting arm easier and more efficient to use. Also, the motion of the lifting arm may be faster as the actuators can move together, upwards.

 

hmmmmmmm, thinking

 

 

 

 

 

Week 7 – 24 Jun: Basic Beaches

Lifting ArmLeave a Comment on Week 7 – 24 Jun: Basic Beaches

Today, we plan to build a base for our lifting arm. But first, we will discover the CG of our arm. We require the CG of arm to ensure that the far edge of the base (pivot) is beyond the CG of arm such that our lifting arm will not topple over when mounted to the base.

 

Without the weight of actuators and worm gear, we derived that the base must be at least 1.62m long from the support tower to prevent the lifting arm from tipping over.

 

We plan to attach heavy-duty wheels to our lifting arm base, so that we can move our lifting arm and use it as a trolley too. This means that the point of contact that front wheels have with the ground must be at least 1.62m away. This may make our base quite large.

 

Alternatively, we consider having 2 3060 aluminum profiles to act as the support tower. If we approximate the weight of the actuators and worm gear as 2kg and 0.7kg respectively and include them in our moment calculations:

 

With a support tower reinforced with another 1.5m 3060 aluminium profile, actuators and worm gears, we calculated that the front wheels must be at least 1.13m away from the support tower to prevent the lifting arm from tipping over. 

 

To increase the safety of our design, we can have base with front wheels about 1.25m away from support tower. The back wheels of the base can be about 0.15m away from our support tower. This effectively means that the distance between the front and back wheels of our lifting arm will be about 1.4m apart and the length of the base can be approximated to be about 1.5m long to accommodate for wheels placement.

 

The width of our base can be about double the width of our lifting arm at its widest point, with the front and back wheels placed just within the sides of the base. This should ensure that the CG of our lifting arm will be within our base, as well as provide enough breadth to place pneumatic units on the base.

 

Also, by balancing and adjusting the pneumatic unit on a 3060 profile, we approximated the Centre of Gravity of the pneumatic unit, as seen in the picture below. This centre of gravity will be kept in mind when we design the clamp system and the distribution of the clamps along the aluminium profile.

 

 

We also designed and 3D printed a mold for one of the attachments on the base, seen below:

 

Week 6 – 19 Jun: Craning our Necks

Lifting ArmLeave a Comment on Week 6 – 19 Jun: Craning our Necks

 

 

 

 

 

 

 

 

 

 

 

 

 

After multiple redesigns and dimensions adjustments, we have finally achieved a design that can seamlessly pick up loads on the trolley base and lift the to a height of 2.9m for installation purposes, in the working space behind the LED screen and near the top of the train door.

 

This particular design, as crafted in SolidWorks, includes a set of 2 actuators between the support tower and first arm, a double actuator configuration between the first and second arm, worm gear between the second and third arm and ball joint between the third arm and clamping mechanism.

 

Between support tower and first arm:

We have chosen to have two 5000N actuators to support the torque of the lifting arms. These two actuators will be able to support the weight of the arms and pneumatic unit at the (impossible) worst case scenario of full arm extension with a safety factor of 2.5. Our calculations can be seen in the image below:


 

Between first arm and second arm:

We have chosen to use a double actuator configuration for the arms to be able to reach maximum extension and elevation.

The first actuator in this unique configuration will extend and push an intermediary arm sandwiched (with pillow block joint) between first and second arm. The first actuator will give the intermediary arm some additional height.

Then the second actuator, upon extension, leverages on this intermediary arm height to push the second arm higher. We envision that this double actuator configuration will allow use to achieve consecutive and parallel placements of first and second arms. This allows us to maximise height reached with minimum aluminium profile length used, via the Pythagoras theorem.

 

Between second arm and third arm:

We will use a worm gear between second arm to the third arm so the third arm can have the freedom to rotate 360 degrees. Since we estimated that the vertical working space behind the LED screen to be about 40cm, our third arm will be about 50cm to have sufficient clearance from the top of the door. This third arm will be able to rotate to pick up the pneumatic units on the trolley base and rotate to angle the pneumatic unit such that it can be slid into its holding place easily. This third arm will definitely add to the ease of use and versatility of our design.

 

Between third arm and clamping system:

We will use a ball joint, preferably one with a locking mechanism, for greater maneuverability of clamping system (which is attached to the unit). This enables the clamping system and unit to be easily moved and held in place, even as the angle of holding place changes slightly from door to door. This likely improves productivity and speed of installation/deinstallation as small adjustments of the unit can be made easily, without the need to shift the whole trolley front and back or side to side.

Week 6 – 17/18 Jun: Trying to Lift

Lifting ArmLeave a Comment on Week 6 – 17/18 Jun: Trying to Lift

We considered some preliminary designs which use actuators (pictured below).

However, after doing some motion studies on our scaled paper model and some height calculations, we concluded that they were not the most suitable for our purposes of lifting a pneumatic unit from the ground and installing it at a height of 2.9m.

 

For the first and second design (from left), they were not able to pick the unit from the ground.

The first design did not have a long enough second arm and the actuator was placed such that the first arm is not able to go below about 45 degrees parallel to the ground.

For the second design, the placement of the second actuator between the first and second arm caused the arms to be unable to reach the pneumatic unit on the ground too.

 

For the third design, the design was unable to reach the required height of 2.9m due to the insufficient extension of second actuator. We will need to consider how to increase the reach of the second arm without resorting to cranes or longer actuator units (with lower and insufficient force outputs).

 

Week 4 – 5-7 Jun: Safety First

Clamping System, Lifting ArmLeave a Comment on Week 4 – 5-7 Jun: Safety First

On 5 Jun, our risk assessment finally got approved! YAY!!

 

 

To us, bastions of safety and wellbeing, this was a crowning achievement.

 

This meant the procurement of the safety paraphernalia. We went to look for safety boots online. We found some pretty-looking safety boots on Amazon. However, Tony, our experienced 5th to 10th member,said that safety boot sizes can be quite different from our normal shoe sizes, and recommended us to buy them in person. When we borrowed safety boots from SMRT for our Tuas Train Depot site visits, we also felt that the safety boot sizes are usually larger than normal shoe sizes. So we decided to buy them in person instead.

 

However, we faced lots of problems. Places selling cheaper safety shoes did not seem to have a retail or outlet store where we can try them out.  We discovered that Timberland sold safety shoes, however, they were really pricey, at about $200 for a pair. A whooping $800 (about half of MnT budget) to buy shoes for all 4 of us.

 

hmmmm, how?

 

Also, we are reconsidering our first lifting design which has an extruding arm and worm gear. Worm gears, those in our price range at least, are too weak to lift the heavy load of the lifting arm and the unit, with a maximum torque of about 60-100Nm.

 

Hence, we decided to use linear actuators. They are much stronger and can exert a lot more force at about 3000-5000N. This meant that the linear actuators is more likely to be able to lift the lifting arm and the unit (to be substantiated with calculations soon).

 

The drawback of using linear actuators is the decrease in the rotational ability (directly translated to mobility in our first design). To counteract this, we will use a configuration of 2/3 arms connected with linear actuators. To preserve the high mobility of the clamping system, we will still use a ball socket joint between the lifting arm and the clamping system.

 

 

Week 2 – 4 – 23 May – 3 Jun: First Flex

Lifting ArmLeave a Comment on Week 2 – 4 – 23 May – 3 Jun: First Flex

Based on the requirements and space restrictions presented by LTA and SMRT during the first depot visit, we went through a few iterations of possible crane design (as will be detailed in another post – insert hyperlink). As of 3 Jun, we sketched out our best conceived design on paper, detailed some of the parts necessary to built this model and thought through its working mechanism, pictured below:

 

 

 

 

 

 

 

 

 

 

We created this prototype model on SolidWorks too so we can use it to conduct motion studies in the future when we acquire more specific and precise dimensions, pictured below:

 

 

 

 

 

 

 

 

 

This prototype model features:

  1. Between the lifting arm and the clamp system, there will be a ball-socket joint between lifting arm and clamp system. This ball joint will be rotatable by 360 degrees to allow for free rotation and minor adjustments. These minor adjustments will be especially useful when angling the lifting arm to pick up the unit and when installing the unit into the working space behind the LED screen. This ball joint should also be lockable so that the orientation of the pneumatic unit can be safely locked during the sliding motion installation. Then, the technicians do not have to be concerned about holding the pneumatic unit up – they simply have to provide the slight jerk/push for the sliding in motion.
  2. Lifting arm will be a retractable single arm so that the length of arm is adjustable.
  3. Between the support tower and the lifting arm, there will be a worm gear (able to support the torque due to arm and clamping system) which can move the arm up and down. This will allow our future prototype to be fully adjustable and pick up units located almost anywhere on the base platform.
  4. Two support prongs to guide the technicians to slide our lifting unit into the train. We envision the support prongs to be flush against the sides of the door and the floor of the train.

 

Hopeful we can one shot one kill with this design! 😀

 

Edit: Famous last words… :”

Skip to toolbar