Author Archive

Compressor up and running!!

March 23, 2014 Leave a comment

Today we had another successful day.  Getting our 60 Gal Craftsman compressor wired and plumbed.  A new drop is install in the middle of the garage and we are ready to use all the air tools!!!!

Bennett also stripped the blown LT1 and found out we ruined 2 pistons/ cylinders so that engine is useless.  We will be making it into a table!

Also we are up 1.5 brooms

Categories: Uncategorized

Saber Board PCB Design

March 28, 2011 Leave a comment

In an effort to properly interface with the National Instruments SBRIO both mechanically and through software, the team developed a PCB.  The PCB was designed mainly as a breakout board, but also incorporated operational amplifiers for sensor inputs.  The PCB design was developed in the EAGLE Layout Editor utilizing two of the 50pin connectors on the SBRIO.  One of these connectors was primary used for Analog Inputs and the other was used for Digital I/O.  In addition the board was designed to have a 5v input rail in order to provide the voltage necessary to run the many sensors attached to it.  The final schematic for the Saber Board can be seen below.

With the schematic complete, it was necessary to create the footprints for the connectors which would be used.  We used Digikey PN # 67-1062-ND to interface with the 50 pos cable from the SBRIO and PNs #455-2248-ND/455-2219-ND to have secure three pin molex style connectors.  With these connectors the board layout was created and the programs auto routing feature was taken advantage of.  While the auto routing feature was not perfect it did save some time.   Two of the most important things that were done to the auto routing were to eliminate all sharp angles and to create both a ground and power plane on the top and bottom of board respectively.   These planes also served to help reduce both high and low frequency noise on the system.  The final board layout that was compiled can be seen below, the red lines represent traces that were placed on the top and the blue lines are the traces on the bottom.

We printed our boards with 4PCB and couldn’t have been happier with the return time.  We had our boards in head less than a week later.  The boards were then assembled and tested as seen below.

While we were working on this PCB externally to the SBRIO, there was also a development plan for the LabView code which would interface with the PCB.  The VI for this interface was setup to mimic the physical appearance of the PCB with the same naming scheme to ease the overall interface between the PCB and SBRIO.  This will allow future calls to the I/O pins to be intuitive and reduce the amount of errors.  The VI can be seen below and as you can see has the same row/ column layout as the physical PCB.

MicroStrain, Inc. Provides us Inclinometer

October 10, 2010 Leave a comment

The team is happy to announce a new sponsor joining our team, MicroStrain, Inc ! MicroStrain, Inc. makes tiny sensors that are used in a wide range of applications, including knee implants, civil structures, advanced manufacturing, unmanned military vehicles, and automobile engines.  Their inclinometer product line was initially developed to measure angles of limbs to help re-animate the limbs of paralyzed individuals.  We thank them for the inclinometer that they are providing for us, which will allow for robot location and orientation to be determined.

Categories: Uncategorized

Mechanical Subsystems Research Continues

June 23, 2010 Leave a comment

This weekend we took a more detailed look into the different mechanical subsystems of the robot.  Our meeting consisted of a detailed analysis of research pertaining to the individual subsystem as well as the consideration of some concrete designs.  The requirements and restrictions of each system had to be approached and a few design constraints needed to be modified.

Initially we wanted to focus on the simpler systems which could be explored without large amounts of external mechanical references.  These include the foot/ ankle system, the second DOF of the hip and the belly motion device.

The first system we approached was that of the foot/ ankle.  This system has been focused on prior in order to understand the importance of being able to control foot pressure.  Initially we focused on the ways in which we could distribute spring force in order to cover the bottom of the foot.  This dilemma allowed us to create two interesting solutions.  The first of which was to mimic that of a dog/ goat and have “toes with pads” that would be hinged to the rear contact point.  The research proved that the ability to have motion in the foot allowed for better control of the animals CG.

The thought process then shifted to focus on the motion in the foot allowing for better control of CG.  We discussed that even though many animals have motion in the feet it is not the ideal amount of motion.  Now set on expanding possible motion we thought about mimicking that toy that has many pins that you can put your hand in and it will retain the shape.  This idea was talked about and we decided that we want to explore both options of an animal paw and a 3×3 matrix of pins.

However, this discussion quickly shifted to the amount of spring force which would be required and how we would “stop” the motion of a spring in order to gain control.  This discussion went on and the need of this motion was then questioned.  The challenge came with the motion of a foot through a stride and how the springs would get compressed each time.  The group decided that we needed to still do more research as to hole feet interact with uneven ground.

The next system that was discussed was that of the hip.  This system was needed in order to allow for the legs to have camber on the body or add a second rotational DOF.  This will allow for balance via gates as well as help to reduce body sway.  However, the large issue discussed was being able to support the entire leg assembly and still accurately control the angle that he leg makes with the body.

There were two ideas which were discussed fully and can be seen below.

These systems approach the altering of the angle of the leg differently, one pivoting the angle from the body and the other pivoting the leg upon itself.  In the first system the leg would be attached to a pivot point on the body and would rotate on the arc of that pivot.  This would allow the angle of the leg to be modified, but it would also require the leg to rise/ lower in order to accomplish this.  The second system focuses on altering the legs angle by rotating it around the contact point of suspension arms.

Each leg would be attached to the body at 4 points corresponding to two pairs of suspension bars.  These bars would be able to extend/ contract in order to push/pull the leg and create a torque that will alter the angle.  This design is very unique in that if the push/pull action is controlled by cables then the top and the bottom bars could be linked if they are attached to the same motor.

This drawing shows how the bar pairs will be linked on the top and the bottom.  The linear guides will allow for smooth motion and the cables will reduce backlash.  In addition both compression and extension springs will be introduced to the top and bottom pairs respectively.  This will allow for the legs to have a hip “suspension” and will allow for some more of the uneven ground and vibrations to be absorbed.

The next system that was examined was that to of the belly motion device.  First the design requirements for this system were examined.  It was determined that this system was to have two DOF and that one would be mechanically controlled and regulate while the other will simply be a suspension system.  The mechanical device that will allow this motion will most likely be loosely based off the Roof Robot MQP center joint.  However, this system will be much heavier and will require all of planning to incorporate suspension.

The initial thoughts is to control the angle of the body with a motor chain system linked directly to the joint and to have the torsion of the body be restrained by suspension.  This suspension system will probably be based off of fiberglass plates and will have the appropriate slots in the body in order to still provide suspension even when the body is not locked in a straight position.

After discussing the simpler systems we moved onto the more complicated systems of the leg control/suspension and the CG adjustment device.  Each of these devices rely heavily on the rest of the robot, but also need to be very carefully evaluated in order to ensure the desired results.

Through our research we came across the MABEL robot and how it utilizes springs to regenerate forces and apply them toward the next stride.  While this robot is a biped, it is our hopes to apply the knowledge gained from this research into a quadruped that utilizes springs to regenerate some of the force and apply it to the next stride. We spent a few minutes discussing how much spring force we wanted to duplicate and it we wanted to make the robot “super human” by adjusting the K values. This system is still in the research phase, but should move onto design within the next few weeks.

The group has been continually throwing around very unique ideas in regards to aiding in the control of the CG of the robot, outside of just gates.  This idea focused on a tail which would automatically based on the mechanical state of the system move to adjust the CG.  The initial thought was that we could extend the cables, that will be driving the joint, in order to control the tail.  However, the problem which was discovered is that the cables cannot make 90 degree turns into different planes than what it is traveling in.  It is our belief that this idea is improbable, but we will continue to research this concept.

Categories: Mechanical Aspect