Engineering Design Guide (Science Module) Part 1

CAD drawing of science module box
Science Module Half Section View

This blog is going to introduce you to engineering design by providing a step-by-step guide, using the rover science module as an example. I (Nora) will walk you through the functional requirements (FR), brainstorming with morph chart, design parameter (DP) selection and visualisation.  My group-mate Jessica will explain the electrical component of the science module, this includes the printed circuit board (PCB) which connects all the power systems, main computer and module sensors.

Introduction of the science module

The science module is the segment that collects soil samples and conducts scientific monitoring under air-tight conditions. For the European Rover Challenge (ERC), we are required to collect three soil samples at three specific locations and conduct scientific analysis and data-processing in its original condition.

Science module side view

Step 1. List All the Functional Requirements

flow chart about the functional requirements of the science module
Functional requirements diagram

The first step of engineering design is to list out the requirements of the project; what problem is your design solving and how. The main requirements of the science module are to collect soil samples and analyse them. These can then be further broken down into specific solutions to the problem. In the graph above, the two main functions of the science module are then broken down into smaller, more specific problems. The more clear and unambiguous the requirements are, the better.

Step 2. Brainstorm with Morph Chart

a table of drawings of different aspects of the design, claw, box etc.
Morph Chart

A morph chart is a visual guide that that compares different solutions to the design problems. In the morph chart above, each design requirement is given multiple solutions. The morph chart allows us to easily compare the advantages and disadvantages of different solutions. As an example for FR1.1 (scoop 50-150g of soil) multiple types of scooping buckets are available; each with its own pros and cons.

Step 3. List All the Constraints for Each FR

FR 1.150-150g of
 soil sample
Arm degrees of
Arm motor
Air pressure
FR1.2Motor torque
SealingSensing space
 and durability
FR1.3.394mm wide module
top lit
110mm* 130mm*85mm
internal space
Container cannot
touch the wall(load cell)
FR1.3.4Sealing110mm* 130mm*85mm
internal space
Sensor space and durability

Every design has limitations to its implementation. These can include cost, space, power and accuracy. By listing our design constraints we have criteria to select the solution of the functional requirements.  For instance, FR1.1
(scoop 50-150g of soil) is limited by the mass of soil it has to carry, the rover arm properties and vacuum equipment.

Step 4. Design Parameters Selection & Sort the DPs into Different Sub-modules

From the morph chart, we can see that the drill sample and main samples require quite different treatment. Hence we put the component for drill sample and main samples into different sub-modules. In this blog we only focus on the main science module for main samples.

FR1.1 scoop soil sample:
a) Arm: Scooping highly depends on the arm. The current arm has 5 degrees of freedom but isn’t very good at wrist movement (it has difficulties doing the scoop movement because the claw is always facing the same direction; down). Hence reject DP 1  and DP3.
b) Torque: We are using linear actuator for the current arm, which provides adequate torque. From this perspective both  DP2 and DP4 works. However, the sealing method of DP4 sliding lid mechanism is not realistic when the device surrounded by soil. Hence reject DP4
c) Vacuum: To create a vacuum, we need a fan or a combustion reaction. Both plans consume a lot of space. A fan to create vacuum will require high power, which we might not able to provide. Also, it is unsafe to trigger a combustion reaction in the rover, hence we also reject DP5.
d) Arm integration: The arm, in its normal state, has a claw (as shown below) and in DP5, the pincher claw design can be easily integrated into the current arm

Our Mars Rover on a field of green grass
Normal claw on the arm of the NUMBAT Rover

FR1.3.3 Store main soil samples in separate containers under a sealed condition:
a) Space: In DP3, the rotation of chambers about the axis which passes through the edge of the circular chambers will take up a lot of space hence reject DP3.
b) Sealing: Sealing for DP2 would be difficult since the holes are linearly distributed.
c) Inspirations: The design of DP1 is inspired by the design of a toothpick holder, a chamber hole will only open when the hole on top lid align with the chamber hole.

FR1.3.4 Switch between different chamber:
The corresponding chamber switching method for DP1 is rotation.

For all the FRs in FR2, since a camera requires a focal length and the module box will be difficult to fit in both the sample box and the camera. Additionally, transparency of our resin-printed sample box is not enough for camera processing. Thus we have concluded that these sensors are likely to be better options to analyse the soil samples:
– Colour (RGB sensor)
– Humidity (soil moisture sensor) 
– Temperature (temperature sensor)
– Magnetic field strength (magnetometer)
– Sample mass (load cell)

Step 5 is to be continued in part 2 of this series

Electrical Design

The role for the electrical component of the science module is to connect all of the sensors used to measure the characteristics of the soil sample.

Originally, the science module contained 4 types of sensors:

– Magnetometer to determine if the soil had any magnetism (which aids in determining the element makeup of the soil)

–          RGB sensor (I2C) to determine the colour of the soil.

–          Moisture sensor (Analogue) to determine if there is any water content in the soil.

–          Temperature and humidity sensor (I2C) to determine whether there was a change in temperature between the soil and the atmosphere.

The science module has 4 different compartments to collect 4 different samples of soil without contaminating each sample. Each compartment has 4 sensors resulting in a lot of sensors and a lot of wires. To reduce the number of wires for the science module, a PCB was designed to work as a connector board that the sensors could slot into. The schematic (Figure 1) and PCB layout (Figure 2) is shown below.

Figure 1. Original Schematic of the Science Module

Figure 2. Original PCB design for the Science Module

The science module then underwent some revision. The magnetometer was not included in the final version of the science module as it is impossible to obtain an accurate reading as there is a lot of interference caused by the rest of the rover.

Additionally, the 12V to 6V DC converter was also removed from the science module. The converter was originally required to convert the 12V from the generic PCB into 6V for the science moduleā€™s stepper motor. However, as the generic PCB is undergoing a revision to include a 6V output, the converter is no longer required on the science module.

Many of the extra connectors for the original design was also removed resulting in the much simpler design seen in Figure 3 and 4 below.

Figure 3. Final Schematic for the Science Module
Figure 4. Final PCB design for the Science Module

Each compartment in the science module will have 1 PCB (totalling 4 PCBs for the science module). The I2C data line and the analogue signal of each PCB will be fed into 2 multiplexers (one for I2C and the other for analogue signals) in order to choose the compartment of soil to read from. The data, analogue signals and clock lines will then be connected to the generic PCB allowing for information to be communicated between the NUC and science module.

The role of the electrical component of the science module is to connect all of the sensors used to measure the characteristics of the soil sample.

By Nora Deng and Jessica Li. Special thanks for Anita Smirnov and Garwerd Liang.

Author: Nora Deng

Nora Deng is science squad lead of BLUEsat Mars off world robotic team, BLUEsat general manager and a mechanical engineering student in UNSW.

(2) Comments

  1. I find although I have…somešŸ˜Š….technical background….much of what your doing is a bit difficult to comprehend at my age….80….it often reads like science fiction to me…..but fascinating…..continue to keep sending info on FB. ……I read all5the details…..would like to sit in on your meetings šŸ˜„

    1. Thank you for your ever encouraging support, Peter! We love the fact that you like to keep yourself up to date with what’s going on, and we are more than happy to add fuel to the furnace by writing our weekly blogs!

Leave a Reply