Team:Sheffield/Hardware


OPENLUX

OPENLUX is an affordable, DIY, open source microplate reader, designed to work at 600 nm.

OVERVIEW

Our device uses a white LED that emits light through an external 600 nm filter, to then be absorbed by a photodiode. It is currently set up to work only with 96-well plates. An ESP32 microcontroller controls everything from moving the motors to taking in data and displaying it.
TIME
3 s/ well
HEIGHT
250 mm
WEIGHT
<5 kg
WIDTH
400 mm
MANUALS
click here
DEPTH
320 mm

HARDWARE

Light Sources

For the light sources, two types are tested:

  • White light LED with external 600 nm filter
  • LED with an integrated 600 nm filter

Using a white LED with an external filter
The white light LED requires an external filter to let only desired wavelengths through to the sensor; 600nm in our case. Once the software was calibrated, we used the white light LED with an external filter. Having an external filter allows the user to swap out filters to fit the users' needs.

Using an LED with an integrated filter
The LED with an integrated filter does not need an external filter. This option reduces the chance of errors an external filter could introduce such as particulates on the surface of the filter, orientation, uniformity etc. Changing the wavelength is ,however, more complicated as the whole LED must be replaced. Custom external filters can be bought online a lot easier than custom LEDs with integrated filters, OPENLUX therefore uses external filters.

Figure 1 - White LED on the left, 600 nm integrated filter LED on the right

Motors and Movement

The movement of the device can be broken down into two platforms; upper and lower. These platforms are mounted onto linear bearings (3), moving across steel rods. The lower platform moves on the x-axis, whereas the upper one, on the y-axis. The plate holder is attached to the upper platform. The movement is driven by stepper motors (model 28BJY-48) and their driver boards (model ULN2003). In total, there are four stepper motors; two per platform. GT2 timing belts connected to idler pulleys facilitate the motion by the motors. 5 mm wooden dowels were used to mount the idler pulleys. Numbers 1 and 2 represent the X-axis and Y-axis motors respectively.

Figure 2 - Labelled x-axis and y-axis motors

Light Sensors

Choosing a light sensor
Unlike light sources, the type of sensor introduces different levels of complexity. For example, a phototransistor amplifies the signal that passes through it. This is advantageous for detecting minute changes in signals, however, noise is also amplified. Phototransistors also require a complicated circuit setup which introduces more potential errors in the system.

Testing phototransistors
Phototransistors were explored as a potential light sensor. Current leakages, fluctuating readings, and circuit incompatibility due to its transistor-like behaviour are all issues that arise when using phototransistors. To deal with a few of these problems, a low pass filter is added to the circuit to remove high-frequency electrical noise. The filter consists of capacitors in series with a resistor connected across the ground and sensor lines. We also tried reducing the noise through software code.

The main issue we run into when investigating phototransistors was consistent background noise. Even in complete darkness, the phototransistor still picked up a reading of approximately 220 points which reduces the possible sensing range of OPENLUX. This reading was likely due to current leakage.

Choosing photodiodes
Photodiodes are cheap, reliable, easy to integrate into a circuit, and cause very little noise. As we are aiming to make a device that any user can build on their own, photodiodes are ideal for OPENLUX.

Figure 3 - Phototransistor on the left, photodiode on the right

Casing

We lasercut the entire case out of two 600x900x5 mm white acrylic. The case is designed to slot together tightly and, epoxy glue was used to further strengthen the fit. The overall case can be broken down into the following sections:

  • Cover: a rectangular cube made of 5 pieces, (2 pieces of 400x255 mm, 2 pieces 320x255 mm, 1 piece 400x320mm) that goes over the device.
  • Electronics compartment: a box made for the device to sit on, and for the electronics to be housed in. This compartment is made up of a rectangular sheet of acrylic (400x320 mm) as the baseboard with walls on three sides.

To access all of the designs, please visit: https://2019.igem.org/Team:Sheffield/Contribution.

ELECTRONICS

System Controller

Choosing an ESP32
The white light LED requires an external filter to let only desired wavelengths through to the sensor; 600nm in our case. Once the software was calibrated, we used the white light LED with an external filter. Having an external filter allows the user to swap out filters to fit the users' needs.

Figure 4 - ESP mounted on a breadboard

Connectivity
The ESP32 hosts a standalone wireless access point (WAP, often called a hotspot) that makes it possible to configure the device and get it online, all through the user’s browser. The ESP can be set up to connect to local WiFi allowing access to the OPENLUX user-interface from any device connected to the same network.

Figure 5 - Any device with a browser can connect to OPENLUX

ESP Outputs
The multiplexed IO of the ESP32 means the controller is capable of using one physical pin for many different purposes. For example, the ESP32 has two onboard analog to digital converters (ADCs) that provide between 15 and 18 analog pins to the user. These ADCs also have a configurable attenuation, so the voltage range of the measurements is configurable. This allows for a 0 - 1.1V signal to be read as 4096 discrete values, whereas an Arduino is only capable of 1024 values between 0-3.3V, effectively resulting in only ~300 discrete values of precision.

Operational Amplifier

The signal collected by the photodiode from the LED is weak so the photodiode outputs a small value. This reduces the effective range of the sensor. The operational amplifier (LM358) was used to amplify the signal and, therefore, increasing its effective range. Small changes in the sensor reading without the amplifier would not be registered, with the amplifier they can be.

Figure 6 - Operation amplifier in the circuit

Shift Registers

Each motor controller needs four input signals. As the motor pairs (top and bottom layer) drive the belt in the same direction, the motor pairs can share the same input signals. Therefore only eight data lines are needed to drive all four motors. The ESP does not have eight output pins so a shift register (74hc595) is used. The shift register takes three input signals from the ESP and outputs eight signals to the motor controllers. Only three pins on the ESP are therefore needed to control all four motors.

Figure 7 - Shift register (74hc595)

Motor Driver Boards

The stepper motors used to move the platforms have 2038 steps to allow the precise alignment of the plate, LED and Photodiode. The driver boards step up the voltage output from the 5V mains to 12V to drive the motors sufficiently.

Figure 8 - Motor driver boards (uln2003)

Power Board

The ESP32 and motors both take 5V. The current system uses a 5V 4A mains power supply connected to a power supply module. The module provides power to the motors. The ESP32 is powered by USB off of a laptop or power bank.

Figure 9 - Power board

HOW TO BUILD

To get access to the designs, parts list, and instructions necessary to build OPENLUX, please visit: https://2019.igem.org/Team:Sheffield/Contribution.

Here is an interactive 3D model of our device:

PRODUCTION

Endgame: Create an Absorbance-only microplate reader

Phase I: Develop a minimum specifications product

  1. Redesign BrightBiotics, and focus on 600 nm for bacterial growth.
    https://2019.igem.org/Team:Sheffield/Improve
  2. Develop and validate movement and detection separately.
    https://2019.igem.org/Team:Sheffield/Demonstrate
  3. Combine and validate movement and detection.
    https://2019.igem.org/Team:Sheffield/Demonstrate

Phase II: Develop a minimum viable product

  1. Approach potential customers for prototype feedback.
  2. Design and test a filter wheel.
  3. Figure 10 - 3D CAD design of the filter wheel
  4. Combine and validate movement and detection.
    https://2019.igem.org/Team:Sheffield/Demonstrate
  5. Figure 11 - 3D CAD design of the shaking mechanism