Micro Capacitive Sensor

The goal of this project is to develop a printed circuit board (PCB) that utilizes capacitive sensors and an off-the-shelf capacitance-to-digital-converter (CDC) to detect and monitor micro entities. The monitoring process is an autonomous process that displays a plot of real-time capacitance values.

Background
The counting and classifying of particles on the micrometer scale is expensive if automated and difficult if done using non-automated methods. An inexpensive automated counting and classification method for small particles needs to be created to lower the cost and increase the ease of use for this process.

Deliverables

 * PCB designed with CDC chips and multiple capacitive sensor designs
 * An easy to use graphing method that accurately displays capacitance vs time on a PC from the CDC chip
 * A capacitive sensor design, or multiple designs, that successfully detects micro-particles

Assembly Operations/Schematic
The overall system layout will consist of a capacitive sensor (shown as a capacitor in the schematic) that is read by the AD7746 capacitive to digital converter chip. This chip will be interfaced with a host Arduino microcontroller to autonomously gather data. The communication between the two utilizes I2C communication protocol depicted as SDA and SCL.



In the preliminary stages of system development we are using an off-the-shelf development board that hosts its on microcontroller and AD7746 capacitive to digital converter. This will be used to test the PCB sensor design to ensure accurate functionality.



Design Process
The AD7746 CDC chip is used on this project for its high resolution capability at 24 bits. High resolution is necessary in order to detect capacitance changes for the smallest micro-sized particles, which is possible for changes as small as 4aF. The CDC communicates the measurement data via I2C communication protocol to a host micro-controller system. The AD7746 development board was utilized to setup quick communication and display measurements for prototype sensor designs.

The capacitive sensors can be designed in a number of different configurations. The figure below shows a co-planer and inter-digital design. For the purpose of our project, the co-planer design was chosen so that we can utilize the "channel" down to the middle to fasten a microfluidic device.



Placing the microfluidic device over the capacitive sensor allows us to pump a sample over the sensor. As particles pass through over the sensor, the measured capacitance value will change.



Utilizing the co-planer design template, four general sensor designs were considered. The designs consist of aligned fingers, offset fingers, mismatched offset fingers and parallel plate (no fingers). To further test how layout impacted sensitivity each of these designs parameters such as channel width, number of fingers, finger width and length and finger spacing could be altered. Varying these parameters may change the E-field between both plates and can cause certain sensor designs to be more sensitive and accurate than other designs. The purpose of the testing process is to find which sensor design will be the most responsive to micro-particles passing through the "channel".



Testing Process
The sensors can be first tested using the microcontroller on the AD7746 Evaluation board. The picture shown (left) displays this configuration where the evalutaion board microcontroller interfaces with the prototype PCB and mounted CDC chip. Once a suitable design for the sensors is chosen, the next step will be to interface the on PCB AD7746 with our own microcontroller that will then display the capacitance values onto a monitor.



Initially, particles such as table salt (pictured), lemon pepper, and sodium bisulfate (pictured) were placed one particle at a time directly onto the sensor’s surface and the capacitance was read from the AD7746 development board software’s capacitance vs. time graph. The goal was to try and see if one sensor design stood out in terms of its ability to sense micro-sized particles. Since the particles varied in size greatly for some of the materials, this setup was deemed to be insufficient due to the lack of consistency. If the particle sizes vary, the particle size difference could have accounted for higher jumps in capacitance per particle added to each sensor.



Currently, an optical setup (pictured above) is being used that has multiple stages that are screwed down onto a stable tabletop. A tungsten needle is suspended over the printed circuit board that is directly under the microscope and the stages are used to move the needle horizontally and vertically. The tungsten needle is used to place particles on so that one particle can be used to test each sensor, which makes the capacitance measurements more reliable. Clear nail polish was painted over the sensors in an attempt to try and protect the aluminum sensor fingers from particle debris, and testing is ongoing to see if it compromises the characteristics of the sensors too much to consider it as an option.

The next step in testing the prototype printed circuit boards is to use a micropipette to place even smaller polymer bead particles in the channels of the sensor designs. This way, the sensors will be tested with particles that are of the size that is necessary for the eventual final medical counting system (around the size of white blood cells, etc.). If there is a noticeable jump in the capacitance measurement of the sensor once the small microbeads are placed in the channel, that sensor design will be a viable option for the final printed circuit board design.

Spring Semester Timeline
The timeline for the spring semester will consist largely in testing the PCB design and debugging. The end of Fall semester is focused on the design of the prototype sensor design so that by mid Spring semester a final sensor can be manufactured and interfaced with the overall complete system.