Hello,
I am working on a high volume liquid level sensor. The PCB will be vertically mounted in a liquid tank and housed in plastic. PCB real estate on the submerged section of the board is limited and requires 4 sensing pads. I will be placing a reference pad out of the fluid to see if that will be helpful. The key driver will be performance followed closely by current draw and cost. Sensing response time can be slow. I’m not sure what level of “slow” is required to achieve the average current draw we need but let’s use 1 -2 seconds as a starter. I would like to get to an average current draw of 10 – 15uA or less. My first step is to design a board that will export the data to a PC via I2C. The board is 90% complete I just need to review some details with an expert. I know there is an eval board with similar features but as time is short I need to move forward with our design for more accurate results in our environment. My goal is to conduct testing under several different conditions both on the PCB (frequency, sleep time, etc.) and hardware conditions (residual water on housing when empty, residue build up on housing, etc.) to make sure we have reliable sensor.
I have reviewed the application notes on the TI website. Unfortunately, I cannot afford the new capTIvate system. I read up on the RO and RC methods for capacitive measurement. The current IC selected for this project is the MSP430G2102IPW20. This was selected mainly for price. I understand this part does not have the comparator required for the RO method. I have identified the MSP430G2452IPW14R I used on another project which has the comparator at a slightly higher price. Can you please let me know your opinion on the following questions?
- Will the RO or RC method provide better/significant sensing results in my application?
- Will one of the methods allow for better/significant current savings?
- Would you recommend using the MSP430G2452IPW14R for the first prototype and testing both methods?
- I want to confirm that either method can completely be shut down in sleep mode to save power.
- The hardware design seems very simple for the RC method. I am a little confused about the RO method. The part I question is the three resistor array at the input of the + side of the comparator which includes a pull up, pull down, and series resistor to the sensor pad. Can you provide some additional explanation and a real life reference/eval design example?
- Any other MCUs I should be looking at?
- Any other design recommendations or example circuits you could send my way.
Thanks for the help!
Mark
