I'd like to request a review/bashing of this circuit for a high precision soil moisture sensing and analysis device. Where did I go wrong?

Question

Below you will find a buffet of questions and the schematic for my device. (Click on it and you will get the full-size image.) Please tear this up. I'm a self-taught novice putting together a high-precision device. Can you help me validate the circuit?

• There is a pull-down resistor (R15) at location B7 on the schematic, I need help calculating the value.
• I have access to SPICE, would this kind of circuit be suitable for simulation?
• Where have I failed to follow standards?
• Which capacitors are bypass, blocking or coupling?
• The ADC and the LED driver both communicate with SDI/SDO. Can I serve both components on the single input/output of the development board? (Location B8.)
• Can I use different color lines to represent signal and communications?
• I need help calculating the voltage divider for an output that is 1/4 of the input (R17/R18 at location B9).

The component labeled soil probe is under development. Essentially, at this point in time, I am using the sub-circuits at the left side of the schematic to create two separate precision current meters where the load is the dirt. I got the circuit from page 22 of the op-amps datasheet.

Datasheets for all the major components follow:

• MOSFET
• Operations Amplifier
• Low-Pass Filter
• Analog/Digital Converter
• Real-Time Clock
• Arduino Nano Every
• LED Driver
• Voltage Regulator

Answer 1

• C1 definitely shouldn't be there. It disconnects V+ pin from the actual power line. Probably you meant to place it between V+ and ground
• Crystal resonator is supposed to be connected between X1 and X2 pins, not between X1 and the ground.
• U5 probably needs its own crystal between OSC1 and OSC2 (or clock source, check with datasheet)
• If R1 and U1 are meant to form a voltage divider, R1 value is too small. (same goes for R4) Proper value should be roughly the same order as U1 resistance in normal operating conditions.
• Not clear what R15 "pulldown" is referring to - VCC of U6 or SDI of U5. In any case, neither of those need a pull-down.
• In general, the situation with op-amps and MOSFETs is not too clear. Are you trying to achieve voltage repeaters? Or amplify the signal? This can be done a lot simpler without the need for any transistors and will also produce cleaner signal. I don't see any reason why soil probes should be its own circuits (as for your update) If they are just some kind of variable resistance that seems to be enough here. (Only with proper R1 R4 values)
• SDA / SCL lines are not connected correctly. You need SDA/SCL lines going directly from arduino to U6 and both of then pulled-up independently to VCC.
• SPI bus seems to have 2 devices connected but no way of actually switching between them two. (CS pin is hard-wired to ground) This is related to your question #5. If you need to connect 2 devices to a single SPI bus, you need to wire their CS pins to your arduino GPIO lines and use that to switch them at runtime.
• SDI/SDO pins are connected wrong way around to Arduino. SDI should go to Arduino's MOSI and SDO to MISO.
• Getting a 24 bit ADC doesn't automatically make your circuit 'precision'. You need to take a lot of care with power isolation and smoothing. Simply sticking power line info VREF is definitely not enough. So either your ADC is too powerful for this task and you don't really need all that precision or you need to properly insulate analog circuitry from the digital noise. That's a big topic on its own.

"Which capacitors are bypass, blocking or coupling?" - is kinda odd question to ask. It implies that you first randomly slap a bunch of capacitors on and then ask what is their purpose?

Conclusion (based on comments and other's input): Just use Arduino and plug your sensors directly into it using proper R1/R4 for voltage dividers.

Answer 2

In addition to what everyone else mentioned, all the filtering you need will be accomplished by the ADC itself, and by post-processing the data in software. As a starting point, I'd consider connecting the ADC inputs as "directly" to whatever it is you're measuring as possible - this will remove a whole bunch of design risk. If, in the end, the sensitivity is too low, you can always add external amplification - that may improve S/N ratio but does so at the expense of other parameters.

The reason I am seeking precision is because I expect to need to accurately measure very low currents

The number of bits determines dynamic range. What is the dynamic range of currents you'll be measuring?

All you need in your application is maybe 1% accuracy and 0.1% resolution. A 12-bit ADC is plenty for that, if the range is fixed.

Soil impedance measurements allow for error correction when done across a range of frequencies, and DC is a bad idea due to galvanic effects. When using DC, in order not to degrade the sensing cell, you may need to drive a certain amount of charge into the cell, then reverse polarity and extract that same charge, and so on. The net charge after a measurement session should be zero. This should be fairly precise, as the errors accumulate, and will cause a DC drift. That's why AC impedance spectrum may be less of a chore to obtain.

Answer 3

I see many probable errors as well as questionable design implementations. Most have been noted in other answers and comments.

But in order to fully critique the design, you will need to provide a complete specification and theory of operation, which should include all non-obvious circuit elements. Maybe the most important is what is being analyzed (soil moisture, apparently), and why it needs to be such high precision as to require a 24 bit ADC. AFAIK soil moisture is highly variable and depends greatly on the exact construction of the probes and how they are inserted into the sample.

You show an LT1475-5 buck regulator, which is fixed at 5V. You might want an LT1476, which has an internal 2.4V reference. Your feedback network R19 and R20 will regulate to about 5V. It also needs to be connected to the output (C3). It also needs a bypass capacitor close to the device. And you don't show the voltages of your battery packs. You can use LTspice to simulate this circuit element. I think you are using it for a 5V supply, so the feedback network is not needed.

As for using two aluminum plates for moisture sensors, you should consider that they will quickly become coated with a layer of aluminum oxide, which is an insulator, and will probably not function well for resistance measurement. They should probably be made from stainless steel, or platinum, or gold plated. Commercial probes seem to be made of stainless steel. Some material to review:

https://extension.umn.edu/irrigation/soil-moisture-sensors-irrigation-scheduling#tensiometers-1870360

https://dynamax.com/products/soil-moisture/pr2-multi-depth-soil-moisture-probe

https://extension.okstate.edu/fact-sheets/soil-moisture-sensing-systems-for-improving-irrigation-scheduling.html

And here is an Arduino project with a "dirt simple" soil moisture sensing circuit. Perhaps you should try this and get some experience before building a highly questionable device.

https://www.wellpcb.com/how-do-soil-moisture-sensors-work.html

https://sensors-technology.com/qa/what-is-soil-moisture-sensor-and-its-uses.html

Answer 4

It looks like you’ve connected the soil probe to 3.3V from your Arduino. That’s likely to carry a lot of digital noise. Ideally you’d use a dedicated low noise voltage reference, or at least isolate using a small inductor in series with decoupling caps on either side. You have a 0.1R pull-up resistor which seems much too small, although the ideal value will depend on the dimensions of the probe, aim to have the probe’s output be at about half of your supply voltage under normal conditions. You might also want to switch the voltage to the probe, if you supply DC then there will be electrolytic effects such as corroding the probe and turning the surrounding water into gas.