This post is a follow up to this post – where I dissected SparkysWidgets’ minieC circuit.
Thanks to Those that Go Before
While this is a short post, It is important to thank exceptional folks whose sharing of their knowledge has greatly benefited my learnings.
- @SparkysWidgets – Thank you for your help. You have been inspirational. if you are interested in this stuff, PLEASE check this person out! I will be borrowing heavily from the minieC product and schematic.
- The person behind this wonderful post on the electronics behind an EC sensor.
- Chris Gammel – our very knowledgable and gifted instructor of Contextual Electronics. A course I would highly recommend if you are interested in turning your prototypes into PCBs and learning tons about electronics along the way. The more time I spend with Chris, the more I realize what an exceptional person – both in knowledge and in ethics – he is. Chris is amazing at clarifying circuit concepts!
- Mike Engelhardt – thank you for LTSpice. What an exceptional contribution to circuit designers!
Follow Up on minieC Readings
At the end of the post I was curious about the waveform of the digital signal that comes out of the last op amp and then read by the ADC. I thought it should look like this:
but it looked more like this:
I brought this up with Chris during our Google Helpout (note: I do not get paid for endorsing Chris – quite the opposite – I get a lot of value from Chris and hence pay him – you can sign up to talk with Chris via a google help out. The link is here).
Chris asked that I try several values for C7 and then several values for R10.
I tried 10pF, .1µF, 1µF, 10µF, and 50µF. I ran the simulation twice. The first time R0 – the substitute for the E.C. Probe – was set at a resistance of 200Ω. This is about the resistance I would expect for a tomato nutrient bath. The second time the resistance of R0 was set at 1000Ω – what I would expect for the nutrient path if the plant was lettuce.
Here are the results when C7 was set at 10pF, 1µF, and 50µF:
R0 = 1000Ω
As expected, the waveform smoothed out as the value of the capacitor got larger. The 1µF capacitor is what is used in the minieC design. The 10pF looks too noisy. As R0 got larger, the smoothed out voltage value began to give a larger value. Given this, 1µF seems like a “best” fit for this design scenario.
The next LTSpice simulation I ran was to keep the capacitor at 1µF and plot R10 values for 100Ω, 1KΩ, 10KΩ, and 100KΩ.
The first image set R0 at 200Ω (closer to the tomato range):
R0 = 200Ω
In the second simulation I set R0 to 1KΩ – what I expect for lettuce:
R0 = 1KΩ
Given the results, it looks like Sparkyswidgets’ choice of 10KΩ for R10 makes sense.
That’s it for now. A simple update to share my LTSpice results to determine the “best” values for smoothing out the DC current going into the ADC without smoothing out the value too much. It looks like the choices made were the right ones.