Plant Science, continued

Well, I was hoping to give y'all an update about the science fair project I was hoping to help my sister with. I recruited a very talented friend of mine to help build the sensor/amplifier interface that would allow my sister to use my labjack to gather CO₂ concentration data from the plant sample container.

Kevin and I actually had the sensor working relatively well. I think it was still breaking in, even after about 48 hours of operational time. We logged some data from it over the course of about 36 hours as it just sat in a room, and the voltage it gave off varied dramatically over that time; it appeared to be an exponential approach to a steady value.

Qualitatively, the sensor behavior was what we expected; an increase in CO₂ concentration yielded a drop in voltage, with the voltage drop increasing inversely with concentration. We tested this relationship using our breath, as well as a soda bottle filled with vinegar and baking soda.

Unfortunately, something went wrong between the time we shipped it to Missoula, and the time my sister and my dad finished assembling the device and interfacing it with the labjack. The voltage divider we chose for the photosensor appeared to have been poorly designed, as even mild darkness would peg the labjack reading (the labjack analog inputs max out at 2.44 volts).

I went down to Missoula to try and fix things, but I think it's possible I actually ended up being the one to finally destroy it all. I hooked up the OPA-111 amplifier, but I fear that I may have reversed the voltage rails; the positive rail was supposed to receive about 14V, and the negative about -6 (relative to ground, of course).

I think I may have reversed the connections, which fried the opamp; I actually saw the smoke escaping from it, as well as the insulating tape we'd carefully wrapped around it begin to melt. So yeah, that got fucked.

This heat, I think, also destroyed the CO₂ sensor, because afterwards there was an infinite resistance across the heater terminals- implying that the heater circuitry was fried. The reason it was possible for the opamp to fry the sensor was proximity: in order to minimize noise (which is a very real concern when you are dealing with currents that are less than a trillionth of an ampere), we mounted the OPA111 directly to the sensor pins.

Unfortunately, this arrangement means that when I messed up with the amplifier wiring, I also killed the sensor and now I have nothing to show for all of my (and kevin's) hard work. It's really quite frustrating, because I was looking forward to seeing the kind of values that Shannon recorded, as well as maybe trying some scientific trials of my own.

Honestly, I'm frustrated enough with myself that I feel compelled to buy another sensor and opamp, just to prove to myself that I can't make such a sensitive circuit actually work. It's not like it was overly complicated- the OPA111 was simply acting as a non-inverting voltage follower; this design was chosen so that we could make the cable between the sensor electronics and the interface electronics fairly long, without having to worry about noise. Any other signal conditioning we could do with a cheap Radioshack opamp package.

I don't really have much more to say on the subject for right now, other than that I'd like to get the setup working someday. One of my plans for this summer is to begin establishing myself a hydroponic garden; it would be interesting to measure the effect of CO₂ and humidity on plant growth (I'm going to be aiming mostly for strawberries and tomatoes).