Seeking sensor advice for environmental control (Temp, RH, CO2)

Projects General Guidance
1

This is my first post. I am excited to be here.

My final goal is to create an environmental test chamber for plant propagation by retrofitting an old refrigerator. Any comments, questions, advice, or resources are welcome, although this is not specifically why I am posting. Eventually, I would love pH and EC monitoring, and a whole host of other features including remote data logging and control, but the prototype will be basic to see if I can get the fundamentals achieved: control over temperature, RH, and CO2 levels, within a band.

I have seen The APDuino Project and it seems like a lot of what I want, but I don't think the sensors used in this build are accurate enough for my purposes. As I am very very new, I am not sure to what extent they can be changed with that build. I also worry about rights eventually interfering with my research.

I am trying to lay out the project, and have been looking at the sensors available, as these seem like a key bottleneck to accurate measurements. Frankly, I am overwhelmed by the choices available. I am also too new to know what to look for in terms of reliability and the honesty of accuracy claims.

Ideally, I would like to have the following tolerances:
Temp +- 0.25C (may be willing to settle for 0.5C)
RH +- 2.5%
CO2 +- 50ppm

Because I would like to be able to have confidence in my readings to
Temp 0.5C
RH 5%
CO2 100ppm
(feel free to correct me if my understanding of tolerances is off, still a student)

My general reactions to the sensors I have seen are a happy disbelief at the low prices coupled with a healthy dose of scepticism that they will read true.

All comments, questions, criticisms and advice appreciated.

2

Welcome to the forums EasyBee,
There are a number of threads on this forum that discuss and detail various aspects of plant propagation and hydroponics. Spend some time and I am sure you will come across these. There are a number of projects that I have helped with over the years, using these type of sensors, mostly privately away from the forum. There will be many people here that do have a lot of practical experience with these sensors and systems that may offer assistance.

Take a look at the Atlas range of sensors to see if they will suit your needs for EC and pH.

For CO2 I use, for completely other systems, a nice little unit from Dynament, Dynament CO2. I am currently working on a project using these coupled with Arduino, and I am impressed with the results.

For temperature and RH, then you may have already looked at the specifications of sensors such as the DHT22, and it other derivatives, some of which offer slightly better specifications. The DHT22 type sensor has very similar specifications to another sensor I use in my large scale industrial projects using a sensor from Vaisala, such as the Vaisala HMT-331. They also have a series CO2 sensors.

I design and commission the control and automation of large large scale industrial product storage, where these sort of sensors are required to monitor and maintain environmental conditions, so having sensors with good long term accuracy and reliability are important. These installations are not based on Arduino, but standard industrial equipment, though the same concepts apply to both forms of technology. Actually, it is quite feasible to use Arduino level technology as the basis of these larger systems. It comes down to knowing precisely the requirements.

What you can do is select reasonable quality sensors and then calibrate them against a calibrated instrument.

Another point to bear in mind is the overall 'trueness' of sensor readings based on where do you place the actual sensors in the environment, as this could be quite critical.

I'll be keen to monitor the thread, and hope you have good luck with the project.

Paul - VK7KPA

3

Thank you rockwallaby, for the advice. The Atlas sensors look great, so do the Dynament and Vaisala. The DHT22 was actually one of the ones that gave me pause, because the specs looked decent, but the price was so low I figured there was some "undocumented feature" that would render it unreliable. Is it really as good as it says?

If I am understanding your calibration comment correctly, I can trust reasonable quality sensors to be consistent, and once calibrated, can be interpreted as true going forward.

  • How frequently should I recalibrate? Or is this even an issue?

  • How should I assess if something is of reasonable quality prior to purchase?

Great point on the location of the sensor! I had been thinking about this, but really have no expertise to guide me, other than to test atmospheric variables at the growing point of the plant. In my case, I will have plant growing in an open tray, with air flowing horizontally across in one direction, and some air bubbling up through a nutrient reservoir beneath the plants. It is important to me to achieve uniform growth patterns, thus the CO2 is being added through the air bubbled through the tank, distributed relatively evenly across the bottom.

Initially, I was thinking I would place the sensors at the air intake, post plant. This would provide CO2 and humidity levels altered by the plant. I am inclined to place it there, unless there is a better place. The drawback of this is the shadow it creates on the plants, affecting growth.

Thanks for your interest in my project!