Arduino Climate Control

[bobinca]

I'm working on an attic grow and it is off to a rough start. Temperatures ranged from 30's to 80's in the last month. The plants are still alive but not doing too well.

My goal is to have 2 climate controlled tents in the attic, one for flower and one for veg with each tent holding 3 plants.

Currently I have 1 tent with an arduino that logs temperature, humidity, light, soil moisture and CO2.

This is my first 24 hours of data since I brought up the arduino yesterday, I'm not sure why after the lights came back on the light sensor is only showing 4.5k lux instead of 60k. The spike at the beginning for CO2 is when I opened the tent and was breathing on the plants.

My gear so far:

Virtual Sun 36in x 20in tent
Hydro Crunch 315-Watt Ceramic Metal Halide Light

DFRobot Mega 2560 V3.0
Soil Moisture sensor
DHT11 Temperature and Humidity Sensor
VEML7700 Light Sensor
MH-Z16 NDIR CO2 Sensor
W5500 Ethernet shield
SD2405 RTC

I will switch to LED light once the weather gets a little warmer, I have a Horticulture Lighting Group HLG 100 V2 Quantum Board QB192

I plan to add a 3rd tent that will contain a heater, dehumidifier, humidifier and a/c. Outside air, veg and flower tents will all exchange air with this tent. I'm doing it this way because my space is so limited that a 3rd small tent works better than increasing the size of the other tents. 36x20 inch is really the largest tent I can fit.

The arduino will be responsible to turning on/off the outside air fan, humidifier, dehumidifier, heater and a/c. For now I'm leaving the light control with a timer. I might be too ambitious for my first attempt at the software, but I'm hoping to minimize electricity usage, so if the outside air is acceptable then it should use that. I also plan to have the arduino control watering the plants.

 

[Old Salt]

Sounds ambitious. You may be able to use some of the code I published on this site, if you plan to add an RTC.

 

[bobinca]

@Old Salt I updated my list of components, I forgot the ethernet shield and RTC.

For temperature control, I'm implementing a 24 hour clock where each hour can have its own ideal temperature set. Something like 75F when lights are on, 70F when veg lights are still on but flower lights are off and 65 when all lights are off. Right now it is just an idea, I'm not sure how things will shake out with veg and flower tents sharing the same air.

I already get the correct time with NTP, so the RTC wasn't really necessary, but I added it just in case the device restarted and internet wasn't available, I use it just as a back up clock.

I see that you wanted automatic watering so that you don't have to worry if you are gone for a couple of day. How's that working out? What is the longest you have gone? I have a 3 week trip coming up in the summer and was hoping to continue my grow during that time.

 

[Old Salt]

see that you wanted automatic watering so that you don't have to worry if you are gone for a couple of day. How's that working out? What is the longest you have gone?

I'm surprised at how well the timer is working out. I can go for as long as my reservoir will last, 4-1/2 days for a 19l / 5 US gal pail. I can probably push it up to ten days by using a larger reservoir. My limit is the ten days I can prepare the nutrient solution in advance of my departure.

 

[Latitude17]

My limit is the ten days I can prepare the nutrient solution in advance of my departure.

Is this a self-imposed limit or a limitation in the nutrient line you're using? Are you watering multiple times per day?

 

[Old Salt]

Is this a self-imposed limit or a limitation in the nutrient line you're using? Are you watering multiple times per day?

It's a nutrient limitation. I'm using General Hydroponics' Flora Trio. I water twice daily. I pump about a liter from the drain trays for every 18l of watering. This is the 5-10% run off I wanted.

 

[bobinca]

The soil moisture sensors seem to be working. You can see in the graph when I watered (the numbers went down). I find it interesting that the sensors report numbers that are consistently off from each other. My guess is that the soil compactness around the sensor could make a difference. I will have to see if a particular sensor always reports higher when I transplant them in the next week.

 

[odam2k]

What are you using for soil moisture sensors?

 

[bobinca]

What are you using for soil moisture sensors?

Capacitive sensors, Gravity: Arduino Analog Capacitive Soil Moisture Sensor- Corrosion Resistant -DFRobot

If they don't work out, I might try your stainless steal pin approach. I think that looks pretty resilient.

 

[Latitude17]

Ditto ^

 

[bobinca]

I got a case for my arduino that holds the mega plus one shield, which for me is an ethernet shield. Sensors and relays will be connected to the arduino over ethernet cables. In the picture I have 1 6ft ethernet cable done. For the 8 wires, 2 are for power, 2x are i2c and 4x are for pins, which for now I only use one digital pin for a temperature/humidity sensor. The light and CO2 sensors use i2c. I ran 1x i2c cable outside of the case as an optional display plug. I like the display for debugging purposes, but I imagine most of the time I will run the arduino without a display and will rely on it logging data to a server to monitor what it is doing.

I plan to make a couple more ethernet cables, 1 cable will handle the moisture sensors, 1 cable will handle the relays (like pumps, heater, etc) and 1 cable will be for outside temp/humid sensor. I'll be curious to see how long I can make the cable for the outside sensor and have it still work. I'm hoping for at least 15-20ft.

 

[Old Salt]

I'll be curious to see how long I can make the cable for the outside sensor and have it still work. I'm hoping for at least 15-20ft.

Why not use RF with a commercially available sensor like the ones from AcuRite? They also have complete systems with a USB interface that you can connect to your Arduino. You wouldn't have to pierce your wall, and you'd have more flexibility in placing your sensors.

 

[bobinca]

Why not use RF with a commercially available sensor like the ones from AcuRite? They also have complete systems with a USB interface that you can connect to your Arduino. You wouldn't have to pierce your wall, and you'd have more flexibility in placing your sensors.

I hadn't considered that.

Since my grow is in the attic I will just probably just place a sensor near a vent in the roof, maybe I will seal the duct to the vent and place the sensor inside the duct to help insulate it from the attic. But if I can't get a good reading because of the heat in the attic I will consider a remote sensor. Or maybe at that point I will just grab the current temperature from google

 

[bobinca]

I think I finally figure out how to mix in fresh air. Currently I have a veg tent and an empty tent set up. A fan sucks air out of the veg tent. A carbon filter is attach to the duct on the inside of the veg tent. The air is exhausted into the empty tent and there is a return duct from the empty tent back into the veg tent. This maintains a negative air pressure in the veg tent so that odor is controlled. With the veg tent at negative pressure and the empty tent at positive pressure I was able to open a 6 inch port in each tent and fresh air is pulled into to the veg tent and filtered air is blown out of the empty tent. So recirculating air and mixing in fresh air happens with just one fan and all I have to do is control opening/closing a port in each tent. Next I'll figure out how to get my arduino to control opening/closing those ports.

I don't need a/c right now, but in a couple of month I expect that to change. I've researched portable a/c's and I would really like to find one that could work for me, but so far I have had no luck. I saw some pet ones that looked good, but they ended up having bad reviews. One of the main challenges is that my entire grow is on a single 15 amp breaker. So with lights, fans and A/C, my target is to keep it all under 1000 watts. Currently I have 1 fan around 100 watts, and will use 200-300 watts for lighting. That leaves me with about 600 watts for the A/C.

So far my best bet seems to be

Frigidaire FFRA0511R1 5, 000 BTU 115V Window-Mounted Mini-Compact Air Conditioner with Mechanical Controls

It is only 400 watts and it has manual control which makes it easier to turn on/off with the arduino. Of course I will have to build my own enclosure for it.

 

[Old Salt]

Why not use the empty tent as an air mixing/conditioning space. Draw air from your veg tent, and send some to the empty tent, and vent the remainder. Use stepper motors to control the air flow for the empty tent, and vent. Open the empty tent only for intake. Connect the empty tent to the veg tent as its source of 'fresh' air. Mount the air conditioner to the side of your empty tent for temperature/humidity lowering. Add a heater and humidifier to the empty tent to raise the temperature and humidity as required.

Don't forget that room size has no real impact as far as the air conditioner goes. You'll have far more air moving through the empty tent than the room the air conditioner is rated for.

 

[bobinca]

I think I see what your a saying, instead of the intake port being in the veg tent, I put it between the fan and the empty tent, when it is open the fan will push the air out and the intake into the veg tent will pull air from the empty tent creating the negative pressure in the empty tent necessary to pull air in if I open a port in the empty tent. This would have the added benefit over my design in that if the intake is a little shocking (like too cold), it will have a chance to mix with the hotter inside air before being introduced into the veg tent. Thanks, I like it.

 

[DrewT]

You guys are on to something! I'm liking this direction. The "pre-mix" zone approach is something I've been zeroing in on for a couple months. I've been doing some rough calculations for my application and so far any conventional cooling device is either too powerful (window AC) or too weak/ineffective (swamp cooler, peltier). I'm planning to try something like Bob in terms of zones, but will use a chest freezer to do the cooling (hear me out, I know it sounds like a dead end at first).

I live in the Pacific Northwest, so summers don't get that hot and most months I can pull cool air from outside to reduce temps. During the warmer months, I only need supplemental cooling for 6-8 hours during the day and occasionally at night. I was thinking about putting a chest freezer into the pre-mix zone, with a vent fan punched through the lid. Fill the freezer with water blocks, let it ice up at night and then circulate air through it (and into the pre-mix) as needed during the day. I plan on sealing and venting my light fixtures, thus removing a majority of the heat at the source. I may be completely underestimating cooling needs for the grow area but it is pretty small (50ish cubic ft).

 

[Old Salt]

You guys are on to something! I'm liking this direction. The "pre-mix" zone approach is something I've been zeroing in on for a couple months. I've been doing some rough calculations for my application and so far any conventional cooling device is either too powerful (window AC) or too weak/ineffective (swamp cooler, peltier). I'm planning to try something like Bob in terms of zones, but will use a chest freezer to do the cooling (hear me out, I know it sounds like a dead end at first).

I live in the Pacific Northwest, so summers don't get that hot and most months I can pull cool air from outside to reduce temps. During the warmer months, I only need supplemental cooling for 6-8 hours during the day and occasionally at night. I was thinking about putting a chest freezer into the pre-mix zone, with a vent fan punched through the lid. Fill the freezer with water blocks, let it ice up at night and then circulate air through it (and into the pre-mix) as needed during the day. I plan on sealing and venting my light fixtures, thus removing a majority of the heat at the source. I may be completely underestimating cooling needs for the grow area but it is pretty small (50ish cubic ft).

Rather than blocks of ice, I recommend containers filled with Smurf Juice (windshield washer fluid). I used it in hot water bottles to ice my wife's knee following her knee replacement. Sealed containers will not let the ice melt and evaporate increasing your humidity.

 

[bobinca]

I'm finally getting ready to have my arduino take a more active role. Instead of just logging, the sensors it will control mixing fresh air, the light and a heater. Hopefully this will help stabilize the temperatures. I attached a screen shot of the graphs, temperature has swung from 63 to 90 in the tent and humidity is a disaster at 9% to 43%. I also plan to exchange my MH light with a LED. Combined with the heater, hopefully this will give me the control I need over temperature. I plan to put in an a/c in the next few weeks. It is definitely getting warmer in the attic.

 

[bobinca]

I upgraded my setup

1) I installed the dampers to control when fresh air is allowed
2) I added a heater
3) I added a dehumidifier
4) I added an Exhale CO2 bag to help since I'm trying to keep the room a bit more sealed (tents are not air tight)
5) I added an outside air temp/humid sensor
6) Lights are now controlled by the arduino instead of a timer

So far in the first hour I noticed the outdoor temperature sensor, which is near vent, but not actually outside, is influenced too much by the attic temperature. I will have to figure out another idea, maybe a remote sensor like @Old Salt suggested. Right now the outside air is reading 85, but it is actually 75. The arduino won't let in the air even though the 80 degrees inside the tent is too hot and the outside air would help.

I also wiggled the soil sensors while installing the new cabling I made. Looks like when they are loosened in the soil it affects the reading quite a bit, making it look a lot dryer. I think this makes sense and explains why some times my 3 plants can have wildly different numbers. I think the actual number isn't useful. For manual watering I look for a relatively unchanged number for 24 hours, then I know it is dry.