greenhouse Greenhouse: lighting on Overcast days

TylerInNiagara · Nov 12, 2013

Ok so I am using lights to keep my Greenhouse warm it’s easy to unplug on warm days or set a timer for supplemental lighting (adding hours of light) but I would like to have something that consumes very little power and can either thermostatically turn things on or based on how overcast it is at the time.

Any Ideas? Thanks Tyler

queequeg152 · Nov 12, 2013

big boy greenhouses do this with fancy pyranometers, idk there might be a cheap photocell version of what you are looking for.

as far as a thermostat... if you want 0 power consumption the asnwer is easy... get a cheap mechanical thermostat.

you could use one of these things. the thing attached to this fan.
http://www.ebay.com/itm/AIR-VENT-53320-ATTIC-AIRE-GABLE-MOUNT-POWER-FAN-1620CFM-THERMOSTAT-HUMIDISTAT-/300945112625

i forget what they are called but they control humidity and temp... mechanically.

there is also really nice thermostats that use coils of copper and a capillary tube, those are somewhat more expensive, but much more rugged.

theghostpepperstore · Nov 13, 2013

If you really want to get fancy, look into Arduino or Raspberry Pi as a controller. I was actually just looking into a Pi kit that does Ph and nutrient control for hydroponics along with acting as a timer for ebb and flow tables. There are all kinds of photocells that are available for it, and plenty of code repositories. You would just need to set it to turn the lights on below a certain voltage level.... but then you run into another problem. The lights will raise the voltage of the photocell and you will end up in a on/off loop. You would have to mount the photocell to the outside of the greenhouse, I guess.

Personally, I think a thermostat would be the way to go. I bought a cheap digital thermostat off Amazon for about $30 that just plugs into the outlet and has an outlet for any 110V 15Amp plug. It also had a battery backup that keeps its settings in case of a power outage.

TylerInNiagara · Nov 13, 2013

http://www.youtube.com/watch?v=Dp7U-tKFKwQ

Greenhouses of the future lit with LEDs, Aalto University Research

Inda-Gro Master Sensor and ADR Controller Presentation

theghostpepperstore · Nov 13, 2013

The controller in that second video is exactly what I was describing with the Raspberry Pi, except it is a DIY solution that shouldn't cost more than $100 and some free time. I hate to even think what something like this costs

I liked the fact that he pointed out that HID lighting isn't going to work very well in a system like this, it just takes way too long to fire. The sun might come back out before it is fully fired and that is going to stress your ballast. LEDs or fluoros would be the way to go.

IMO, LEDs have a long way to go before becoming an industry standard in professional greenhouses. I just purchased new lighting for my greenhouse and had I gone the LED route, it would have been 5X more expensive and the reduction in electricity use would never make up for the capital expense over the course of the lifetime of the lights.

queequeg152 · Nov 13, 2013

http://www.amazon.com/Ventamatic-XXDUOSTAT-Adjustable-Thermostat-Ventilators/dp/B00336G1ME/ref=sr_1_1?ie=UTF8&qid=1384376190&sr=8-1&keywords=duo+stat

this is the thermostat/humidistat i was talking about above... srry i thought this thing was alot cheaper than 30 bucks.

this is an example of the other type of thermostat i mentioned. again they tend to be very rugged, and will stand up to moisture and heat etc. all day. the cheap one i mentioned above, it far less... robust. it relies on a bimetal coil to sense heat, and a plastic film material to sense humidity. Ive never had one fail... but they are known to.

W/ respect to the supplementary lighting you see in professional type greenhouses. from what i understand... they are mostly used in upper lattitude greenhouses to supplement lighting during short photo periods in winter months? for heating etc, ive seen that they like to burn cheap biomass in highly efficient forced air boilers .

overcast days perhaps, could be detected with a simple net enabled device retrieving weather information from a weather station....or even perhaps pyranometer data from a local station? IDK if this is possible or not, im really not up on this sort of thing. seens more ideal than a photocell arrangement.

Plz allow me to play devils advocate w/ respect to the arduino/PI controllers...

IMO stay away from arduino/ PI systems as far as PH control goes.
PH controllers are fairly specialized pieces of equipment BUT they very common in industry... because they are so common, they are very cheap on the used market.

a good example is the GLI 53... it was VERY VERY common... there are 100's of these things on ebay every month. i have personally, about 3 ph controllers and a hand full of conductivity controllers. ive never payed more than 50 bucks for any of them.
they used to drop these things into sewerplants... paper mills... effluent monitoring stations etc. theres so many out there up for grabs.

yokogawa also makes a number of good ph analyzers, exact same form factor, and same sensors.
rosemount makes them... look for "solucomp" analyzers etc.
thermo now makes them... these are heinously expensive usually, full color screen... data logging onto an sd card etc.

a ph controller needs to be able to calibrate and generate a slope, keep this in mind when you consider an arduino set up. you will need a way to enter new calibration information, or have it do this automatically like a traditional ph analyzer unit

Also, MOST if not all of the cheap ph electrodes you get with your arduino PH stamps etc. are IMO junk. IMO you need an electrode rated for 100% duty, and "process" immersion. these sorts of electrodes are necessarily more expensive. with that said, i have not tried using a cheap gellie electrode for continuous usage. it may just work well enough... if you know what i mean? i mean if it lasts 6 months, fine. its probably worth the money, but keep in mind they are not designed for continous usage. they have fairly small reservoirs of electrolyte.

the continuous rated process type electrodes will tend to be refillable,have replaceable salt briges... some even have replaceable glass. most will have a much thicker glass that will hold up to repeated electrode cleanings, and hence can essentially be regenerated many many times, once its fouled... they will also have an integrated ATC thermocouple for very easy temp. correction.

some of these electrodes are rated for absurd operating conditions such as 300 psi and 200+ Celsius, this is of corse not necessary for hydroponics, however many of these electrodes are rated as such as a matter of fact... being that their main usage is for industrial processes that tend to be very very harsh.

brand new these electrodes are heinously expensive, however again they are very very common in industry and hence can be had for pennies on the dollar on the used market.

look here:http://www.ebay.com/sch/i.html?_trksid=p2050601.m570.l1313&_nkw=accuglass+&_sacat=0&_from=R40

rosemount makes very good sensors... look for accuglass stuff with appropriate threads, and the correct glass.
GLI also makes a very very good sensor, if you are patient you can get one for under 100 bucks.
yokogawa makes some good sensors altho dont be confused by " reference" electrodes. you do not want a reference electrode!

TylerInNiagara · Nov 13, 2013

I would like to keep this simple (low power under 160 watts) (Not internet connected) I'm not always around and if I have someone less technical watching things it would be great to know they can repair, reset or diagnose things. I may also want to setup a second or third Greenhouse...I have a power bar with timer and battery backup I guess I could plug one end into the power bar the other into a thermostat have the timer set to 7 on 7 off then maybe buy some more sun blaster lights with reflectors.
[SIZE=24pt]Farm Innovators Thermo Cube Thermostatically Controlled Outlet[/SIZE]
http://www.amazon.com/Farm-Innovators-TC-3-Thermostatically-Controlled/dp/B0006U2HD2/ref=sr_1_1?ie=UTF8&qid=1384383715&sr=8-1&keywords=outdoor+thermostat

theghostpepperstore · Nov 14, 2013

TylerInNiagara said:
I would like to keep this simple (low power under 160 watts) (Not internet connected) I'm not always around and if I have someone less technical watching things it would be great to know they can repair, reset or diagnose things. I may also want to setup a second or third Greenhouse...I have a power bar with timer and battery backup I guess I could plug one end into the power bar the other into a thermostat have the timer set to 7 on 7 off then maybe buy some more sun blaster lights with reflectors.
[SIZE=24pt]Farm Innovators Thermo Cube Thermostatically Controlled Outlet[/SIZE]
http://www.amazon.com/Farm-Innovators-TC-3-Thermostatically-Controlled/dp/B0006U2HD2/ref=sr_1_1?ie=UTF8&qid=1384383715&sr=8-1&keywords=outdoor+thermostat

That thing will definitely work for your purposes.

This is what I have. http://www.amazon.com/Lux-Heating-Cooling-Programmable-Thermostat/dp/B000E7NYY8/

I like the fact that it is programmable and that it can be used to heat or cool. I can use it for a space heater in the winter and a exhaust fan in the summer.

queequeg152 said:
W/ respect to the supplementary lighting you see in professional type greenhouses. from what i understand... they are mostly used in upper lattitude greenhouses to supplement lighting during short photo periods in winter months? for heating etc, ive seen that they like to burn cheap biomass in highly efficient forced air boilers .

Most of the professional greenhouses I have had the opportunity to tour have had supplemental lighting, and none of them were north of Pennsylvania. I think it really depends on what the greenhouse is being used to propagate. Lettuce for example, isn't really sensitive to photoperiods but any commercial greenhouse lettuce operation is going to want to have the same photoperiod year-round. Lettuce grown under 12 hours of light won't grow as quickly as lettuce grown under 16 hours of light. During commercial production, you would want to have a very predictable crop time which means you need to have the same length of "daytime".

queequeg152 said:
Plz allow me to play devils advocate w/ respect to the arduino/PI controllers...

IMO stay away from arduino/ PI systems as far as PH control goes.
PH controllers are fairly specialized pieces of equipment BUT they very common in industry... because they are so common, they are very cheap on the used market.

a good example is the GLI 53... it was VERY VERY common... there are 100's of these things on ebay every month. i have personally, about 3 ph controllers and a hand full of conductivity controllers. ive never payed more than 50 bucks for any of them.
they used to drop these things into sewerplants... paper mills... effluent monitoring stations etc. theres so many out there up for grabs.

a ph controller needs to be able to calibrate and generate a slope, keep this in mind when you consider an arduino set up. you will need a way to enter new calibration information, or have it do this automatically like a traditional ph analyzer unit

Also, MOST if not all of the cheap ph electrodes you get with your arduino PH stamps etc. are IMO junk. IMO you need an electrode rated for 100% duty, and "process" immersion. these sorts of electrodes are necessarily more expensive. with that said, i have not tried using a cheap gellie electrode for continuous usage. it may just work well enough... if you know what i mean? i mean if it lasts 6 months, fine. its probably worth the money, but keep in mind they are not designed for continous usage. they have fairly small reservoirs of electrolyte.

I agree with you here 100%, I wasn't necessarily saying it was a good idea, just saying it is possible. I was actually looking for process immersion electrodes at the time you wrote this. The beauty of the Arduino is that you could control any of these high grade sensors with it. The electrodes they sell with the "kits" are obviously crap. There are already several open source software packages available that allow you to enter custom variables to build the proper curve logically. I have no idea if any of them are any good though, I am not ready to start looking at the scripts yet.

Here is what I was looking into for a photosensitive relay to turn lights on/off. Remember, I am not pursuing this, I was doing this just as a thought exercise, so it is pretty half baked with lots of room for improvement. One of the key points to this design would to program a long "sample" time and a minimum "on cycle" time so that a passing cloud isn't going to cause your lights to come on for 30 seconds then go back off. You would want the Arduino to sample for a few minutes, produce an average then make the determination to turn on of off. Maybe create a logic loop that says "if [on] then wait until 3 contiguous sample sets of [below threshold] then [off]"

https://www.sparkfun.com/products/11021 -- Arduino Uno $30

https://www.sparkfun.com/products/8688 -- Light Sensor $8

https://www.sparkfun.com/products/11042 -- Relay $8

The knowledge that you built it yourself PRICELESS!

With that said, I feel like I unintentionally hijacked Tyler's thread. If I continue down this path I will start a seperate thread for discussion.

theghostpepperstore · Dec 2, 2013

Tyler, I decided that I wanted to start playing with Arduino and Raspberry Pi for some future projects. I went ahead and mocked up an Arduino board for your purpose. I figured I would drop this here so you can see just what can be done and how easy this would be to build yourself.

Arduino Uno R3
(cheap Chinese-made clone) - $12 shipped - eBay

Untitled by GhostPepperStore.com, on Flickr

Relay Shield for Arduino - $14.50 + shipping - DFRobot

Untitled by GhostPepperStore.com, on Flickr

Analog Ambient Light Sensor - $4.50 + shipping - DFRobot

Untitled by GhostPepperStore.com, on Flickr

Total Cost to this point is about $35

This is what it looks like all together. The relay shield plugs straight into the Uno. Also the shield very conveniently breaks out the 6 analog inputs along with a ground and 5V rail which makes the light sensor plug-n-play, almost. (I had to rearrange the pins in the female connector to make them match)

Untitled by GhostPepperStore.com, on Flickr

I don't know if you have any knowledge about computer programming, so I tried to put as many comments into the code as possible.

byte relayPin[4] = {2,7,8,10};
//D2 -> RELAY1
//D7 -> RELAY2
//D8 -> RELAY3
//D10 -> RELAY4
const int numReadings = 10;

int readings[numReadings];      // the readings from the analog input
int index = 0;                  // the index of the current reading
int total = 0;                  // the running total
int average = 0;                // the average

void setup(){
for(int i = 0; i < 3; i++) pinMode(relayPin,OUTPUT); // sets up relays 0-3
Serial.begin(9600); // initializes communications for serial monitor
    // initialize all the readings to 0:
for (int thisReading = 0; thisReading < numReadings; thisReading++)
    readings[thisReading] = 0; // arranges readings into array
}
void loop() {
// subtract the last reading:
total= total - readings[index];
// read from the sensor:
readings[index] = analogRead(A0);
// add the reading to the total:
total= total + readings[index];
// advance to the next position in the array:
index = index + 1;

// if we're at the end of the array...
if (index >= numReadings)
    // ...wrap around to the beginning:
    index = 0;

// calculate the average:
average = total / numReadings;
// send it to the computer as ASCII digits
Serial.println(average);
delay(500);        // delay in between reads for stability
if (average > 1000) // sets the threshold for turning lights on. Light sensor value increases as light decreases, so 1000 is very dark. Scale is 0-1024
{
digitalWrite(relayPin[0],HIGH); // turns relay 0 "on"
delay(900000);    // sets minimum "On" cycle to 15 minutes
}
else{
digitalWrite(relayPin[0],LOW); // if ambient light is bright enough, relay stays "off", 5 second delay then process starts again
}
delay(5000);
}

Basically what is happening here is the Arduino is measuring the ambient light about twice a second. Once it has taken 10 readings (5 seconds) it averages the number then decides whether or not it is dark enough to turn on the lights.

If it is dark enough, the light comes on for 15 minutes before taking another set of readings and determining whether to turn the light off or leave it on. If it is not dark enough, there is a 5 second delay before the next set of readings take place.

I set the threshold really high for testing purposes, but all of these variables are easily changed. You could also just as easily use a temperature sensor as the "input". You could even make the decision based on temperature AND light level.

The only thing I wished this had at the moment is a decent clock function, but I guess you could just have the power source for the Arduino connected to a timer if you had certain hours of the day that you know for sure you do not want the lights coming on.

I tested all the code this morning and it works like a charm.