Can I monititor 16 temp/ humidty sensors?

So I am very new, and am trying to determine whether Arduino is the right approach for my project. So sorry its a very basic question. I want to be able to get periodic readings from temp / humidity sensors, like the DHT 11 or DHT 22 I've seen. Have read how with Arduino I can do several things - multiple temp sensors and one temp/humidity sensor. Have read how I can get a shield to allow for more pins (inputs) I think up to 16. I'd like to have a go at it, if someone could tell me is this possible.

I am creating a micro-climate, aprx 75 cubic feet, for growing tropical plants. I would like to make a grid of sensors, and compile data at different hights (z) so that I can plot either temp or humidity for the volume - to the end that I might compare the effectiveness of design modifications for heating and ventilation. Have this great idea to mount the sensors on telescoping antennae, so I can ge data in different planes. One more great idea...

The sensors are cheap. The computer is very reasonable. The graphing software I think I can use for free. But I'm new at this, and am looking for someone to tell me whether this is feasible using the Arduino platform. I may also explore getting professonal help with this, if I am encouraged to proceed. Thanks for any help with my decision.

Yes this is perfectly possible.
You may want to look at a SD card reader to store your results on as well. Otherwise you seem to have thought if the right bits.
Good luck.

Do this anywhere near right and you will be able to add outputs to influence the greenhouse atmosphere. Relays are also cheap.

Yes, perfectly do-able. I suggest using a 74HC4067 (a 16 channel analog multiplexer/demultiplexer) to read your 16 analog sensors into just one arduino analog input. You'll need four digital pins to address the multiplexer.

Works fine for me, although I'm using it as a digital multiplexer to input 16 switches instead of 16 analog sensors.

A DHT sensor uses 1 data pin, as the arduino UNO has 20 IO pins that can be used for digital IO it would work quite well. An MEGA would have even more pins so no multiplexer is needed (keeps the code a bit simpler.

Never used a DHT together with a multiplexer, but it should work.

check my DHT lib here - -

personally I would go for the DHT22, is more precise and supports negative temperatures. For humidity these sensors are not that good. Based upon reactions on the forum lets say only between say 30-70% they have a 2-5% accuracy.

Do you really need 16 sensors? (why?) How big is the greenhouse?

That idea is completely workable except for using a DHT11. DHT11 humidty values are totally fictional. Use a DHT22.

I would start out without using the multiplexer, though if you end up needing more pins, that'd probably be a quick way to get more of them, and 16 pins just for sensors is a lot of pins. Do you really need 16 of them?
If you go for multiplexer, you would need to read how the mutliplexer/demultiplexer worked, to make sure it would work correctly, since the signal gets driven from both both ends to get the bidirectional communication.

Make sure that the conditions around the electronics and sensors remains non-condensing, or things will crap out fast.

Make sure that the conditions around the electronics and sensors remains non-condensing, or things will crap out fast.

+1 :slight_smile:

Heat shrink, silicon glue and heat-sink metal as needed?

Have this great idea to mount the sensors on telescoping antennae, so I can ge data in different planes.

Ensure that you give the sensor ample time to stabilize ... Internet rumors say the DHT22 humidity stabilizes quicker than the temperature. Data sheet will likely be truthful... Just add a fudge-factor.


Hey all,

Thanks so very much for all the help. My first problem, assessing feasibility, is solved in the affirmative. Next if possible I’ll just try to keep from making a fundamental mistake in choosing the hardware components.

The DHT22 as I understand it is a digital output sensor, so I don’t think I need an analog multiplexer.

I’ve read some about the limitations of these sensors, but for my application they are more than adequate. I am building a tool to help me design a micro climate. It would be generous to call my project a greenhouse. It is better described as an orchidarium, where a pretty bad habit growing green and white paphiopedilum plays itself out. I only have about 25 square feet of bench space, but must do a very good job of keeping it controlled. Circulation, light, humidity, temp., to say nothing of the care and feeding… As I close the envelope around the plants, I want to know how the placement of vents and how location of heat sources, etc., etc., is impacting conditions.

I aim to model with 3d graphing software the volume of my display case, so I can find warm spots or cold spots - dry spots or wet ones. I want to study how changes in the display case and its components impact the climate. I can’t just leave a circulation fan on high in there, and figure everything will equalize, because it’s crowded in there and the plants can’t tolerate that kind of wind. I’m a hobbyist. Some people do crosswords.

Back to the electronics, I’ve noticed that the Mega 2560R3 has 50-something digital input/output pins. That’d be more than enough. I have read on one of the replies how the Uno with 20 pins would be sufficient. I don’t truthfully see myself automating control of the ‘greenhouse equipment’. Maybe, but it’s hard to imagine. The only other output I can think of at present is the SD card writer. I have seen a $20 shield from Adafruit that would work with either the Uno or the Mega, I think. The question is whether the Uno has enough input/output pins for the 16 sensors and the SD card writer.
It seems a waste to use something as capable as the Mega for what I’m up to. Unless I did start remote controlling it all.

If the Uno would work with the card writer and 16 DHT 22’s, then I need to quit talking about it and start buying some parts. And again, many thanks for the experience you’ve shared in reply to my initial post. I will no doubt be back again, after doing my best to figure out the programming from the may published helps.

This is why its most encouraging to learn that I can use arduino boards for my project - because I can think about what my next problem is, search that specific problem, and read the experiences of people who have posted their projects on line.

Hi again,

was just looking at photo of an Uno R3 board. Looked like 13 or 14 digital i/o pins, and another set of 6 or 7 that were marked as analog. If my inputs are all digital, is it correct that I can use some of the pins marked analog? Would this be possible, but perhaps really increase the complexity for me when I try to get the thing working for the first time?

I'm thinking if it saves me complexity when I figure out how to program and run my application, then it'd certainly be worth the extra cost for the Mega.


Yes, analog pins can be used as digital inputs or outputs.

Yes, analog pins can be used as digital inputs or outputs.

...simply by declaring them digital in your code. No additional complication in the wiring.

I love that word "monititor" 8)

I love that word "monititor" 8)

Maybe he stutters. Don't mock the afflicted. :slight_smile: