5 replies to this topic

[jlprest]

I am a noob I and I am working on my first idea using the Netduino and would appreciate any advice, suggestions, etc. I have a 3 level (vertically) plant stand used to grow herbs that has 3 separate plants on each level roughly in a triangular configuration - weight is about 10# and it is about 18" tall (picture attached). The problem is that one side faces away from the window so it has to be roatated every few days to ensure light to all the plants. My plan is to install a rotating base which I can turn with a motor. I have purchased a cheap timer motor (2/3 RPM, 120V) as a first try and will need to devise a method to connect to the base (friction, gears, belt, etc). I would like to have a light sensor that will control when the motor turns on and off to prevent rotating when dark, etc. I also thought I might install a motion sensor as a safety feature to shut off the motor should something interfere with the rotation (e.g. cat). Eventually I might be able to include moisture sensors and/or water dispensers as I know someone has already created something similar. I assume I will need a relay to control the motor and the light and motion sensors. Is this too ambitious for a beginning project? Any ideas regarding configuration, equipment, programming would be appreciated. Jackie

Attached Files

•  plant stand.jpg   21.08KB   18 downloads

[gbreder]

Hi Jackie,
the answer is: it depends...
My first real project after blinking a LED was to automate my floor heating system. It took about one year (and several hundreds of dollars) and now I have a wireless heating controlled by Netduino. Feels fantastic!

Your project seems makable, but you should define you project requirements first.

• On what is the rotation speed depending?
• How does your plant stand detect obstacles (like cats)?
• ...

If all your requirements are defined you can divide your project into smaller projects (one for each requirement) and combine them later to the end result. If this project is a good start depends on your patience and if you can handle several small projects.

Regards
Guido

[ErikN]

Hi Jackie,

I think this sounds like an amazing first project!

Guido's advice about breaking down the tasks is essential. Without it you might start to feel overwhelmed at what you've taken on and I really don't want that to happen - I'd love to see what you design!

First I'd identify what are the 'must have' features. These will be your baseline for accomplishment. The remaining features can then become additional upgrades you can make or can be brought in if you're already nearing completion of your core features before you've reached any limit on time, budget, patience, etc.

In your case, you already have a motor but that would have been one of the first things I would have suggested researching. Since others might follow your journey, the way I'd approach it is to:
1. Identify the power capability of the environment. How will this be powered?
> In your case, you've chosen a 120V motor so I'll assume from here on out you're using standard US AC power socket. Have you checked to ensure the motor is AC rather than DC input?

2. Identify the components that are critical to the function of your project. What do you need to make it work?
> In your case, you've identified (and sourced) a motor and presumably the plant stand. In your 'nice to have' features, you will need a moisture sensor and possibly a water drip source.

3. Refine your static design. Where do all the parts go?
> In your case, the only thing that stands out is that the plant appears to have multiple independent soil containers. Soil not exposed to direct light will not dry as fast - and won't need as much water - as soil that is in the light. If you water all the plants equally when a single sensor detects dryness, you could be over or under watering plants. Is this a problem? Should you use multiple sensors?

4. Identify your intended location of the project. Where will it go?
> In your case, will the plant have enough room to rotate? Will the plant brush against any objects that could tip over? What other environmental factors (curious cat) could impact the rotation?

5. Figure out which environmental factors pose a risk you need to account for in your design.
> Your project could involve electricity and unattended watering. What safety mechanism(s) do you have to prevent water from overflowing the stand?

If you take a look at these questions, your initial scope (create a rotating stand) is much less difficult than when you factor in your additional features of being able to detect soil dryness and water the plants. This is why it's good to separate out the features that aren't critical to completing your project and getting satisfaction! Besides, you can always go back and tinker, tweak, add, edit, etc.

So, now that we know at a minimum we will be working with:

Powered Motor capable of spinning a 10lbs load.
Linkage to a platform which will hold the stand that needs to be rotated.
Light sensor to power down in the evening to save power; not rotate unnecessarily.
Microcontroller with the logic.
Interfacing components.

The interfacing components get a bit tricky. While there are parts to do pretty much anything you want available quite readily, it can be a bit daunting trying to find the /right/ components for what you're trying to accomplish. 'Should I use relays? What about a transistor? MOSFET? Why is isolation of power so important?'

For powering the motor from AC (if this was a correct assumption) you have a couple options and your initial thought of a relay is what I would have thought as well. There are a couple options in this arena though. You could use some screw terminals and strip an extension cable to plug into the outlet and run through the relay. I tend not to like this approach only because I'm skittish around mains. I've been looking at the Powerswitch Tail (Adafruit) which is the same thing as using a relay but it's already wired and sealed. As long as your motor doesn't require more than 15A (also check the 'stall current' spec on the motor in case it gets stuck. This is the amount of current the motor will draw while attempting to force the rotation). The only exposure points are where you'd plug in the signal leads. Bonus: the work to prevent feedback on the signal lines when he coil is discharging is done for you and you don't need to worry about this at all! Since the board does act as an LED, I believe you should put a small resistor on the line. Some of the other people here who are stronger in hardware could weigh in on this. Using the Powerswitch Tail contrasts what you'd do if you use your own relay as you'll have put in your own protection to prevent your microcontroller from getting fried. This has been discussed previously and isn't anything new that you'd have to solve on your own should you chose to go that route. In fact there are probably more than a few compatible Relay Shield modules which have done this work for you as well.

How will you power the microcontroller? Since the board can be fully powered from either USB or a barrel jack and you'll already need access to your mains for the motor, you can easily use either a plug-in USB charger or a wall wart adapter providing the right power. These requirements are documented.


So now what do we have?

Given my preferences for non-cut/spliced mains wires, the design looks like it'd be a Powerswitch Tail and a DC adapter plugged into a power strip. The microcontroller would send signal to the tail and run the timer to kick off spinning based on light readings and possibly other factors. Since the cord and micro will be connected and both are powered from the wall mains, the micro can't live on the spinning platter in order to keep the wires from tangling unless you used a servo or stepper that could be reversed. That's no problem here, unless you want the moisture sensor... (See what I mean about pre-defining your musts?)

What's left?

Figuring out how to pair the motor to a drive on the plant stand, timing the rotation to ensure you get the right turn radius per cycle and figuring out how to keep the lighting evenly distributed in order to keep soil moisture levels fairly consistent. Figure out how to connect a moisture sensor without tangling wires!

Alternatives?

While looking up the spec for the Powerswitch Tail, I noticed a new product - the Smart Cord! It's essentially the same as the Powerswitch Tail but with a 10A capacity and Bluetooth control! This means no wires from the micro to the power tail (but requires the addition of a Bluetooth module, figuring out the protocol, etc.) Using this, your micro is no longer required to be directly connected to control the motor. Now the micro /could/ be mounted on the spinning tray and be attached to moisture sensor(s) and whatever else you would like to have. But - there's a new issue of how to power the board since plugging into the power strip would bring us back to twisted cords. If you're okay with changing a battery from time to time, there are some very powerful LiPoly rechargeable batteries available from various sources which I think would fit the bill though you might need a boost circuit to get the required voltage for the board. Going down this road, you can have an LED on your micro light up to warn you when the moisture level is too low. Getting two batteries would be great as you can have one charging (there are simple USB chargers for them) while the other is running the micro. If you wanted to get super fancy, you could probably find a battery gauge circuit for your batteries which would trigger when the power was getting too low and then light up another LED.


I'd say, just to get started and get that rush of accomplishment, look at just doing the motor and interfacing to spin the plant. Get it revolving and marvel in figuring it out and getting the linkage of the motor to the platter working correctly. Then think about possible ways of adding sensors for feedback. I don't think I'd go down the road of automatic watering of a spinning plant source - it just seems to easy to make a mistake and end up with flowing water and dirt out of the stand. I reserve water pumping for static projects that don't move and have sufficient drainage (usually a closed-loop using the same water unless the drainage is, say, a whole yard which won't flood from an outside water source getting stuck open).


I know this is a super long post but that's because I think this is an awesome project and I'd love to see how it comes out! Just know the community will be here to help you along the way if you need it.


Happy making!

[Paul Newton]

Hi Jackie, I just had a crazy idea. You could put a strain gauge under one corner of the base so that as it rotates you can tell which side of the plant is heaviest. As long as the plant was balanced on the base to start with, you could use the heaviest side to tell which side is growing the most and give that side less time in the sun. Well I did say it was a crazy idea... Paul

[ErikN]

You could put a strain gauge under one corner of the base so that as it rotates you can tell which side of the plant is heaviest. As long as the plant was balanced on the base to start with, you could use the heaviest side to tell which side is growing the most and give that side less time in the sun.

That IS a crazy (awesome) idea! Do the herbs grow at different rates/weight though? Maybe the same idea would work as a relative wetness indicator though. You could spin the platter and store the readings as a benchmark (with all the soils appropriately watered first) and then as it rotates, it could indicate when one side has less water than the rest. Maybe spend less time with this section in the direct light until it gets watered again. It's less accurate than a (set of) moisture sensors but it could help keep a neglected plant from becoming too dry (an unexpectedly warm/bright day in the window while away). Otherwise it will spin on schedule. It could also warn if the weight seems to be getting too light in general (all the soils are getting too dry).

[jlprest]

Amazing replies - thanks so much. After reviewing the suggestions I think I should set stages as suggested. Once each level accomplished move to next. Knowing myself I can have problems continuing if the project seems overwhelming. I suppose there is some hazard in not incorporating all features in a project from the beginning in case a previous step is incompatible requiring major reworking but you gotta start somewhere. 1. Develop method and control of rotation, I know I may have jumped the gun in getting the motor and may have to consider changing if it doesn't work. My considerations were that most motors have much higher rpm than I need which would bring reducing gears, etc into the equation which could be more than I want to tackle in a first project. I am fairly adept at woodworking so I am thinking about a wooden base with a lazy Susan bracket that the plant stand could sit on and the netduino, etc would be within the base. Initially I probably will wire the relay or smart cord to the netduino but bluetooth sure sounds awesome and might be included in step 2. I had thought of pointing a motion sensor at the plant stand and stopping the motor if it did not move . After the comments I may need a separate method to stop it if someone or something comes within a certain distance. Potential issues I could see are ? interference from multiple sensors of same type, sorting out appropriate actions from each one. 2. Develop method to monitor moisture and possible temp and alert if beyond certain value. Possibly alert would be a simple buzzer initially but could develop into a text, email? If multiple mosture sensors perhaps LED to indicate triggering sensor?(For when I am smarter) 3. Develop watering system which interfaces with above. Incorporate strain gauge to alert if system becoming unbalanced. I can see several issues as mentioned. Each level is actually a common space (dirt wise) but the plants (3 per level) do utilize moisture at a different rate so requirements vary. Ideally there should be a sensor for each plant but then I need a method to water each plant separately also. I can envision a 3 level water supply, stopping the rotation so a particular plant is under the spigot and then dispensing water based on the moisture level of the soil. This one may have to wait until I finally reach genius (currently planned for 2018)(joke) I appreciate the help and will post updates as things progress. Jackie