My diy automated system is getting closer.

[chuey316]

Few more additions and tweaks and I'll be ready to start wiring up sensors/relays and testing with an empty tent before starting my next run. The goal here was to build a system that even someone who has never grown anything, and has zero cultivation knowledge could use to achieve a successful harvest with minimal input, then apply the experience from each grow to make the next better.

The system completely automates the environmental variables, including temp, RH, VPD, PPFD taking bulb hours into account etc, send alerts to PC, Android, and Alexa devices for voice alerts for routine maintenance or any issues with any part of the system, from hardware to software, or environmental targets being out of optimal ranges. Will also track grow costs in real time by referencing the amount of ingredients in the medium against the bulk cost paid, and tracking kwh per device. It has a drying mode as well to regulate the temp, rh, and vpd.

Every aspect of the system has built in fail-safes, and most have fail-safes to the fail-safes. Grossly over engineered, but it is setup to continue running all automation even in the event of a total network failure. The only thing that could stop it would be a power outage, which will be addressed in the future via a UPS.

No doubt a few changes will need to be made here and there as it is coded using what I believe to be optimal environmental targets and swings based on personal research here and elsewhere, and could potentially need refining in areas, but overall it should be fairly good to go. The only thing it won't do it prune, harvest and trim.

The next step will be to add code for curing utilizing air pumps, though I may do that on a separate system all together. Haven't decided yet.

 

[MercDod]

Few more additions and tweaks and I'll be ready to start wiring up sensors/relays and testing with an empty tent before starting my next run. The goal here was to build a system that even someone who has never grown anything, and has zero cultivation knowledge could use to achieve a successful harvest with minimal input, then apply the experience from each grow to make the next better.

The system completely automates the environmental variables, including temp, RH, VPD, PPFD taking bulb hours into account etc, send alerts to PC, Android, and Alexa devices for voice alerts for routine maintenance or any issues with any part of the system, from hardware to software, or environmental targets being out of optimal ranges. Will also track grow costs in real time by referencing the amount of ingredients in the medium against the bulk cost paid, and tracking kwh per device. It has a drying mode as well to regulate the temp, rh, and vpd.

Every aspect of the system has built in fail-safes, and most have fail-safes to the fail-safes. Grossly over engineered, but it is setup to continue running all automation even in the event of a total network failure. The only thing that could stop it would be a power outage, which will be addressed in the future via a UPS.

No doubt a few changes will need to be made here and there as it is coded using what I believe to be optimal environmental targets and swings based on personal research here and elsewhere, and could potentially need refining in areas, but overall it should be fairly good to go. The only thing it won't do it prune, harvest and trim.

The next step will be to add code for curing utilizing air pumps, though I may do that on a separate system all together. Haven't decided yet.

If your making this for others to use can you make it so the user can input their own optimal ranges?

I also noticed you comment on bulb hours (MH and HPS). I think most people are using LED now, would that need any tweaking for those peeps?

Is it possible I could custom input the wattage, distance, PPFD, DLI for my particular light so the program could advise me on the light settings if I need a particular PPFD or DLI?

Could DLI be calculated for custom times ( 18/6, 19/5, 12/12, etc...) and ramp up or down as customed programed by user.

Would I be able to control the amount of veg time I wanted as well as Flower time?

Would it work with both photoperiod and Autoflowers?

If I never use a heater, can I turn off that function?

Not being a jerk here, it just seems you have your sh%t together and im interested. Maybe you might give me some pointers on making a system that is more taylored to my style.

Nice job!

 

[PlumberSoCal2]

Few more additions and tweaks and I'll be ready to start wiring up sensors/relays and testing with an empty tent before starting my next run. The goal here was to build a system that even someone who has never grown anything, and has zero cultivation knowledge could use to achieve a successful harvest with minimal input, then apply the experience from each grow to make the next better.

The system completely automates the environmental variables, including temp, RH, VPD, PPFD taking bulb hours into account etc, send alerts to PC, Android, and Alexa devices for voice alerts for routine maintenance or any issues with any part of the system, from hardware to software, or environmental targets being out of optimal ranges. Will also track grow costs in real time by referencing the amount of ingredients in the medium against the bulk cost paid, and tracking kwh per device. It has a drying mode as well to regulate the temp, rh, and vpd.

Every aspect of the system has built in fail-safes, and most have fail-safes to the fail-safes. Grossly over engineered, but it is setup to continue running all automation even in the event of a total network failure. The only thing that could stop it would be a power outage, which will be addressed in the future via a UPS.

No doubt a few changes will need to be made here and there as it is coded using what I believe to be optimal environmental targets and swings based on personal research here and elsewhere, and could potentially need refining in areas, but overall it should be fairly good to go. The only thing it won't do it prune, harvest and trim.

The next step will be to add code for curing utilizing air pumps, though I may do that on a separate system all together. Haven't decided yet.

That's cheating

 

[Grownsince95]

I may sell them, or just go open source. I haven't decided yet, but I'll need a grow or two on it either way to make sure everything is good. As for the ranges, they are all adjustable from the dash.


There's an option for LED as well. By clicking the drop-downs, you can choose between veg, flower, and drying stages, as well as MH, HPS, and LED lighting source.


Yes. The wattage, distance, PPFD, and DLI are all user defined and editable through the dash. Right now the system uses the ppfd range set by the user, and uses the light type, wattage, bulb hours, and light distance to calculate PPFD/DLI, and show the calculation on the dash. I'm currently coding in the advice to raise or lower lights based on the PPFD/DLI, as well as a switch to toggle between a calculation, and an actual PAR sensor for the most accurate data.


Yes. This is already coded in.


Yes. There is a selection for number of veg, flower, and drying weeks in the dash. I am currently coding the addition of a manual advancement or regression of the current week as well, so if plants are growing faster or slower than expected, or recovering from training, the week can be changed to give more or less time.


Yes, it will.


Yes, all devices have a on/off toggle to include or exclude them from automation. If I decide to sell or distribute the code I will likely add an AC support, but I don't currently have the use for it personally, so it is missing at the moment.


Not taken that way at all. In fact I love these types of questions because it makes me go back and check the code for any good additions if needed.

So it's operating fans, heaters, pumps etc? How is it "talking" to all the equipment? Are there smart plugs? What is it going to be compatible with?

 

[chuey316]

If your making this for others to use can you make it so the user can input their own optimal ranges?

I may sell them, or just go open source. I haven't decided yet, but I'll need a grow or two on it either way to make sure everything is good. As for the ranges, they are all adjustable from the dash

I also noticed you comment on bulb hours (MH and HPS). I think most people are using LED now, would that need any tweaking for those peeps?

There's an option for LED as well. By clicking the drop-downs, you can choose between veg, flower, and drying stages, as well as MH, HPS, and LED lighting source.

Is it possible I could custom input the wattage, distance, PPFD, DLI for my particular light so the program could advise me on the light settings if I need a particular PPFD or DLI?

Yes. The wattage, distance, PPFD, and DLI are all user defined and editable through the dash. Right now the system uses the ppfd range set by the user, and uses the light type, wattage, bulb hours, and light distance to calculate PPFD/DLI, and show the calculation on the dash. I'm currently coding in the advice to raise or lower lights based on the PPFD/DLI, as well as a switch to toggle between a calculation, and an actual PAR sensor for the most accurate data.

Could DLI be calculated for custom times ( 18/6, 19/5, 12/12, etc...) and ramp up or down as customed programed by user.

Yes. This is already coded in.

Would I be able to control the amount of veg time I wanted as well as Flower time?

Yes. There is a selection for number of veg, flower, and drying weeks in the dash. I am currently coding the addition of a manual advancement or regression of the current week as well, so if plants are growing faster or slower than expected, or recovering from training, the week can be changed to give more or less time.

Would it work with both photoperiod and Autoflowers?

Yes, it will.

If I never use a heater, can I turn off that function?

Yes, all devices have a on/off toggle to include or exclude them from automation. If I decide to sell or distribute the code I will likely add an AC support, but I don't currently have the use for it personally, so it is missing at the moment.

Not being a jerk here

Not taken that way at all. In fact I love these types of questions because it makes me go back and check the code for any good additions if needed.

 

[chuey316]

So it's operating fans, heaters, pumps etc? How is it "talking" to all the equipment? Are there smart plugs? What is it going to be compatible with?

I operates fans, heaters, humidifiers, dehumidifiers etc. All of the logic is stored locally on the board, but the targets can be edited at anytime from a pc or phone, and the new settings are then saved so that even in the event of no internet connection or a complete network failure, the board will continue to run the automation based on the targets set by the user.

Smart plug code is included, but they are only used to control the devices in the event that the main board is unable to for some reason, like a relay getting stuck on, or a code glitch. Basically it will run completely independent of smart plugs, but using them adds a fail-safe layer to ensure that if there is an issue then the devices will still be controlled according to the logic and the set targets.

The real-time grow cost calculator however relies on the use of smart plugs as it monitors the current draw to calculate the cost based on the kwh cost that the user defines. So without smart plugs the system will still run the same, but the cost of electricity will not be included, and the device takeover fail-safe would be gone.

 

[Grownsince95]

I operates fans, heaters, humidifiers, dehumidifiers etc. All of the logic is stored locally on the board, but the targets can be edited at anytime from a pc or phone, and the new settings are then saved so that even in the event of no internet connection or a complete network failure, the board will continue to run the automation based on the targets set by the user.

Smart plug code is included, but they are only used to control the devices in the event that the main board is unable to for some reason, like a relay getting stuck on, or a code glitch. Basically it will run completely independent of smart plugs, but using them adds a fail-safe layer to ensure that if there is an issue then the devices will still be controlled according to the logic and the set targets.

The real-time grow cost calculator however relies on the use of smart plugs as it monitors the current draw to calculate the cost based on the kwh cost that the user defines.

You didn't answer two of my questions so I'll ask one more time.

How does it communicate with my equipment and what is it compatible with?

For instance, how do I connect my spider farmer light and old AC infinity fan? Then I have a humidifier, heater and aquarium pumps that all need to be plugged in as well.

 

[chuey316]

How does it communicate with my equipment and what is it compatible with?

There is a main board that controls a relay board to operate each device, as well as the optional smart plug per device that can be used to take over individual control in the event that the main logic board cannot do it for whatever reason. The only thing that would not be covered at this point would be a failed smart plug, or a relay that is stuck open. Though an alert would be sent, and a person could just remove the bad smart plug and let the relay control the device, or if a relay is stuck open, they could bypass the relay, and use only the smart plug until a new relay is installed. Either way would allow the automation to continue.

For instance, how do I connect my spider farmer light and old AC infinity fan? Then I have a humidifier, heater and aquarium pumps that all need to be plugged in as well.

The devices are plugged into the smart plugs (optional, but fully recommended), and wired through the relays to be controlled. As of right now the devices are heater, humidifier, dehumidifier, light, and fan, but I am working on a way to add new devices from the dash, though it would require either an additional relay board or one with more relays which isn't a big deal and would work just the same.

Also right now, even though the main board runs independent of any external software, the dash and abilities to edit targets is coded to work with a pc running Linux mint 20.3, though I may make a windows or raspberry pi version as well. Linux is just what I use as it's far more stable, so that's how I made it.

 

[chuey316]

That's cheating

If there's one thing that I have learned as I get older, and acquire more obstacles from past injuries, it's work smarter not harder. Anything that can takeover some of the work, and keep me from having to go up and down the stairs to tend to things is a win.

My next step outside of this project will be to build new SIP wicking buckets to decrease the amount of watering that I need to do. I'll still need to top water for dressings and in the seedling stage of course, but filling a reservoir vs individual manual top watering throughout the grows will be a welcomed change.

Now if I could only come up with a way to get these 5gal jugs to the store to be filled, and then transferred to my basement without me having to lift them I'd be set.

 

[Grownsince95]

There is a main board that controls a relay board to operate each device, as well as the optional smart plug per device that can be used to take over individual control in the event that the main logic board cannot do it for whatever reason. The only thing that would not be covered at this point would be a failed smart plug, or a relay that is stuck open. Though an alert would be sent, and a person could just remove the bad smart plug and let the relay control the device, or if a relay is stuck open, they could bypass the relay, and use only the smart plug until a new relay is installed. Either way would allow the automation to continue.

The devices are plugged into the smart plugs (optional, but fully recommended), and wired through the relays to be controlled. As of right now the devices are heater, humidifier, dehumidifier, light, and fan, but I am working on a way to add new devices from the dash, though it would require either an additional relay board or one with more relays which isn't a big deal and would work just the same.

Also right now, even though the main board runs independent of any external software, the dash and abilities to edit targets is coded to work with a pc running Linux mint 20.3, though I may make a windows or raspberry pi version as well. Linux is just what I use as it's far more stable, so that's how I made it.

How are smart plugs "optional" if that's my only way to connect to it?

What equipment out there right now connects to your "relay board"?

What kind of connection is it? 3 pin? 5 pin? USB?

I'm asking because you do know this is on the market already right?

 

[Grownsince95]

If there's one thing that I have learned as I get older, and acquire more obstacles from past injuries, it's work smarter not harder. Anything that can takeover some of the work, and keep me from having to go up and down the stairs to tend to things is a win.

My next step outside of this project will be to build new SIP wicking buckets to decrease the amount of watering that I need to do. I'll still need to top water for dressings and in the seedling stage of course, but filling a reservoir vs individual manual top watering throughout the grows will be a welcomed change.

Now if I could only come up with a way to get these 5gal jugs to the store to be filled, and then transferred to my basement without me having to lift them I'd be set.

Btw I think this a great idea and it's always satisfying to design and execute something yourself!

 

[chuey316]

How are smart plugs "optional" if that's my only way to connect to it?

The smart plugs are not the only way to control the devices. That is handled by the logic board. The smart plugs are for power monitoring and a fail-safe to be controlled in the event that the board is unable to.

What equipment out there right now connects to your "relay board"?

If it uses 110v power it can be connected.

What kind of connection is it? 3 pin? 5 pin? USB?

The logic board is USB powered, and the relays control the hot wires from each device, and connects to the logic board via jumpers to the header.

I'm asking because you do know this is on the market already right?

Yes I know. I didn't realize that when I started, but this will do most if not all of what the others can do and then some, plus it's a fraction of the cost.

I'm at about $50 all in and could have done it for less had I been ok with waiting on shipping for different parts.

 

[chuey316]

Btw I think this a great idea and it's always satisfying to design and execute something yourself!

This was my reason besides the lower cost for continuing when I seen that there were automation systems available.

Figuring out ways to make it better than what I could buy and watching it all come together is very satisfying, but the real test will be on the first couple grows with it.

 

[shooter64738]

Im in a similar position and mostly trying to help disabled growers. Automation is the only way for me (or them). Im running behind, but hope to have a full autonomous grow running from seed to smoke starting next month. Still have some model training to finish first tho.

This looks really good! Takes quite a bit of effort to get it all together. Good job!

 

[chuey316]

Im in a similar position and mostly trying to help disabled growers. Automation is the only way for me (or them). Im running behind, but hope to have a full autonomous grow running from seed to smoke starting next month. Still have some model training to finish first tho.

This looks really good! Takes quite a bit of effort to get it all together. Good job!

Thanks. Appreciate it. That was the plan here. I was in a work related auto accident in June of 2024 and have been off work and in physical therapy for lumbar injuries since, so I wanted a way to automate everything so all I would need to do is prune, and the occasional top watering, though the code for automatic watering is there, I'm just not using it as of right now.

Are you building yours diy as well? What type of code, C++, yaml? If you ever wanna bounce ideas around don't hesitate to msg me. I may have features that you may want to add, and likewise you may have some that I could have overlooked as well.

Good luck with your build. Hope it works fantastic.

 

[cpurola]

You are, will be, an asset to the farm. Thanks for sharing. I'm totally a hands on, not disabled (altho' old) person and can see this in my future. (hopefully not for a few years)
I've been growing outdoors for 6 years but dipped my toe in indoor growing and it is very challenging.

 

[Brother_Antioch]

If you're looking for a Beta tester for, say, next fall for an indoor hydroponic setup in a 2x4 tent, I happen to know a guy...

Cheers,

BA.

 

[chuey316]

Thank you, but the true assets are people here and elsewhere that share the knowledge of the proper environmental variables for growing this wonder plant. I just built the system to use that info.

Like stated by Grownsince95, there are systems on the market to do this that may not have as many features, but are straight forward to setup.

This is just a cheaper way to automate, that has far more redundancy and fail-safes than any consumer grade option, or commercial grade for that matter, plus new features or hardware can be added at anytime.

 

[shooter64738]

Thanks. Appreciate it. That was the plan here. I was in a work related auto accident in June of 2024 and have been off work and in physical therapy for lumbar injuries since, so I wanted a way to automate everything so all I would need to do is prune, and the occasional top watering, though the code for automatic watering is there, I'm just not using it as of right now.

Are you building yours diy as well? What type of code, C++, yaml? If you ever wanna bounce ideas around don't hesitate to msg me. I may have features that you may want to add, and likewise you may have some that I could have overlooked as well.

Good luck with your build. Hope it works fantastic.

Like most things this started as one thing and over the years has exploded into something else. I have spinal cord damage at c6-c7 so everything mid chest is kinda wonky and walking is very exhausting/impossible.

Mine is not so much diy anymore, but if you're crafty and have a 3d printer I guess you could. :)

Most of the hardware control is done with C++, running (mostly now) on some flavor of stm32, all connected via can bus. The hmi/front end and ml runs on a small windows touch screen pc. I did have to do some odd stuff with a pi compute board for the cameras.

My wife helps me quite a bit and any tasks I delegated to her I found a path to automate it. Then several years later I've created a whole eco system.

 

[chuey316]

Like most things this started as one thing and over the years has exploded into something else. I have spinal cord damage at c6-c7 so everything mid chest is kinda wonky and walking is very exhausting/impossible.

Mine is not so much diy anymore, but if you're crafty and have a 3d printer I guess you could. :)

Most of the hardware control is done with C++, running (mostly now) on some flavor of stm32, all connected via can bus. The hmi/front end and ml runs on a small windows touch screen pc. I did have to do some odd stuff with a pi compute board for the cameras.

My wife helps me quite a bit and any tasks I delegated to her I found a path to automate it. Then several years later I've created a whole eco system.

Nice sounding setup. One of my next additions will be ai cameras to monitor for issues and give instructions.