Grow room automation...Whatcha think???

Hello all,

Just wanted to kick this around.

I am a semi- retired engineer and though it has been a while since I was involved in data acquisition, I have a knack for closed loop control systems and growrrom automation is not at all as complex a system as others have made it sound.

Anyway, I grow in soil and have no real wish to change that, I would like to try and apply what I know and see it anyone else would like to add to what I have started.

I have done some research on getting one of the three primary componants:

1. A DAQ unit either a USB or preferably a PCI card.

2. signal conditioning/routing board

3. computer w/ XP

Below is a system overview. I feel all the outputs can be digitally controlled since most are either on/off anyway.

It would be nice to have a variable fan speed but thats another issue.

The software is already written for DAQ's and is simple to operate.

Next is process software to do something withthe data acquired. This of course involves writting process algorythms.... piece of cake.


Mosst of the sensors are off the shelf items so don't be fooled with supposed high tech like some automated system manufacturers (name withheld but it had something to do with growing and tronixs I beleive) want you to believe. And as far as the company that I have not mentioned goes, they are simply modifying, interfacing and repackaging existing subsystems...its good for them and free enterprize but the reputation they have here blows. Most of the on/off stuff is avail able through home automation suppliers.

I admit that the dosing machines are a bit trickier...but not some much...really we are talking about pumps here for pH up or down, adding nutes. So a method of metering the dosage is required. Unfortuantely, paristalic pumps are not cheap. but 12v car window washing fluid pumps are cheap and are simple to get- they are not submerged (least the one in my car is not) so a draw tube is installed in the fluid container and drawn up and through the pump's output side...metering becomes an issue though. Any thoughts on that.


Anyway, whats your thoughts?

minds_I




This is a simple pH dosing algorithm:



EDIT:

Forgot to mention the pH, TDS/EC probes are expensive for the process control sensors. Most if not all the other sensors are available online cheap all plug and play.

I would still use a perastolic(sp) pump as they isolate the nutes from the pump by keeping them in the tube the whole time.... they are 60-100$ new

Also there are a few DiY PH/EC/TDS probe projects on diff diy electronics websites

minds_I said:

I admit that the dosing machines are a bit trickier...but not some much...really we are talking about pumps here for pH up or down, adding nutes. So a method of metering the dosage is required. Unfortuantely, paristalic pumps are not cheap. but 12v car window washing fluid pumps are cheap and are simple to get- they are not submerged (least the one in my car is not) so a draw tube is installed in the fluid container and drawn up and through the pump's output side...metering becomes an issue though. Any thoughts on that.

Click to expand...

This is actually very easy and inexpensive IMO. Here's a thread on this forum that covers that subject thoroughly. The thread is about pH but it also applies to Hanna's EC minicontrollers as well. Don't know about hooking them up to a pc though. The high end pH minicontroller does have data output capabilities so that may be enough. You'd have to check out the EC units to see if that option is available.

As the above have said, dosing pumps are cheap. You wouldn't need too much flow from them and the smaller ones can be got for around $60. Probably cheaper if you look hard enough or buy enough of them lol. IMO paristalic pumps are a must for moving acid etc.
I see you're growing in soil so PH and EC automation probably isn't as big a deal as it would be in hydro. Enviroment and climate control (Temp, humidity, co2 etc) should be your consideration and i will be intrested in following this thread as it progresses.
Personally i'd rather have 2 seperate units. One for everything hydro and the other for the enviroment/climate just incase you decide to ditch hydro or to counter the inital start up/equipment cost. Just some food for thought.

Hello all,

Ask and you shall recieve...

These might suit a small scale (which I have been assuming) grow environment.

http://cgi.ebay.com/Ismatec-Perista...103300214QQcategoryZ78220QQrdZ1QQcmdZViewItem

Last bud was 19 bux plus 10.53 shipping.

Has several it says

Note this unit is an on or off situation of a 120VAC. Easily intergrated into the system. The down side is that in order to get a accurate dose would require dialing the pump in for volume/min rather then volume/rev. Still a nice find. May get a better deal on say 5 units (pH up/down. N solution, P solution, K solution).

minds_I

that pump should do the trick... I am confused by the large dosing range it said it does .05 to 5.5 ml per min iirc .. by adjusting the tubing... I am sure ti will make more sense after I read the manufacturers page instead of the ebay sellers write up... one interesting thing I read was it can pump two hoses at once.... guess you would need a on the lines if you wanted to use it to dose more than one type of liquid

minds_I said:

Note this unit is an on or off situation of a 120VAC. Easily intergrated into the system. The down side is that in order to get a accurate dose would require dialing the pump in for volume/min rather then volume/rev. Still a nice find. May get a better deal on say 5 units (pH up/down. N solution, P solution, K solution).

Click to expand...

If you can get that thing for cheap get it...looks awesome. Now that you've posted it here don't be surprised if others start bidding...lol. There is no down side. To get accurate dosing just drive it with a $125 Hanna pH minicontroller. The minicontroller does it all for you.

Ona Nadagin - The APT pumps are the same way regarding changing the flow rate. They have charts for three different tubing sizes and flow rates on their website.

Adversereaction - That link through Hanna USA takes you to a store asking $241 for the pH mini controller!! That's abit steep, unless it comes with the probe which i don't think it does. Automated Aquariums has the same unit for $120. I've been looking round and have found it for $118 at a few places but if i was going to buy it i would pay the extra $2 through Automated Aquariums for the better customer support as they actually use these and build them.

Minds_I - Have you got some links concerning the enviroment controls/sensors you are planning on using. I have no idea what a algorythms is or what the DAQ unit is??? You're speaking a foreign language! lol

try aquitronica it can do all that plus another room or 10 or 100 rooms it can do nutes online aswell =) no pc requierd

http://www.aquariumobsessed.com/

Use your windshield washer fluid pump. Have your pump control program set up to check EC every 15min. Use an IF THEN ELSE statement to cycle the pump for a very short period if the EC measured is less than the EC setting. This way the unit will change the EC by small amounts over a longer time and will have less overshoot.
Good luck! I'll be tuned in to see it working.

Hello all,

Grownz, as to the environmental sensors...temperature, humidity and pH/TDS/EC sensors are available on the net. As to which ones to use...well, still looking into that.

Pretty much the plan is to use off the shelf parts... I am game to help design a virtual system that covers our needs.

As far as the DAU- data acquisition unit....its just analog/digital I/O device that allows the use of a stand alone computer to sample the sensors, make decisions based on the data obtained and make process control commands and impliment them.

An algorythm is just computer code that preforms a specific set of instructions.

The whole point of the thread was to have a central stand alone computer system to monitor, control and command a grow unattended. The other point was to maybe cobble a system together virtually.

Simply buying seperate pieces of equipment is fine and all, but you have no central control.

Could be fun.

I have found this for the input side of the system.

http://www.cyberresearch.com/store/...ta-acquisition-modules/UMDAS_0802L_5397.2.htm


This is just but one option... there maybe better cards for less (older slower cards...how fast do have to sample..8kS/s is screaming fast considering the project at hand).

Still have yet to find a decent price on pH/TDS/EC...fact is, I have not been able to find a price as these things are for industrial application and as such there prices are not posted with the literature.

http://glo.mt.com/mt/filters/produc...l_4_pH_polymer_0x000249740002871b0005241d.jsp


DIY probes??? as with Hanna and similar meters I question the durability durability of the probe.

http://www.octiva.net/projects/ppm/

For a automated system, how long would you wish thhe system to operate unattended..1 week, 2 weeks a month?

minds_I

minds_I

I wouldn't wanna leave a grow unattended for more than 3-4 days. What about raising the light? If you can find away round this then maybe longer. But then you still have to worry about pests. Doubt there's any affortable automated solution for pests. Then there's electrical faults etc. Even if i dropped 10K on automated room i'd still check in at least weekly... Id say if you want automation with as little work as possible a hydroponic sog would be the best route. Just my 2 cents.

Grownz said:

Adversereaction - That link through Hanna USA takes you to a store asking $241 for the pH mini controller!! That's abit steep, unless it comes with the probe which i don't think it does.

Click to expand...

Wow...WTF??? I didn't visit any other site today besides Hanna and I just copied the URL and pasted it here. It's fixed.

Hello all,

Grownz, what first comes to mind is camerato visually confirm the need and if so, send the appropraite command from a remote location or on sight at the computer to raise the lights.

This assumes you have the appropriate equipment to raise the light. I would use a shielded hood for extra protection.

A miniture wench of some sort utilizing gearbox motors all available fairly cheap.

10k on a simple system is way overkill. As a project, I would limit costs to say 1500 bux. But thats just me.

So, as to pests... thats another issue as you mention... but a weekly visit is always a good idea.

Electrical faults...there is always a danger of that...AFCI circuit is what I would recommend.

minds_I

PS: Did I mention some DIY is involved and some electrical/mechanical abilitites are needed.

heya Minds I I like that like to the DIY EC meter... you can trust the DiY ones they are just a modified voltmeter

Good idea with the camera. That could maybe cover the pest problem too cos you would see the problem and know you had to visit your room. Having a camera has quite a few benifits.
This is all very intresting. This all requires alot more work and knowledge than the Hanna controllers obviously but would probably be cheaper and offer more central control as you said. The more i read the more i like it. I can see how it all goes together now. I will be watching and learning.

You can often find some decent deals on peristaltic pump heads on Ebay. I picked up 8 w/ steppers, no drive electronics, for under $50. I'm using a 1-wire moisture meter from hobby-boards.com for EC/TDS. $27.50 but you have to add your own probe. There's so much free software available now, that one of the hardest problems is deciding what to use. I don't control PH, as the PH is normally pretty predictable. Plus the probe will probably have to be replaced every 6 months to a year. Especially under continuous usage.

If you want to be a serious grower, please forget some DIY probes. These sensors' manufacturing are very complex and lots special know-how's and equipments need. Hundreds Scientists and Engineers digging them out.
Not easy to do. Trust me, do n't waste time and money to do it.
E.G., PH probe need special glass that you can't purchase from any shop and store. If as a commercial action, you should need order from a Japanese or Taiwan special glass companies to get good and cheap glass. However, they will ask how many tons you will order? And search them also take you lot time. Google is not good site to find them out, Too special! Big USA/Canada companies actually purchase Taiwan's products and called this: OEM (ordered Manufacture). But they will tell you these are made in USA. 90% is liars.
If you really want to save money, you should do like these big guys, to find the final manufactories, but it will take you so much money. They(Big companies) sent millions sales and buyers in Asia whole years and hiring 10 millions local people to cut the cost off. As us, normal persons, we should produce more weeds to save money.
For TDS probe, some high value metal (Pu) film you need as basic material. Very expensive. DIYs can only support us 1~2 days (maybe weeks if you lucky enough) working time, and unstable too. Compared with these, I would prefer to using a commercial probe. It will save my bucket eventually.
If you want to become an Ele Eng student, you can try DIY.

So you have made and used one and speak fom personal experience eh?..


I agree that buying a store bought probe is a good idea they only cost $25-50 , but as an EE I can assure you it is not as difficult as you make it out to be, the probes are just galvnic cells and do not have to be made of glass but it is one of the best options.

EC probes and the circuit to drive them are easy. Works fine, lasts a long time. PH probes are a different matter, but the amplifiers are easy and probe parameters are well published. Amplifier schematics are published and parts costs are minimal. An advantage to rolling your own is that if it breaks it can be easily fixed. I've built EC circuits and PH amplifiers. No big deal.

For automation software, check out Mr. House. Free open source. The downside is you have to learn Perl. There are quite a few inexpensive microcontrollers available to implement stand alone controls. The picaxe series are easy to use and cheap. Relatively slow, but the smallest unit retails for under $5 US. The next unit up is about $10.

Excitation for a EC meter NEEDS to be AC. Theres a schematic on the web of a CMOS 555 based EC meter. A couple of useful links.

http://www.emesystems.com/OL2ph.htm
http://www.emesystems.com/OL2mhos.htm

I used the conductivity circuit and it works fine. One caveat. The output is nonlinear because of the charge/discharge characteristics of a capacitor. I used Excel to do a polynomial curve fit and implemented the correction in VBscript. Works great. The hobby-boards moisture sensor uses the same excitation circuit w/ the addition of a temperature sensor and the 1-wire interface.