dstroy

Well-Known Member
I got the dimmer circuit fixed, and I'll share it with you so that anyone can build it. It was a little more involved than I thought it would be initially but all the parts are really cheap.

I added two switches and two 10kohm potentiometers to the front panel of the control box. The switches are an easy way to switch the different grow areas from veg to flower without having to upload a different program to the Arduino when it's time to flip, all they do is tell the Arduino if they are open or closed and the program switches the photoperiods to 12 or 18 hours. The 10kohm potentiometers allow me to manually dim the lights.

So instead of using transistors (PWM, pulse width modulation) I ended up using an LM324N op amp (additive voltage, 0-10v). It has 4 channels available (perfect!) so that each meanwell driver has its' own dimming circuit.

I'm not sure why the PWM dimming circuit didn't work, but probably had something to do with there only being two channels out and I have 4 drivers so I had to wire them in parallel.

But this one is working great!

Each individual circuit looks like this:
LM324N.png
and it looks like this when it's all put together
IMG_1764.JPG
Don't mind my messy protoboard, I'm not the best at point to point wiring.

I added some more stuff to my serial monitor:
IMG_1766.JPG
(all of this stuff gets datalogged, at 5 second intervals)

and the wiring of the switches/pots
IMG_1767.JPG
IMG_1765.JPG

The only other thing that I've done is to add large (10A) flyback diodes for the solenoid circuits.
 

dstroy

Well-Known Member
Hi ladies/gents,

Just wanted to let you know that I've been super busy but haven't forgotten about this, I've been working on a schematic of this whole control box thing and have changed out some small parts, (terminals, resistor ratings etc.). It's been in heavy testing with everything connected and as soon as I'm comfortable that everything is working as intended I'll release the schematic and full parts list I used.
 

smokeingdog

Well-Known Member
Making more progress, control box is all done now. I've got the dimmer circuit wired up, but something is screwy with it. It dims, but not down to 100%. With the multimeter hooked up monitoring it, the current from the driver goes from 1.4A to 1.2A over the 15 minute sunrise/sunset. I've verified that the voltage going to the led driver dimmer is going from 10v to 0v over 15 minutes like I wanted but the current is only dropping 14% so I'm going to have to figure out what the deal is. The documentation for the drivers have a graph that shows current should be going to almost 0 when 0vdc is applied to the led driver dimmer.

I've decided to use 3/8" tube for the nute line "backbone". I've got all the parts I need for that save for one epacket coming from china with adapters that I need for my solenoids.

The tube coming from the backbone, going into the bucket and supplying the 3 misters is going to be 1/4" the only other thing that I want/need are some check valves that crack open at 15ish psi so the misters don't dribble. I've got enough valves so that each bucket has its own and can be shut off and moved around without interrupting any of the other buckets.

The other thing I'm on the fence about is adding some 40mm noctua fans that I have to the control box. Sometimes the vreg on the Arduino gets kind of warm, but it is only supplying power to two relays. I'm just thinking about longevity, I should power those two relays with the 5v power supply that is installed but that terminal block is really full, not as bad as the 120v terminal block, but still really full. Each terminal pair for each block is rated for 10A but more than two connectors on each is sloppy and potentially dangerous.

The wife has sorta decided on what beans she wants: blueberry and strawberry sour diesel.

She's not sure yet though, she's been smoking her way through the menu from the dispensary finding out what works best for her. http://greenleafcare.org/

They only sell clones a couple times a year, it'd be great if the timing lines up so that we can get some from there.
All this is greek to me,no clue at all,i wouldn't know were to start
 

dstroy

Well-Known Member
I'm confident this is stable enough to use now, after much testing. I'd never really coded before this except way back in high school which is why this probably looks juvenile to some of you. Getting closer to finishing the list of components I used, and some things I would do differently. Current uptime on this code with everything hooked up and running is 7 days.
 

Attachments

dstroy

Well-Known Member
Improved stability, reduced CPU usage and decreased heap fragmentation. This was accomplished by turning each function into a state machine, on or off. If the state hasn't changed, nothing is done except for rule testing (ie if the lights are on, only test to see if they should still be on and don't use digitalWrite() unless the state has changed). This increased memory usage by 11 bytes to 922, still plenty of memory left.
 

Attachments

dstroy

Well-Known Member
ok, so since I'm not planning on paying for OR designing a custom pcb to house everything that I've put into this box I highly recommend that you get something like:

https://www.amazon.com/gp/product/B01CU3OB5Y/ref=oh_aui_detailpage_o02_s02?ie=UTF8&psc=1

or the Rpi3 equivalent if you're using that. I also have cord end terminals (bootlace terminals?) that go on all the wires that go into the screw terminals. This prevents "whiskers" aka stray strands of stranded wire from touching other terminals. The cord end terminals are common in Europe and Asia, and are making their way around the US, they're great at what they do, and you can get tools to crimp them on the cheap from amazon.

Once again, I'm going to plug the high quality potentiometer, especially if you're going to be using them to control something critical like lights. You don't want your lights to turn on and off repeatedly because your "one hung low" china/ebay special $0.10 potentiometer decides that it's done doing its job. Yeah, the good pots are more expensive ($10-20) but isn't good pot always expensive?

image1.JPG image2.JPG

the screw terminals are a lot more reliable long term than pin headers.
 

dstroy

Well-Known Member
Another lesson learned. The relays I'm using are only good for about 50,000 cycles. If your HPA solenoids are cycling every 3 minutes, that comes out to 104 days before it (the relay) reaches its' life expectancy. A much better choice is to use MOSFETS, specifically RFP30N06LE. I installed five so I have one extra just in case, it's spare channel. These should last a very long time. I've made some more progress on the schematic but it's a lot of work and I only have about an hour a day to do it.

IMG_2241.JPG

This is a good tutorial:
http://bildr.org/2012/03/rfp30n06le-arduino/

I edited the source code, made some stability improvements and changed it to work with MOSFETS
 

Attachments

dstroy

Well-Known Member
Getting some pretty strong signs from the universe here. My wife found 7 seeds in some Grape Ape she got from the dispensary. Of the 7 only one is white (premature) aaand here we go. Popped one to see how strong it would be, trying the other 6 now. The seeds seemed smaller than normal, but I'm no expert.

I'd love to post the pics I took but I'm having some trouble uploading them. I'll try again later.
 

dstroy

Well-Known Member
I still can't upload pictures, but I guess that's an everyone problem and not just me.

I just spent a ton of money on amazon to get a chiller setup for my cloner and my two aero reservoirs. My cloner reservoir is hovering around 78f (I checked with my laser temp gun and my EC and PH meters) right now. It's only under a two foot 4 bulb t5, but I have to put it into a cabinet that can lock to comply with the law.

It's going to be closed loop (no cooling water touches nutrient water), using stainless steel immersion coils. Right now I've got the chiller coming, a cooler to put all the pumps in, two 25ft 3/8" coils and a small 3/16" coil for the cloner along with four pumps and coolant lines and fittings.

I knew it was going to be a bitch to create the "perfect" environment at home, but now I think I've got everything I need.

On a side note, I'm switching from contactors to SSR because I should have done that in the first place.

I also bought a 5 stage RO, because I don't need DI water, it's not going through electronic equipment. I got a chemical PH test kit to go with this because my meter (HM digital PH-200) isn't supposed to be used with RO water, so I'll buffer the RO water first and test it with the chemical kit and then when it's buffered I'll be able to use the digital meter.

Once I'm satisfied that the cooling loops I have will not leak I might put some water wetter in with the coolant but I don't think that's a good idea in case there ever is a leak.

The seed that I popped first has it's second set of true leaves coming out.

Three out of the other four that germinated have their first set of tiny true leaves coming out, the fourth's taproot is about two inches long but still too short to put in a neoprene collar.

I don't want to raise up the lights and end up with spindly seedlings.
 

dstroy

Well-Known Member
I make sure to wash my hands before touching anything to do with plants. We're thinking about using gloves on top of that.

I just changed out the water in the cloner, it's been about 7 days. No evidence of slime, everything feels clean.

My water comes out of the tap @ PH 7.5, it takes 0.6mL PH down to PH a gallon of water to 5.8

Per gallon:
0.6mL PH down
2.0mL Clonex
0.6mL Mammoth P

ppm comes out to about 450

I was getting some PH drift which is normal, and just correct the PH daily. 0.2mL PH down corrected my cloner water by about PH -0.18 each dose @ 2 gallons volume.

The established plant looks normal, root system is developing nicely.

Sprouts are dark green (like evergreen) and have bright purple stems.
 

dstroy

Well-Known Member
I've got the cloner and chiller setup. I'm running the cloner 1 min on 3 off. Lights are 2ft t5 4 bulb fixture 18/6.

Tomorrow I'm starting the dyna gro feeding schedule.

The first plant is showing signs of calcium deficiency.

Lessons learned are: clonex seemed to work fine for the first three weeks, but once roots are established and healthy I should switch to veg Nutes.

One out of the five plants is really picky.
 

Attachments

dstroy

Well-Known Member
I started finishing the lids for my buckets. That's going well. I've got some replacements for these solenoids in case any go bad.

The plants are looking better, stems aren't so purple anymore. I think I've got the calcium problem under control, not seeing any spots on new growth. I'm running dyna gro nutes at 1/3 recommended strength, roots are looking well.

Reservoir temp is right around 73f.
 

dstroy

Well-Known Member
3/5/2017
Plants are looking good, new growth looks normal.

I added low pressure switches (on when pressure is low, off when pressure is above the setpoint). I thought about adding pressure transducers, but they're not necessary and are expensive. I've got it set up now so that if one pump fails and the pressure drops in one of the lines that the pumping manifold will switch back and forth between the two lines and respective nutrient solutions to keep the accumulators charged.

I'm still working on a battery backup.

image1.JPG image2.JPG
 

dstroy

Well-Known Member
3/7/2017

res change today. 2 gallons.

I lowered the chiller target temp from 70f to 68f. This brought the cloner water temp to 71f, I think I might need a higher flow pump for the cloner's heat exchanger.

tapwater was
ph 6.71
ppm 458

added
per gal
8.3ml grow
3.3ml pro tekt
0.6ml mammoth p
0.5ml mag pro
0.4ml ph down

ph 5.8
ppm 1300

side note: I finished all the lids for my buckets, and built the ball valve tee thing that every bucket needs

to do:
build legs for the control box
finish the design on the pump manifold

image1(1).JPG image1.JPG IMG_2346.JPG
 

dstroy

Well-Known Member
3/9/2017
ph 5.63
ppm 1010 .5conv

added 0.4ml ph up to approx 2.5gal water

I raised the chiller target temp back to 70. I added 3ml of bleach to the coolant, which is completely separate from the nutrient solution. The pumps for all the loops are located in the cooler. I've got insulation for the lines coming from the chiller going into this and for the larger cooling coils.

IMG_2360.JPG

The stem of the plant on the right in the front has broken, you can see that when the terminal shoot came up out of the cotyledons it never completely broke through and constricted the stem right below the first node. This happened on 3/7/17 and the pictures are from today. It looks better than it did yesterday. Roots look good.

IMG_2355.JPG IMG_2357.JPG IMG_2358.JPG IMG_2359.JPG
 

dstroy

Well-Known Member
I forgot to mention that I also set the low pressure switches to come on at 65psi and go off at 80psi. I set the pump bypass pressure to 125psi, and the pump shutoff switch to 100psi. If the pump shutoff switch fails it will just pressurize everything to 125psi, if the bypass fails the physical safety valve I have opens at 130psi.
 

dstroy

Well-Known Member
3/10/2017
ph 5.5
ppm 1100

added 1/3 gallon distilled water

ph 5.53
ppm 900

added 0.4ml ph up

FIMmed front left and rear right plants.

The plant with the broken main stem looks the same up top, there was some root growth though.

image1.JPG
 

dstroy

Well-Known Member
3/11/17
ph 5.64
ppm 924

added 0.4ml ph up

terminal shoot growth is continuing on two FIMmed plants

3-11-17.JPG
 

dstroy

Well-Known Member
3/12/17
ph 5.7
ppm 950

added 0.3ml ph up

3/13/17
ph 5.5
ppm 1070

added 0.5ml ph up

plant growth is robust, pulled back fan leaves of #2 and #4 to open up the canopy for #1 and #5.
I think #5 is dying, I think I should cut it up into clones
IMG_2387.JPG IMG_2388.JPG .

I'm starting to worry about vertical space availability in the tents.
I really can't take these things out of the cloner for another couple of weeks. I separate the roots daily, and try to not let them ball up so I can pull them out of the cloner.
 
Top