Automated driver dimming, smart phone

hour

Well-Known Member
I built a few versions of these for multiple veg tents I had. Total cost per unit is like $4.50. Sadly I can't find the code I was running on these, and I built them 2 years ago... so I need to check out a bunch of flash drives for a backup. Each box supports two drivers or driver groups (you could connect multiple dimmer leads together and then pigtail in to one dimmer lead).

I've dismantled most of my setups, but have a single 4x4 flower tent running with a 480 (light 1) and 185 (light 2) meanwell driver, so I'm using both jacks on the little box to control them independently. Some variants had headphone jacks on the side to read temperature from DS18B20 sensors. This one uses a single DHT11 sensor for temp/rh. The DS18B20s are more sensitive but I don't need that level of accuracy. All units can override dim settings if temperature exceeds a limit, which will show up on the app. It also triggers a push notification to my phone if a temperature is exceeded for a defined length of time. If my AC fails or something I don't have to turn the light completely off, it'll just dim, tell me, and keep record of it.

I wrote three modes in to the firmware. Fixed, sweep, and random. Fixed is what you'd assume, a set level. Sweep will go between the high and low wattage settings back and forth. Random gently bounces around between the high and low wattage, sometimes going bright, then brighter, then dim, then bright, dim dim, bright, you get it.. Completely random and is pretty cool to look at. Like clouds passing overhead. After a random time at a certain level, it'll gently ascend/descend to the next setting in a proportional timeframe. Not jerky.

I initially built these for veg tents and found that using random or sweep really helps with acclimation to LED.

Phone automatically switches 'Via Network' off if I leave my home network. If not at home, the polling period for new data is increased. Locally I get new data every 3 seconds.

Setting up the wattage was tedious. I made a calibration mode that loops over the steps (values 0-255, steps of 17 = 11 hard steps). Using a kill-a-watt, I got the wattage value for each driver at these different steps. Even if the driver is sweeping during a call to get data, it gives me an accurate wattage +/- 2 or 3w.

Settings persist if powered off, and you can tell it to run at a low wattage at a certain time for X minutes to achieve a soft startup of all your drivers when the timer kicks on.

Anyway.. just made the phone app a little more plain and remembered I've never really shared this anywhere.






 

Renfro

Well-Known Member
Hey, if my lights use 3 drivers each, can it be used to control all of them at the same setting? I know that wiring three dimmer lines in parallel requires an adjustment resistor.
 

hour

Well-Known Member
Hey, if my lights use 3 drivers each, can it be used to control all of them at the same setting? I know that wiring three dimmer lines in parallel requires an adjustment resistor.
Not sure what happens with three, and can't remember if I noticed anything weird stacking two. Like maybe each driver independent is 228w at the wall full blast, dimmer set to 0 (full brightness, 255 being least bright). You'd expect 228*2w or there about if you paralleled them, but you're saying you'd get a lower reading? And worse as you add more drivers to the same channel?

I don't have three identical setups to test with or i'd give it a whirl. It's easy to add more channels though and would pbb be easier than fucking around with a resistor.. just drill another hole - solder jack, wire it up further down the breadboard identical to the first one. Modify firmware a little to include a second, third, fourth identity for the lead. And iterate over an array of leads for returning status (I return my data in JSON format). Since my units were built for veg tents that lined a wall, I used one per two tents (single 185 driver in each). I never wanted to build one with more than two inputs because of having to use 2.1x5.5mm extensions. Those definitely caused squirrely results cause of wire resistance, I guess.
 

Airwalker16

Well-Known Member
Think he just means each driver hooked up to the same main AC and controlling all 3 together in unison as if say 3 240's or 320's were used to run strings of 4-5 cobs or 2-3 boards, each.
Not sure what happens with three, and can't remember if I noticed anything weird stacking two. Like maybe each driver independent is 228w at the wall full blast, dimmer set to 0 (full brightness, 255 being least bright). You'd expect 228*2w or there about if you paralleled them, but you're saying you'd get a lower reading? And worse as you add more drivers to the same channel?

I don't have three identical setups to test with or i'd give it a whirl. It's easy to add more channels though and would pbb be easier than fucking around with a resistor.. just drill another hole - solder jack, wire it up further down the breadboard identical to the first one. Modify firmware a little to include a second, third, fourth identity for the lead. And iterate over an array of leads for returning status (I return my data in JSON format). Since my units were built for veg tents that lined a wall, I used one per two tents (single 185 driver in each). I never wanted to build one with more than two inputs because of having to use 2.1x5.5mm extensions. Those definitely caused squirrely results cause of wire resistance, I guess.
 

hour

Well-Known Member
Think he just means each driver hooked up to the same main AC and controlling all 3 together in unison as if say 3 240's or 320's were used to run strings of 4-5 cobs or 2-3 boards, each.
Idunno, he was talking about dimmer lines in parallel... I run a 185 and 480 pigtailed together in to a male power inlet like on the back of a computer tower PSU. They both get power when the timer kicks on, dimmer operates them independently. Something tells me he was talking about dimming getting boofed when you parallel a bunch of dim leads together.
 

Airwalker16

Well-Known Member
Idunno, he was talking about dimmer lines in parallel... I run a 185 and 480 pigtailed together in to a male power inlet like on the back of a computer tower PSU. They both get power when the timer kicks on, dimmer operates them independently. Something tells me he was talking about dimming getting boofed when you parallel a bunch of dim leads together.
I don't understand how it can control the drivers independently when they both connect to the same power cord.
 

Renfro

Well-Known Member
In my situation I have 3 drivers per light and his system uses two inputs. Ideally I would want to be able to dim each driver independently but I would settle for all three together. I really do want to incorporate this tech into my light build if it's doable and reliable.
 
Top