These are chat archives for CommonGarden/Grow-IoT

Oct 2016
Jake Hartnell
Oct 15 2016 00:12
Grow.js is a mess, but I've got test-thing posting messages to a script in another terminal.
But to do that I had to install it with Bower and add it to imports.html
Don't think that's too ideal...
Jake Hartnell
Oct 15 2016 02:16
Eventually, I think there show be two views: 1. a dashboard view: lot's of devices and charts for sensor data etc. and 2. a flow view: a wiring logic view like node-red but for more devices that the raspberry pi.
We hope to get some funding to help build the later.
Nice docs though.
Oct 15 2016 02:37
I am building my environmental sensor and remote power strip hardware now, I got prototype boards on the way from osh Park. I am shooting for a middle ground of people who already own individual timers and control units, or if they don't to add automated and timer controlled ventilation etc.
Not for light control essentially.
Commercial who uses Gavitas which have their own hardware controller etc.
In all my searching, I have yet to find a "real" remote viewable sensor hardware in this space that isn't just a 16x2 serial monitor using zigbee or some other short distance radio.
Oct 15 2016 02:44
Has to be remote controllable and viewable from different counties to be worth anything at all to me. It's not so much the logic of everything talking together, it's also just things working as intended.
Trying to do too many things in a piece of individual hardware, invites the Swiss army knife approach, everything is good enough, but nothings great.
And what that really means is nothing is truly essential if they do nothing GREAT.
Oct 15 2016 02:50
Take that ART DNIe timer you posted, I use alot of those.
Those are absolutely essential.
No timer comes even remotely close to their reliability and accuracy.
If a maker made replacement fails to fire even once in a million cycles, it's worthless comparatively
That timer is bulletproof. I have dumped gallons of nytrient solution over them, still just kicking along.
If you can replace that timer for less than $35 in hardware and have it be that reliable, and still make a profit after logistics etc, hats off.
But I wouldn't use anything not field tested for at least half a million cycles in that particular hardware piece.