Anyone using GrowLink Pro on commerical scale?
I have some questions abt your opnions regarding climate control using their BACnet module, their new climate senor (ESM-2 from E+E), TDR mositure sensor for rockwool (from Acclima), and control software.
I love that you have me blocked cause I’ve been running grow link for multiple runs, and love sharing my experiences.
The TDR sensor is built for rockwool and other fast draining media, people have actually complained about the TDR in coco.
The environmental sensor is accurate enough to keep me happy and give a reasonable VPD estimate.
Control is pretty seamless across BACnet. Everything is centralized. The UI is super clean, easy to program, and integrate journaling with data, pictures from integrated cameras etc. If you try to manually adjust your system from a device controller growlink will reset it’s config the next poll.
The controller build is kinda comical. From just glancing at the PCB I can tell it was either a particle.io or onion.io build. You can actually get all the sensor modules for a good amount less from the manufacturers. I can post links later this weekend. The relays and connectors are kinda jank. The software makes it all worth it though. Their fertigation skid is awesome and I wish I could afford one.
I unblocked you the other day because I reacted too strongly before 
. I responded that way bc of a beloved family member’s life-long condition. I have given you a few thanks for the posts you’ve made in the past few days 
Thanks for your insights!
Do you have any issues or concernes with GrowLink’s TDR sensor? Interestingly, I confirmed with GrowLink that they created a volumetric water content to voltage (VWC/V) curve specifically for Grodan’s Gro Block Improved line. And that’s the VMC/V curve they’re using for all rockwool now. I believe they created a VWC/V curve for coco coir, but it needs to be enabled on the backend.
I also spoke with Grodan about their updated 6-prong TDR sensor. The Grodan TDR sesnor has considerably greater accuracy and measures a greater media volume than the 3-prong TDR sensor GrowLink uses. I will use the GrowLink TDR sensor to control irrigation events, crop steering, and for analytics, and Grodoan’s e-Gro platform (TDR sensor and data logger) for advanced irrigation analytics not offered by GrowLink, for example, its irrigation demand predictor.
I will use Grodan’s e-Gro TDR sensor and platform as a validation check on GrowLink’s TDR sensor. I expect some accuracy drift from GrowLink over time and under high EC caused by strong dry backs (for example, during drought stress in late flower to increase cannabinoid biosynthesis). The e-Gro TDR sensor should offer greater accuracy under drought stress and improved stability over time.
Do you use the prong insert tool to pre-make holes for the TDR prongs? I’m concerned about skewing the TDR measurement accuracy by changing the prong inter-distance when inserting the prongs without pre-making holes.
Have you created a VWC/V curve for your rockwool brand/line to verify GrowLink’s VWC/V accuracy? If not, once we get ours installed, if you use Grodan’s GBI line I’m happy to share the VWC/V curve I’m going to create with you.
Do you have the ESM-1 or ESM-2 climate sensor? And did you get the radiation shield? If so, did you get the aspirated (fan) or non-aspirated model? If you have the ESM-2, have you verified its accuracy with NIST certified or otherwise high-accuracy temp and RH sensors? They claim the ESM-2 addressed all the issues with the ESM-1, including the poor accruracy stabilty over time. I confirmed that I could use the reserach grade Apogee temp and RH sensor instead of the ESM-2, which I may do if I don’t like the ESM-2.
How many climate sensors do you use per room? And how many would you use if you wanted to increase? I’m using two-tier mobile carriages, with eight carriages per flowering room. So I’m starting with four climate sensors per room, but I may increase it to six. I originally planned for eight per room, but that’s probably overkill, lol.
Also, you may be stoked to learn, if you don’t already know, that GrowLink is updating their software for full integration with the Apogee infrared radiometer (canopy temp measurement). I ensured that fact before I agreed to give them a few hundred grand. Using the Apogee sensor, they will track canopy (leaf) temperature and use the canopy temp to calculate leaf VPD correctly. Especially considering leaf temp can typically exceed air temp by 2-4’F under high PPFD and high CO2, causing a substantial increase in VPD, and 6-10’F(!) under water stress, which wildly increases VPD. While they claim that had been long planned, I told them I wouldn’t use Growlink without it.
They also claim I can control a BACnet-enabled dehumidifier based on VPD. Pretty freaking cool!!! So air temp would use my setpoint (say, 80’F), and the humidity would depend upon my VPD setpoint (as affected by leaf temp).
That’s great to hear about BACnet. The Fujitsu AC/heat pump units purchased prior to my arrival are BACnet-enabled. By the way, I emailed Subcooled today to ask if they can offer BACnet control for their 550 of 705 subcooled dehui units. Their dehui use an AC condenser unit, so they’re not blasting hot air into the room.
Do you find GrowLink BACnet control provides climate stability that suits your needs?
I use the ESM-1 and probably why it falls into the “good enough” category.
I use a coco config for crop steering.
Between BACnet/modbus and api, there really is no magic you can’t create. I run mine on Mitsubishi heat pumps, so I don’t have a huge subset of to judge, but the stability of the system is amazing. I actually run mine off am APC batter backup and never go down.
I would probably use an enviromental sensor each tray at the least. Maybe even each light if you can afford it.
Interesting. That’s beyond my understanding and experience. I’d love to learn more 
That didn’t even occur to me. Yes, please! I know the manufacturers, but I didn’t think about buying the sensors directly. I could increase the number of climate and TDR sensors per room without adding an extra $25,000-ish to the proposal. With that in mind, I bet I can reduce the proposal cost by more than $40,000!
For sure! Are you using Dostatron or something else for stock tank injection? And are you using mainline injection for direct fertigation or batch tanks? I have a ten-injector skid planned to use different ionic ratios by growth phase, with a veg/mom profile and a flowering profile. With that number of injectors, I can modify the ionic ratios in the early and late flowering stages by using separate SOP (purified in my lab before dissolution), CaSO4, and MgSO4/NH4SO4 stock tanks. I’m using one primary batch tank per room.
Nice. I appreciate that you shared your experiences and opinions. It’s very helpful.
Where are you located? Once I have our GrowLink installed you’re more than welcome to stop by and check it out. I bet you could show me a thing or three!
I personally use a peri-pump setup, in a hot/cold tank setup with sensor controlled tank valves. It was the most affordable way for me to get on the fly fine control over my feed. I hate plumbing out dosers and would prefer to just cook batches and send em out.
Sweet. That sounds rad. 
I’m excited to run the H.E. Anderson injectors. I have only used Dosatrons and Etatrons so far.
Absolutely wish I could afford their skid. Those injectors are so awesome.
Yeah, I’d like to, but with 104 LEDs per room, I would need a few modules because I think they max out at 32 senor inputs? Plus, it would be crazy expensive.
I would like to have two climate sensors on each tier of our eight 26’ long carriages in each flowering room; for 32 climate sensors per room. When I asked the GrowLink rep to quote me on that, he laughed and said it’s overkill. I’m not so sure, considering it’s a tiered grow.
Considering you will show me where I can buy them directly, I will def start out with at least two sensors on both tiers of 3 carriage rows. Giving me at least 12 climate sensors per room.
I mean I guess you could call it over kill. I just call it accurate. Every piece of equipment is a variable in the equation. You are running a really large room. 50kw is usually the max I personally see per room/bay.
Make sure it does exactly what you want out of the box. I’ve only used one in a greenhouse and they had to do a lot of what they consider custom stuff ($$$$) to arrive at something serviceable. Edit: the last time I worked with one was 2 years ago, things may be different now…
I don’t know what their component lifespan is compared to a real PLC. For example Eden uses a Siemens PLC, true industrial equipment.
We do all of our control PLC’s in house. The Logic for a grow is actually very simple, even to control injectors. There are a bunch of freelance ladder logic guys out there for reasonable prices, you may want to get a quote. Being able to change logic to work exactly like you want can be very helpful.
I love this, I use Honeywell and Siemens. They have been making controls for over 50 years. No green tax and products that work as advertised.
I feel like for a majority of operations, going with a custom PLC build, is going to be far more expensive, and you are definitely going to invite OpEx unless you have in house engineering or competent IT. Alot of the real work from grow link comes out of it’s analytics and ability to cross platform. Things that are far more difficult and expensive with PLC’s when your aiming for a polished user friendly end product.
Up front costs are balanced by the quality of the end product. 
You can have 10 reliable cars and one quality car that really stands out for something above and beyond. I can accept the fact that a simple logic controller is going to be a bit more reliable than a microprocessor/microcontroller but we can solve for almost all of the microprocessor issues by adding batteries and solid state memory.
Gate arrays are wonderful at doing a specific task over and over. But that is litterally the extent of their abilities. To do anything else you need to add components and integrate them appropriately(microprocessors). A majority of the time adjustments can not be made easily or on the fly, and if they can , you put alot of time money and effort into a custom built control system you are always on the hook for support.
Don’t get me wrong PLC’s are great, and I use them for control in places like my crypto mine, where inputs are pretty basic. Temp/humidity. I feel like it’s a really big ask to bring everything growlink has done into a polished solution using off the shelf PLC and a custom install.
I totally understand CapEx/OpEx, anything custom is going to inflate your OpEx side by requiring specialized support.
Thanks for all of your comments!
At this point, lead time, ongoing and immediate customer and product support, detailed documentation, a user-friendly GUI, clean analytics, and remote/cloud access are paramount for us. I’m sure a custom PLC solution would be better because it puts me in complete control, and I can choose what features and Logic I want. But unfortunately, I know next to nothing about PLCs, even though I have used them in the past.
@vortal seems to have a solid argument about the unforeseen issues and costs of trying to create a GrowLink clone using PLC. But in truth, all of that is way above my understanding. I know I could learn about and get up to speed on what you guys are discussing, but I don’t have the time or bandwidth to do so while trying to finish all designs, plans, and SOPs for everything myself.
I’m not sure how long it would take a ladder logic freelancer to create something similar to GrowLink (meaning all the features of GrowLink software). I assume it’s not a quick process, especially if the freelancer isn’t familiar with growing or what GrowLink offers? I guess I would have to tell the freelancer what I want (easy) and provide information to complete the task (difficult considering I don’t understand PLCs?).
@emdub27 Do you have any links you can share where I can learn about ladder logic (a new term for me) and find freelancers?
I completely agree. Once I know the cost for the ESM-2 climate sensor with radiation shield directly from E+E, or a distributor, I can decide on the final quantity of sensors per room. I prefer to max out the available inputs, which I think is 32 (with a single module) per room?
Automation direct has several videos on youtube if you want to educate yourself. I run automation direct Domore plc’s for my personal projects. Ladder logic is basically an electrical line diagram, so if someone on your team is proficient with electrical, they will probably pick it up the quickest. The cool thing is it is only limited by your imagination. We datalog every single parameter once per minute.
The main reason I chose to go the custom route was because I was unhappy with available options. For instance when I originally dealt with growlink, controlling light deprivation by temperature and vpd in addition to time was a novel idea to them. I’ve experienced similar limitations with Eden and Trolmaster, they just don’t quite do everything I need. The PLC controls everything from grow all the way through dry and cure.