Let's talk DIY lights


I will need to get a few more grows under my belt in my 4x4 tent to determine what I prefer/determine is optimal for me all other things being equal; this is probably an adventure I would consider next summer/fall. My first grow I filled the entire tent, and had way too little light. This next grow I am just doing two plants and basically committed one light to each plant (in a SCROG). I intend to do what you essentially said in your last sentence - throw two plants in the middle and leave room around the edges. I have lots of planned upgrades to the growing system that will determine how much of the usable space I will actually use in the future (switching from DWC to RDWC, etc.). In the mean time I figure now is the time to do research, learn, and decide down the road how I want to improve my yields and maximize my investment.


The more you know/understand the better off you’ll be @Bogleg, so feel free to ask away. As far as planning for next summer, the latest and greatest will probably change at least twice by then. So don’t start buying little pieces of the puzzle now. Your best bet will be to stash a few extra bucks here and there when you can until you’re ready.

Right now quantam boards, and citizen cobs are the hot topic on diy end. The hlg-550 retail $1050 currently has coupon code for 20% off, and diy build would run you around $600. They are getting ready to introduce new light, hence the sale. But that’s the typical margins you’re looking at with diy. For reference, you could pretty much do anything you damn well pleased in a 4x4 with a light like that. It should provide plenty of coverage wall to wall. 2 plants now, scrog in half with a couple of plants on other side, or whatever. You’d be good, and hassle free at a little over $800 wouldn’t be bad. If you’re a tight ass and don’t mind a little elbow grease, you could save $250 and build it when components are available.

If you’re more budget minded, building something like 200 watt cob bars is a pretty good option. You can usually get into a pretty good setup for about $1/watt give or take. If you want to give up a little efficiency to save on build cost, probably even a little cheaper. But that would give you the ability to build a bar or 2 and grow your 2 plants, an add another bar if you decided to go back to something like a full tent. Or even just build the one fixture into a square and deal with any shortcomings until you have a little extra cash. Then reconfigure the same cobs later if you decide to go a different route.

You’re only really limited by your budget, and how creative you want to get.


As in all things, I would ideally like to simplify the overall environment as much as possible - i.e. reduce the number of fixtures I am dealing with/managing. At some point I will squirrel away enough money for a second tent and break out to separate veg and flower tents; in that case I would probably move my current King Plus LEDs to the veg tent and look at something new for the flower tent. Part of my issue is making sure I have a growing system that allows me to easily manage the water/reservoirs. DWC with a full tent was tough at times. Once I figure that out, I can get back to using all of the available space and get lights that cover it.


I bet that was a handful! Several ways to go about that, but definitely not in my wheelhouse. From watching others, running flood system with external res seems like method of choice. But that’s just from watching others.

As far as the lights go, keeping up with the tech is half the battle. Building a light for a flowering specific room answers some the questions you’d eventually get to. You’ll want something that produces higher intensity in the red wavelengths to help bud development.


Yeah that’s the plan - RDWC/undercurrent system (I have been following @Snowman’s thread and really like what he built) with an external reservoir so I can easily add/change/modify the water.


Should make managing day to day activities easier.


Oh man it’s going to be nice! Just unzip the tent and it’s right at your feet. No more battling foliage abd branches!!! Whooo. Going to be adding the babies to the system this weekend! @Bogleg


@dbrn32 I wasn’t sure where to tag you so I went here. Since you have more experience than myself I found a driverless LED from amazon. I know they aren’t that new. Color spectrum seems acceptable. Except for IR and UV. I could add those later maybe. I’m thinking minimum 24 spread out for a 4x4 space. Maybe I order 1-2 to see what they’re like?
The top one is not driverless and found it after my original post.


Someone here is using them. @skgrower maybe? If nothing else he may know who.

I have little knowledge of them either way. Of the ones I’ve seen, they use a driver it’s just built into the chip. Good, bad, or indifferent, I wouldn’t expect to get the same performance out of them as the flagship models. But they’re a lot cheaper too, so they could very well stack up dollar for dollar.

I would see if we can locate whoever is using them and get their thoughts. Otherwise, I would say it wouldn’t hurt to grab a couple and try.


Thanks @dbrn32 I put 4 in my cart to give them a try, along with some heat sinks. If I can locate some driverless 3000K -2700K ones, I could make it a better spectrum. I’m not entirely sure how important IR/UV is, but I’ve seen those sold separately also. I figure 4 is a good start to see if it goes anywhere. Thanks for stopping by.


The uv and ir definitely isn’t important enough to worry about on a trial level, if at all.

We could go back and forth about uv, and not come to a solid conclusion. About the only thing solid you’ll find is with the addition of uvb, and even then there are some arguments about how and when to use it. To my knowledge, there aren’t even any affordable uvb options in led. You would have to go with a fluorescent reptile type of bulb.

The ir, or even far red wavelengths can be beneficial, but you’re talking about fine tuning a light spectrum at that point. It’s not like it will double your yield or anything. The biggest benefit I’ve seen proven from them is the potential to shorten a grow by a few days.


Has anyone caught any buzz on the new Cree cobs? Other than what’s posted on their webpage, of course.

@Dumme @latewood @HappyCamper @skgrower @MAXHeadRoom feel free to tag anyone that may have any info…


Which one are you referring to. I use cbx3590’s and absolutely love them.


I believe they are similar platform as cxb, but using metal base over ceramic and different string configurations. Would post link, but I don’t think it’s appropriate for forum rules. Here’s screenshot from their page though.

Have you tried any of the 90 cri? I just received a few, will be a bit before I can get them into a fixture though.


Those “driverless LED” COBs say right in the description that constant current drivers are recommended. I think they are only driverless in that a driver is not included in the price! If you try to drive them with a constant voltage power supply, very bad things may happen including thermal runaway that blows up the COB. At least us a small transformer that limits the current output of the power supply. And DO NOT wire them in parallel, ever. Very small differences in one COB to the next can send all the current for two or more COBs to a single COB and blow it up, Then the next lowest resistance COB gets all the current and blows up, and so on.

As for the price: 2100 lumens for $17.38. That’s 121 lumens per dollar.
Digikey has Vero 29 series that put out 14892 lumens for $22.58. that’s 659 lumens per dollar.

And Digikey may not even be the cheapest source for high power LED COBs.


@1BigFella The colored light probably won’t ever match up on a luminous efficacy basis. But I wouldn’t be surprised if the numbers were a lot closer if we were calculating ųmol/joule. If majority of the knock off cob is red, it will have a much higher photosynthetic efficacy than luminous efficacy. On a ųmol per dollar measure, I’d probably stick with name brand stuff other than cree. But I buy cree too lol, so there’s always that.

Using constant voltage drivers and wiring in parallel are more advanced methods that require some experience and electrical knowledge to implement safely, for sure. But it can and us done pretty regularly, by installing fuses. A ct can be used, but usually at an efficiency cost. I’m not super familiar with low level models, but I know our bigger models are just as if not more dangerous. They need to remain loaded otherwise ratio will go backwards and do completely opposite of the purpose they’re installed for the way I understand it. Is that accurate?


@dbrn32 Check this out, everything else on this grow is the same. 90cri on the left 80cri on the right.

Citizen 1812 cobs I think


Hey @Daddy just a couple of questions,
#1 are these clones or seed plants?
#2 did you defoliate the plant on left?
What is the difference in 80cri and 90cri?
Does the 90cri have less blue spec.( idk how to say this without being a dumba$$ .)is that the reason it stre5ched more? Or is it just a representation of how the 90cri plant looks to have more bud,and the 80cri more leaves?


@Daddy @Nug-bug

I think i know exactly who that grow belongs to, followed it from start to finish. Although I don’t remember for sure, I think they were started from clones.

It looks like the plants on left were defoliated, but I’m pretty sure both were treated exactly the same. The difference you’re seeing is the left side finished a little earlier. It was hacked and right side was let go for another 9-10 days I think. The 90 cri did stretch more, but yield was about the same due to 80 cri not being ready at the same time. The 80 cri did show significantly more bud swell in the additional amount of time it had.

The 90 cri actually has higher intensity in blue wavelengths than the 80 cri. Weird huh? It shows differently than just about everything we know and suspect. The reason for the stretch I’d because the 90 cri citi cobs peak deeper in the reds, almost 650nm. As opposed to like 620-630 nm with 80 cri. So while it has more blue, the deeper reds caused that stretch. I believe dawg said the difference was about an inch per week from late veg through transition, and then leveled off. Which seems to be supported by pics.

If you have read this entire thread, you can see we discuss the cri quite a bit. I think this grow supports the benefits of the 90 cri cobs. Like we concluded, higher cri isn’t necessarily better. But the deeper reds from going to 80 to 90 cri in most cases do look to have horticulture benefits. If you jump to day the 97 cri, you don’t see the same increase. You’re essentially giving up light intensity to create more yellow and green wavelengths at that point.


Thanks for the explanation @dbrn32 , I have read along with this thread from the start,but for some reason I can’t retain the material? Gonna need to go back through the topic , I’m assuming it will help me understand a Lil better instead of a post here and there.