LED headache... what would you do?

[ruwtz]

I have a 3.5 x 7ft table in a brand new veg room with a pair of MarsII 400's. And yes I saw the shonky rep, but I also saw plenty of people happy with them too for veg only and they were affordable ($379/pair on Amazon). I figured they might be the best of the worst of the Chinese blurples. They cover the table ok, but could be better and I have considered adding a third panel or T5 to improve coverage. At capacity I expect 36-40 plants on this table, LST'ing + moving out at around 24" tall / 5weeks, but that is probably ambitious or stupid or both.

Hanging around here i've learned a lot about cobs over the past few days and now i'm thinking I probably could have done better than what I have with the same money by improving light/heat efficiency ratio and saving $$$ at the wall.

As I see it, there are three choices at this stage:

[1] STICK - make it work, see how I go with the 400's, hope nothing craps out in the first month and get a bunch of cycles out of them before upgrading to cob;

[2] TWIST - strip 'em now and repurpose everything in the 400's with a DIY cob install, a la @Growmau5

[3] RETURN and spend the cash on a DIY cob array instead. The 400's are currently set up but unused as seedlings are still in the nursery. I hear MarsHydro do everything to resist a full return, especially since nothing is defective at this point, so I might need to be smart as to my reasons if I go down this route.

What would you do?

 

[Growmau5]

my hack video was sort of a solution to making the best of a bad situation with 1-2 year old mars panels with blown diodes. So #2 would not be the best solution since the heatsinks in the mars arent very robust and the drivers are likely mid to high 80% efficient.

I would try to return them, $379 can go a long way for diy parts although you are likely to spend double that to outfit your space with high end cobs.

if they wont return them, try a run with them. mars will grow some good quality bud, they just probably wont be as productive on the weight side as your other options.

 

[ruwtz]

my hack video was sort of a solution to making the best of a bad situation with 1-2 year old mars panels with blown diodes. So #2 would not be the best solution since the heatsinks in the mars arent very robust and the drivers are likely mid to high 80% efficient.

I would try to return them, $379 can go a long way for diy parts although you are likely to spend double that to outfit your space with high end cobs.

if they wont return them, try a run with them. mars will grow some good quality bud, they just probably wont be as productive on the weight side as your other options.

Good advice, cheers. Funny I was just watching more of your starter vids and trying to calculate a DIY setup for my needs. Working out PPFD for a light is way over my head (SPD graphs, conversion factors etc - mind blown), but I see people with good results and high efficiency scaling incrementally upwards to 250-300ppfd by the end of veg using dimmable 36v cxb5390's at 700ma every sqft.

Am I generalizing correctly there?? Seems like heck of light for my fairly small veg table, but I could be way off trying to figure this out with my limited brain capacity.

I've spent everything I have on this new op so pretty much out of budget at this point, but research is free so here I am working out next steps.

 

[weed-whacker]

return, although u won't get enough cob for that price, although probably still more light than u have now

try some citizen cobs and just power them hard, as u can afford more cobs add them in and lower the current

 

[dandyrandy]

One downside to cobs are growth rate. No jokes aside I went from 4 Mars 300 old style several years ago. My first diy was Vero 29's which I still have. Now 16 3590's. I really have to manage the canopy.

 

[testiclees]

It's super optimistic to hope for good results with 800 mars watts in

One downside to cobs are growth rate. No jokes aside I went from 4 Mars 300 old style several years ago. My first diy was Vero 29's which I still have. Now 16 3590's. I really have to manage the canopy.

same. I twisted, beat, tortured the fuck outta this kali mist to not take over whole grow area. V29 & 3590

 

[dandyrandy]

It's super optimistic to hope for good results with 800 mars watts in

same. I twisted, beat, tortured the fuck outta this kali mist to not take over whole grow area. V29 & 3590

Nice. Long bloom on that one?

 

[bertaluchi]

Just add a couple cobs and keep your Mars lights. I use a mix of CREE COB and Bridgelux 5watt diodes. Works well.

 

[In Shape Vet]

anyone here start with led and finish off with HID when penetration is wanted?

 

[testiclees]

Nice. Long bloom on that one?

I hope so. The breeder states up to 90 days. I'm on day 56. Really not too easy to manage a prediction but 30 more days would not surprise me.
It grew so fucking crazy I had to move into a 15 gal after 3rd week of flowering. It was completely rooted up.

 

[testiclees]

anyone here start with led and finish off with HID when penetration is wanted?

HID isnt going to improve "penetration" if you're running multiple light sources like cobs or CMH in a highly reflective environment.

 

[ruwtz]

Im not about to roll over my Gravitas in flower any time soon, so I'm just talking veg here and the problem of explosive growth is a problem I would love to have.

I'm very close to returning these guys. Very close. If this was your 3.5x7ft veg table what cob array would you be putting into it? Still getting my head around ppfd etc but the electrical side of things is easier to me (wattage, drivers etc)

 

[ruwtz]

Great input here from @dandyrandy @testiclees @Growmau5 @weed-whacker - thank you all!

Obviously I couldtn't resist spending the rest of the day reading, learning and trying to make decisions.

Props to @bggrass for knocking up the COB LED calculator which has simplified a lot of numbers I could not have otherwise reached by myself.

Now i'm ready to spend some cash here so lets assume that side is not a problem until it becomes one

If you don't mind I'd like to share what I got out of the calculator. I did post this on the calculator thread but it doesn't look as busy as it deserves to be over there.

So for a 3.5x7ft flood table veg room only (total 24sqft) assuming no other source as i'm ditching these MarsII panels.

Speccing out CXB3590 36v 5000k + 4.6 profile heatsink

8 of these at 700ma gives me 262vf, 305w at the wall, 122par/watts, a ppfd of 244 and 74% cob efficiency.

Does this sound like an improved setup for my veg table? I was thinking a 6ft bar with 2 rows of 4 so each cob covers 3sqft and dimmable so I can work up to maxing out 244ppfd by the end of veg.

Do you think this is enough???

 

[ruwtz]

I'm guessing its not enough current and I could double this up to 1.4amps for 279f, ppfd 461, 230 par/watts and 65% efficiency.

Better?

@ttystikk has been good to me so far so maybe he won't mind me dragging him into this too?!

 

[ttystikk]

I'm guessing its not enough current and I could double this up to 1.4amps for 279f, ppfd 461, 230 par/watts and 65% efficiency.

Better?

@ttystikk has been good to me so far so maybe he won't mind me dragging him into this too?!

Well I confess I'm not really up on the finer points of roll your own chip and amperage combinations. I had one specific type built and I know that one and very few others.

 

[ruwtz]

Well I confess I'm not really up on the finer points of roll your own chip and amperage combinations. I had one specific type built and I know that one and very few others.

Fair enough. As you were.

 

[ttystikk]

Fair enough. As you were.

Now when we talk applications...