Help on HLG-480/Vero29 build

Hi guys, first i'm sorry if there is some post like this but the amount if information here is HUUGE and i end up a bit confused,

i have a hlg-480H-54A, already bought it and i'm not 100% sure if what cob should i buy,

i thought about Vero29se, and my doubts are:

• 3000k, 3500k or maybe other?
• 80CRI or 90CRI?
• B or C? i should use B right? ~54v?

my plan is 4 cobs on ~120w ea with active cooling

Pedro Mello said:

Hi guys, first i'm sorry if there is some post like this but the amount if information here is HUUGE and i end up a bit confused,

i have a hlg-480H-54A, already bought it and i'm not 100% sure if what cob should i buy,

i thought about Vero29se, and my doubts are:

• 3000k, 3500k or maybe other?
• 80CRI or 90CRI?
• B or C? i should use B right? ~54v?

my plan is 4 cobs on ~120w ea with active cooling

Click to expand...

You should be using B version with that driver, the C is around 70V, too much for that driver.
Veros are an excellent choice, other alternatives would be luminous cxm22 or citizen 18xx-series.
Spectrum: 3000K 90 Cri or 3000k 80 cri would be the most flowerleaning spectrums, 3500k both cri's are more fullcycle. All should work ok both for flower and vegg, 90 cri might stretch a bit more in vegg but less in flower although all is straindependent.

4x120w on active heatsinks? First: 120w cobs are seriously intense and need a lot of distance to cannopy. A friend did it for supplementation to hps and got light burn no matter what he tried.
Second: on that driver you need to do paralell connections. When you connect in paralell you need to use Cobs which have the same voltage, and as voltage lowers with heat, you should have the same thermal management on each cob. If not you can get a cob that heats up and draws more power than the rest, and as it draws more power it also heats up more, drawing more power etc: thermal runnaway. Basicly: if your hoping to expand with a few cobs later you should use the same active cooling as the first ones, at least if you wanna be carefull and make sure. Or get new heatsinks for all chips. Id go for 6 vero bs, at 80w each, w passive heatsinks.

Rocket Soul said:

You should be using B version with that driver, the C is around 70V, too much for that driver.
Veros are an excellent choice, other alternatives would be luminous cxm22 or citizen 18xx-series.
Spectrum: 3000K 90 Cri or 3000k 80 cri would be the most flowerleaning spectrums, 3500k both cri's are more fullcycle. All should work ok both for flower and vegg, 90 cri might stretch a bit more in vegg but less in flower although all is straindependent.

4x120w on active heatsinks? First: 120w cobs are seriously intense and need a lot of distance to cannopy. A friend did it for supplementation to hps and got light burn no matter what he tried.
Second: on that driver you need to do paralell connections. When you connect in paralell you need to use Cobs which have the same voltage, and as voltage lowers with heat, you should have the same thermal management on each cob. If not you can get a cob that heats up and draws more power than the rest, and as it draws more power it also heats up more, drawing more power etc: thermal runnaway. Basicly: if your hoping to expand with a few cobs later you should use the same active cooling as the first ones, at least if you wanna be carefull and make sure. Or get new heatsinks for all chips. Id go for 6 vero bs, at 80w each, w passive heatsinks.

Click to expand...

First of all, thank you for the anwser. I'm just asking myself some points, hoping you could explain for me.

I understood what you said about thermal runnaway, but considering that I live in Brazil, a fucking hot place all the time kkk, you think the passive heatsinks will handle? besides that the cobs costs the double for me. I'd prefer to buy fewer, if possible.

You said about the distance to canopy, how many feets do you think I need? Cause I have a big space, like 6.6 feets height.

Maybe the space I'd lost justifies the cost I'd have for get more cobs as you advised

Thank you, man!

About heatsinks: get active or passive as long as you use the same or similar performing heatsinks on all cobs; having one run hotter than the other is what can make thermal runnaway an issue, not how hot it gets in brazil

About lightspread/distance: depends on your space/optics and more: here 100w cobs +reflectors were to intense for 50cm height. Many here would recommend 35-50w per cob, 1 per 30x30 cm at + 30cm

B cob is a good match for that driver. a -48A would give you more output and is a little more failsafe

turn the voltage screw of the 54A down to just above where your cobs start to dim. that way if you lose one or more cobs voltage will be limited and you wont lose em all

CobKits said:

B cob is a good match for that driver. a -48A would give you more output and is a little more failsafe

turn the voltage screw of the 54A down to just above where your cobs start to dim. that way if you lose one or more cobs voltage will be limited and you wont lose em all

Click to expand...

Thank you!!!
It is something very useful to know.

If I make a circuit with a system that turns off every cob if one fails would this problem disappear?

Pedro Mello said:

If I make a circuit with a system that turns off every cob if one fails would this problem disappear?

Click to expand...

youd be overthinking the solution big time

CobKits said:

youd be overthinking the solution big time

Click to expand...

but do you think that my yield would be very prejudicated using the voltage lower than the maximum of 54V as you suggested?
In this case, the best way to solve that would be to buy more cobs, like run 8 or 9 for these driver?

Pedro Mello said:

Thank you!!!
It is something very useful to know.

If I make a circuit with a system that turns off every cob if one fails would this problem disappear?

Click to expand...

You would need quick blow fuses on every + wire or a thermal cutoff on every heatsink. Do as cobkits sed and just turn the voltage pot down. This will limit the current the cobs can pull.

Using CV +CC A type drivers requires some planning.You should have got the 48A type imo.

Pedro Mello said:

but do you think that my yield would be very prejudicated using the voltage lower than the maximum of 54V as you suggested?
In this case, the best way to solve that would be to buy more cobs, like run 8 or 9 for these driver?

Click to expand...

Instead of 4 Vero29 at 120w you should use 6 Luminous CXM22(19$) at 80w. Maybe cheaper when you pay additional custom fees as to buy them directly in brazil.

Randomblame said:

Instead of 4 Vero29 at 120w you should use 6 Luminous CXM22(19$) at 80w. Maybe cheaper when you pay additional custom fees as to buy them directly in brazil.

Click to expand...

the problem is that I've already bought 6 veros 29, gonna run them at 80w ea aprox
just cogitating if I should buy 1 or 2 more...

what do you think?

thanks for your time, man!

At 80 watts luminous and veros are about the same in efficiency. Just make sure you use all the same type of cob if you are using paralell connections. And same heatsink solution.

Pedro Mello said:

the problem is that I've already bought 6 veros 29, gonna run them at 80w ea aprox
just cogitating if I should buy 1 or 2 more...

what do you think?

thanks for your time, man!

Click to expand...

The more the better! More COB's on the same driver always means more efficiency and with lower driven COB's you can reduce the distance which means less light gets lost on the walls. With less distance you get also much higher PAR readings from the same light. 480w at 24" would be ~700-800μMol/s/m², 480w at 18" would be ~900-1000μMol/s/m² and with only 12" distance it would be ~1200. With 80w you need probably 18-24" with 60w per COB 12-18" would be enough.
You need to look at it from another point of few..
In the end, higher upfront costs always pay back hella fast! You're investing $100-200 more in COB's but you've got it back in better yield after the first run. How much is a zip in brazil? Believe me, saving on light means in the end only you lose potential yield ... and that every time!