DiY LEDs - How to Power Them

Want to avoid tax sending from Digi-Key to Europe?
Let's do group buy - maybe someone need's four pieces - Veros 29?

Would this driver be alright for a single Vero 29? Thinking about getting a few and running them at 2.1a.

Unless i can find the higher bin Cree CXA3070 but they seem impossible to find in smaller quantities.

http://www.ebay.com/itm/70W-Watt-High-Power-LED-Driver-AC85V-265V-50-60HZ-Waterproof-/380520934764?pt=LH_DefaultDomain_0&hash=item5898d0ad6c

The Vero29 at 2.1A would be about 38.6Vf, so that driver might not be a good match for it (would work with Vero18 )

This should work with the Vero 29 (and CXA3070), although I am not sure exactly how much current it will actually put out. I have tested the smaller brother of that driver (1.5A version) and it was very good, 1.45A and 89% efficient.
http://www.ebay.com/itm/70W-watt-led-driver-Power-Supply-AC-100-265-V-2100-mA-DC-20V-40V-Waterproof-IP65-/191076829554?pt=US_Lighting_Parts_and_Accessories&hash=item2c7d110972

The 80W version (2.4A) of this batch looks good at $28
http://www.ebay.com/itm/371178730397?var=640326403815

For $15 more you can get the Mean Well dimmable that puts out 2.1A
http://www.mouser.com/ProductDetail/Mean-Well/HLN-80H-42A/?qs=QHrm/geNk%2bg8XXUxOaNP3w==

Perfectdeal_us ran out of the 50W drivers again. She said 20 days until they are restocked. In the meantime here is one I have on order to test that looks promising:
http://www.ebay.com/itm/Great-LED-Driver-Power-Supply-Adapter-AC-100-240V-to-DC-24-40V-Waterproof-50W-/201221153486?pt=US_Lighting_Parts_and_Accessories&hash=item2ed9b722ce

Another cheap option with better feed back
http://www.ebay.com/itm/50W-DC28-40V-Power-Supply-AC110-260v-1500mA-waterproof-IP65-LED-Lamp-Driver-/261666621892?pt=LH_DefaultDomain_0&hash=item3cec8bc9c4

This one I ordered, tested and verified to be a good driver. Puts out 1.45A, power factor corrected and about 89% efficient. The only downside is these drivers all ship from Asia.
http://www.ebay.com/itm/171401815098?_trksid=p2059210.m2749.l2649&ssPageName=STRK:MEBIDX:IT

SupraSPL said:

The Vero29 at 2.1A would be about 38.6Vf, so that driver might not be a good match for it (would work with Vero18 )

This should work with the Vero 29 (and CXA3070), although I am not sure exactly how much current it will actually put out. I have tested the smaller brother of that driver (1.5A version) and it was very good, 1.45A and 89% efficient.
http://www.ebay.com/itm/70W-watt-led-driver-Power-Supply-AC-100-265-V-2100-mA-DC-20V-40V-Waterproof-IP65-/191076829554?pt=US_Lighting_Parts_and_Accessories&hash=item2c7d110972

For $15 more you can get the Mean Well dimmable that puts out 2.1A
http://www.mouser.com/ProductDetail/Mean-Well/HLN-80H-42A/?qs=QHrm/geNk%2bg8XXUxOaNP3w==

Click to expand...

Yeah thats what I suspected, there is another one that is a step up to 40v but it also puts out 2.4a.

I will probably end up going with the mean well driver, but it is always nice to look at other options. That driver from Great Led may be something I will look in to depending on how it performs.

Thanks again Supra!

Does anyone know a good source that compares 3000k vs 3500k spectrums and the main differences (pros/cons) between the two? They may be too close to notice a difference in my small 2 sqft setup.

The Vero spec sheet shows only a slight increase in output from the 3000K to the 3500K, 9900-10200 lumens (3% more light). Probably a slight increase in blue and a very slight decrease in the entire rest of the curve but we do not have a graph tof the 3500K to analyze. So since the efficiency increase is only 3%, it may be offset by the photosynthetic efficiency decrease. Also, I have a suspicion that the 3000K might make better buds (denser, less leaf?)

If you had a 3:1 ratio of 3000k to 3500k you think that would provide some suppression to stretching in early stages of flower?

SupraSPL said:

The Vero spec sheet shows only a slight increase in output from the 3000K to the 3500K, 9900-10200 lumens (3% more light). Probably a slight increase in blue and a very slight decrease in the entire rest of the curve but we do not have a graph tof the 3500K to analyze. So since the efficiency increase is only 3%, it may be offset by the photosynthetic efficiency decrease. Also, I have a suspicion that the 3000K might make better buds (denser, less leaf?)

Click to expand...

That is a good question, because the COBs would be spaced apart I am not sure the colors would have much of a chance to mix. You could add a blue diodes next to each COB on a separate string if you have overly stretchy sativa doms or if you are very height limited. Then you could turn it on during veg and during stretch. On the other hand if you have the height, the stretch might actually be helpful by increasing airflow and canopy surface area. My worst performers are always my indica doms they tend to get out vegged and shaded out by the sativa doms. So I have been learning to give vegging priority to the indica doms.

Ah yes that is a good point, didn't think that completely through

I'll probably end up doing a few small strings of red/blues if I have any problems with my only 3000k spectrum build. All I have are a bunch of heavy indica seeds so it will be fun to see how they react.

Someone pointed out that there are .5W (150mA) drivers available. So I got a batch and tested one this morning. $1.25.ea shipped, efficient and can drive a wide range of COBS at reasonable driver efficiency ~86%. These drivers make some sense for the Vero 10s, at 150mA the V10 dissipates 3.75W each and cost about $3/PAR W. If you ran one V10/driver you would get about 155mA and if you ran a pair, about 140mA.

You could also use stuff like this for cloning box, mother box, human-use lighting, under cabinet, porch lighting etc. You could create super efficiency lighting with Cree XML2/XPL at .5W ea. Saves a few watts you can put toward your grow

results:
@3Vf driver squeals and hates life, 30% efficient, 100mA output
6Vf, 58% efficient, 183mA output
9Vf, 70% efficient, 165mA output
12Vf, 80% efficient, 164mA output
15Vf, 84% efficient, 160mA output
32Vf, 85% efficient, 156mA output (CXA3070)
38Vf, 86% efficient, 154mA output
44Vf, 86% efficient, 150mA output
47Vf, 85% efficient, 146mA output
49.5Vf, 85% efficient, 141mA output (maximum Vf)


I zip tied the wires to the unit for strain relief and then heat shrunk it, mostly to cover the AC solder points. Crimped on 1/4" slide connectors and heat shrink over those.

So i have a large house plant in the living room that isn't getting enough light. I want to use a single Vero 10 to help it through the winter. I'm wanting to go passive so i was thinking a 6" slab of the 4.900" profile from hsu. Which of these drivers would be better suited:
http://m.ebay.ca/itm/161348468007?nav=WATCHING_ACTIVE&skus=Type:APC-12-350 TUV 9-36V&varId=460372051506
http://m.ebay.ca/itm/160927552473?nav=WATCHING_ACTIVE

At 300mA it would dissipate about 7.8W. That should be plenty you could probably go with even less. I use 1W LEDs on my aloe plants and it is too much light for the parts of them that get too close.

SupraSPL said:

At 300mA it would dissipate about 7.8W. That should be plenty you could probably go with even less. I use 1W LEDs on my aloe plants and it is too much light for the parts of them that get too close.

Click to expand...

So maybe just the generic driver instead of the meanwell?

O just noticed the drop down box, I was looking at the picture and saw it was 700mA. Yes you could use the mean well apc but efficiency seems low at 82-84%.

@SupraSPL hey thanks for getting back, whats the deal with the bin ?? what does it represent ??

Also just picking ur brain now the vero 10's whats the deal with those as a grow light, i see there cheaper than the 3070's.

I'm really looking for a decent solution to my lighting not a half measure, i want to build a good light that will perform the same or better than my HID ideally I would like to stay within the 600 - 1000w range as im restricted with my power due to location. In an ideal world i'd like to be a lot less than this going LED but if its a case of extra lights for more bud then so be it.


Can you point me at any grow journals using light movers and LED lights, I would be interested to see how they compare.

Thank in advance

SupraSPL said:

Hey Rosco, the jury is still out on the actual bin of the Ali CXA3070s. I have one on order to compare it against Z2, Z4 and AB and will report back when I find out.

Click to expand...

What is everybody's experience with ordering meanwell drivers (from europe)? I ordered 15 HLG-185H-C1050A and some other drivers, but delivery is suddenly posponed to early 2015 when first delivery was promised within 2 weeks (this week). Does meanwell have deliveryproblems or do I have a bad dealer?

I ordered a 185-c700a from jameco and they bumped the order from 2 days to 5 weeks. I think they have access to meanwell's stock of drivers they do not keep in stock but what you see on the vendor's website can get poached by other vendors.

[This is in US though unsure if same applies to eu]

happy75 said:

What is everybody's experience with ordering meanwell drivers (from europe)? I ordered 15 HLG-185H-C1050A and some other drivers, but delivery is suddenly posponed to early 2015 when first delivery was promised within 2 weeks (this week). Does meanwell have deliveryproblems or do I have a bad dealer?

Click to expand...

Ok so I'm fixing a guys Stealth Grow 602's, he says he has 12+ of these things but two have not lighting up fully anymore. I said I would take a look at them for $, I tested the drivers and they all work fine I tested the diodes. I marked the burned out ones, some of the dead ones still let the current through while others I will just add a bypass wire around them. These panels have (1) 30-36V 3.5A and (3) 20-24V 4A constant current drivers inside them. The (3) 20-24V ones drive 7 parallel strings of 10 red LEDs in series (4A / 7 strings = 571mA per string), the (1) 30-36V driver has 6 strings of ten LEDs (3.5/6=583mA), there are no resistors incorporated in this design.

On the one W/B driver there is 1 led on 5 strings burned out and 2 leds on the 6th string out. My question is if I bypass the burned out leds with a wire will this be ok or do I need to bypass it with a resistor or something to keep the strings even and prevent thermal runaway? I dont have the electrical knowledge @goud has so before I just add a wire I want to know if I'm doing the the right thing.


Also I understand that this graphic is 10 LED's running on 7 separate parallel strings
but what is this? see how the strings are not isolated in this second graphic
The other panel I opened is a version 2 of their circuit board and its like the second graphic. No diodes were burned out on this board but the drivers all need replacing. Does this have to do with the circuit design?

the second version is a sort of fail safe strategie for the most failure on leds, an open circuit.
based on Kirchhoff's circuit laws...