cocogrower
Member
4000 k's could be a better option, they lack a bit of 440, but that could be added! So many options!
Bueno Grows w/ DIY COB LEDs
- Thread starter Bueno Time
- Start date
I realized that I forgot to hook up the Kill A Watt to my COBs after swapping drivers from MW LPC-60-1400s to the $13 JHX drivers.
CXA3070 3000K Z2 & Z4 bin one of each
w/ MW LPC-60-1400s 0.63 PF 121W 1.6A 191VA
w/ JHX "1500mA" drivers 1.00PF 119W 1.0A 119VA
Going to measure the actual driver output current with a multimeter too eventually just out of curiousity. I think the output current is a little higher on the JHX drivers causing the wattage consumed to be nearly the same as the MW drivers even though these JHX should be a few % more efficient than the MWs.
I never measured anything with a multimeter on the MWs but Im getting 86% efficient calculation for the MW LPC-60-1400s considering at 1400mA each COB runs at ~52w and with a pull of 121 watts at the wall plug for the 2 COBs on the MW drivers 104w is 86% of 121w pull so thats roughly the efficiency of the MWs I believe they are listed at 85% or 86% efficient so that seems right on the money.
Many of the $13 ebay drivers JHX brand labeled 1500mA were tested by a member on RIU the JHX drivers put out 1420-1460mA once warmed up and are ~90% efficient. So if we assume 90% efficiency (for now until I actually measure it) then the 119w pull from the wall with the 2 JHX driver powered COBs then 119w x 0.9 (90% efficiency) = 107w so maybe I gained 3w at the COBs (1.5w each) but still pulling 2w less at the wall due to the increase in efficiency.
I dont know kind of makes sense to me right now but I want to test the actual current out to see if this is true.
Either way its not going to effect the grow at all its just something Im curious about and its nice to know exactly what everything is actually running at.
Even if there is no real substantial difference in power draw at the wall (1W per driver) the real difference is in cost effectiveness with the $13 JHX being ~50% of the cost of the MW LPC-60-1400 also very important for a large build or important when trying to keep the current draw down on a circuit the JHX drivers perform much better since they are power factor corrected each driver pulls 0.5A instead of 0.8A with the LPC-16-1400s, that is a very significant difference when running multiple drivers or when your circuit your grow is on also powers other higher current household items.
CXA3070 3000K Z2 & Z4 bin one of each
w/ MW LPC-60-1400s 0.63 PF 121W 1.6A 191VA
w/ JHX "1500mA" drivers 1.00PF 119W 1.0A 119VA
Going to measure the actual driver output current with a multimeter too eventually just out of curiousity. I think the output current is a little higher on the JHX drivers causing the wattage consumed to be nearly the same as the MW drivers even though these JHX should be a few % more efficient than the MWs.
I never measured anything with a multimeter on the MWs but Im getting 86% efficient calculation for the MW LPC-60-1400s considering at 1400mA each COB runs at ~52w and with a pull of 121 watts at the wall plug for the 2 COBs on the MW drivers 104w is 86% of 121w pull so thats roughly the efficiency of the MWs I believe they are listed at 85% or 86% efficient so that seems right on the money.
Many of the $13 ebay drivers JHX brand labeled 1500mA were tested by a member on RIU the JHX drivers put out 1420-1460mA once warmed up and are ~90% efficient. So if we assume 90% efficiency (for now until I actually measure it) then the 119w pull from the wall with the 2 JHX driver powered COBs then 119w x 0.9 (90% efficiency) = 107w so maybe I gained 3w at the COBs (1.5w each) but still pulling 2w less at the wall due to the increase in efficiency.
I dont know kind of makes sense to me right now but I want to test the actual current out to see if this is true.
Either way its not going to effect the grow at all its just something Im curious about and its nice to know exactly what everything is actually running at.
Even if there is no real substantial difference in power draw at the wall (1W per driver) the real difference is in cost effectiveness with the $13 JHX being ~50% of the cost of the MW LPC-60-1400 also very important for a large build or important when trying to keep the current draw down on a circuit the JHX drivers perform much better since they are power factor corrected each driver pulls 0.5A instead of 0.8A with the LPC-16-1400s, that is a very significant difference when running multiple drivers or when your circuit your grow is on also powers other higher current household items.
cocogrower
Member
I'm still trying to grasp the details of this last posts. I do love that you are so serious about the technical parts of this project, and that you are sharing your findings! Do you know if the JHX driver is dimmable?
thank you bueno and cocogrower for your very useful info. i have been looking for a suitable cob, and i have found this one: http://si.farnell.com/cree/cxa2520-0000-000n00q227f/led-hb-white-2379lm-screw/dp/2226422 it is 2700k and uses about 20w at 0.55A; this is the test current, max current is 1.25A (46w). is it safe running it on max or should i run lower?
if i would run on about .8 to .9 A i would use 3 of these led's for a total of ~100W and 4 x 55w pl-l's in flower. I am thinking to set it up with the led's in the middle and 2(one 3000k and one 6500k) pl-l's on each side.
I could also use different led's lets say 3000K or one 5000K in the middle so give me your ideas please
if i would run on about .8 to .9 A i would use 3 of these led's for a total of ~100W and 4 x 55w pl-l's in flower. I am thinking to set it up with the led's in the middle and 2(one 3000k and one 6500k) pl-l's on each side.
I could also use different led's lets say 3000K or one 5000K in the middle so give me your ideas please
No they arent dimmable unfortunately. I was looking at some open chassis MeanWell drivers that I think are around 1950mA max or something like that and can be made dimmable easily but they are $36 each plus shipping so I just went with the JHX for now to keep cost down and since the drive the COBs at the current I wanted to go with.
Having a dimmable driver would be really nice though. stardustsailor on RIU has a kickass DIY COB rig that uses those drivers and he set it up so his min. current is around 350mA up to the max the driver will run by turning a knob on the front of his case. He also has an arduino built in with thermocouples to measure, display, and monitor the temp of each COB and auto shutdown the unit if the COBs were to ever get too hot (fan failure, etc). He has an incredible amount of knowledge of electronics and took everything into account when building his lighting, many features that are awesome. My lighting is bare bones compared to that haha.
thank you bueno and cocogrower for your very useful info. i have been looking for a suitable cob, and i have found this one: http://si.farnell.com/cree/cxa2520-0000-000n00q227f/led-hb-white-2379lm-screw/dp/2226422 it is 2700k and uses about 20w at 0.55A; this is the test current, max current is 1.25A (46w). is it safe running it on max or should i run lower?
if i would run on about .8 to .9 A i would use 3 of these led's for a total of ~100W and 4 x 55w pl-l's in flower. I am thinking to set it up with the led's in the middle and 2(one 3000k and one 6500k) pl-l's on each side.
I could also use different led's lets say 3000K or one 5000K in the middle so give me your ideas please
Those 2520s wont be as efficient as the 3070s are but should work OK. I wouldnt run them too high of current because efficiency will drop the higher current you run them. They are rated just under 2400 lumens at 550mA so 3 of them at 550mA each would be roughly equivalent light output (just a little more output) to one of my 3070 AB bin COBs at ~1400mA.
Any particular reason you were considering those 2520s over 3050s or 3070s? I dont know what country you live in so dont know if its hard to find the 3070s or what but those 2520s are 30.15 euros each? That would be equivalent to $40.90 US, you can buy 3070 Z2 bin COBs for that price which are much more efficient giving you more light output for the same wattage as you would run the 2520s at.
OK nevermind they dont have 3070s on that site and want 51-60 euros each for 3050s thats crazy prices! I wonder if you can find the Bridgelux Veros cheaper I noticed some of the European guys say they Crees are more expensive there and harder to find and the Veros are easier to find and cheaper there. Might be worth looking into.
cocogrower
Member
It might have been I who suggested 2520's since he asked for smaller cobs. They are less expensive and pretty efficient if driven low. I'm almost about to buy 10 2530 4000 k 70 CRI. They give 155 l/w if driven at 800mA. My plan is to drive them @350mA, which should make them pretty damn efficient! Was planing to buy vero 29 3000k, but found such a sweet deal on the 2530's
Do you think 4000k would work for flower without supplementation?
Do you think 4000k would work for flower without supplementation?
cocogrower
Member
Yeah, stardustsailor often turn up when I google stuff associated to spectrum and grow lights. He seem like a smart and knowledgeable dude. Will have to find his project!
cocogrower
Member
I've seen 2520 for around 20$ and just found a ten pack of 2530's for 160$ From a Chinese seller. Let's hope it's the real deal!
cocogrower
Member
Hm. misread the binning. its a t4 and not u4. which leaves me with 131 l/w @800 ma.. it was a little bit to good to be true..
Buenotime where/how do you find your high binned chips? I'm looking at findchips.com ebay, chinese "ebays". Am i missing something? High binned chips are un existing or have to be bought in packs of 20-1000.. only looked a few days though, is it like fishing. It should be a verb for this. I've been "chiping" for several days now!
Buenotime where/how do you find your high binned chips? I'm looking at findchips.com ebay, chinese "ebays". Am i missing something? High binned chips are un existing or have to be bought in packs of 20-1000.. only looked a few days though, is it like fishing. It should be a verb for this. I've been "chiping" for several days now!
cocogrower
Member
Sorry for spamming your thread, i will try to let anyone else write something to!
Driving them at 350mA should be quite efficient. I dont know the numbers since I havent looked at data sheets or anything on them but running that low the efficiency should be very good. 4000k isnt bad I am using 3000K and would buy them again so I would recommend the 3000K spectrum at a slight loss in efficiency vs the 4000k or 5000k it produces a better stand alone spectrum for flowering IMO just from looking at some CREE CXA spectrum graphs. Thats not to say the 4000k wont work great I have just never tried it and prefer the 3000k spectrum from seeing graphs of the CXA spectrums and other forum members analysis of the spectrum of the 3000k CXA.
Do it, its a great build and you will learn a lot. He is big on math and analysis of all electronics involved and the science, a lot of the stuff he says makes no sense to me its too advanced for me goes right over my head but the stuff that I do understand is great information.
Nice!
131lm/w @ 800mA isnt bad at all and running lower current the efficiency will be even better.
I bought my first 3070 chip off digikey.com it was a Z2 bin, I would have had to buy minimun 20-100pcs to get the Z4 bin depending on which site I would have ordered from so I stuck with the Z2 bin and then they went out of stock on digikey for a few weeks I was looking for a second 3070 and found a few Z4 bin listed on ebay so I bought one of those. Most recently when I got the 3 AB bins I was very lucky to find and purchase them from an online forum friend at a good price so I picked them up. He was selling Z2s for $40, Z4 for $45, and AB for $50 each.
You either need to purchase a large quantity of chips to get the higher bins or find someone who did and is selling the extras. Unfortunately as you could understand I wont give out the information for the seller that sold me the AB bins for various reasons but if you look hard enough you might find the better bins. I wouldnt sweat searching for the holy grail top bins too much. As long as you get the best bin you can without having to purchase a large quantity I would go with that (unless you dont mind and have extra cash to buy a bunch to get a better bin and sell extras). I dont know what the best bins are for the 2520s but the AB bin for 3070 is the newest and most efficient to be release as far as I know. The thing about getting too obsessed with the best bins is that by the time you find some available there will already be better bins out or soon after they will come out but I know we all like to get the best we can at the time too. I didnt even know the AB bin 3070s existed yet until they kind of fell into my lap and I jumped on grabbing a few of them.
Its all good man. I am willing to try to help anyone if I can. I am not the most knowledgeable on everything LED but I do know a lot about the CXA3070s from doing tons or reading and research over the past few months.
I like the higher activity level in this thread compared to my old grow thread, so I dont mind answering questions or trying to give little bits of advice and knowledge here and there.
cocogrower
Member
glad to hear that! Looking forward to some more pics soon. Also wonder about how many lumens you have got per square foot, and if you have a rough approximation of the amount of red and blue spectrum? And do you feel that the red and blues made a big differens?
I really am not sure on a lumens/sq ft figure but I know the LED guys talk about umoles/sq meter/sec as the better figure of light photons being emitted per area of canopy per second. I havent spent enough time reading to full understand that whole concept. I do know my lumes per sq ft is low compared to what is typically recommended in the HID world.
I just went to a little larger tent so my per sq ft light is less now though I am not using the whole tent footprint either. For the rest of the run I am sticking with the 3070 3000k Z2 and Z4 bin COBs only and as long as I stop being lazy and get to building then I plan to have the red and blue LED bars and all 3 3070 3000k AB bin modules running for next run using the whole 2x3 ft footprint of the tent. Should be running about 185w of LED dissipation wattage and around 210w at the wall plug which includes the driver losses with 6 sq ft of canopy if filled to the max like I try to do. That is low on watts/sq ft but the efficiency is also great in light emitted per watt, better spectrum vs HPS, and not to mention not having reflector losses from trying to reflect half the light emitted from a HID bulb back towards the canopy (horizontal growing with HID bulbs). These factors all play into the equation but Im pretty confident I will get good results even with a relatively low light power on a per sq ft level.
Will have to wait and see though, some of the strains I am going to run next round arent known for their yields though.
Day 47 12/12
BBCJ BB Dom
BBCJ BB Dom
BBCJ CJ Dom
PECC
BBCJ
cocogrower
Member
Beautiful flower pics! hard to believe they have more than 20 days to go! How long will you let them flower?
Probably harvest around day 70ish but will have to wait and see how fast they finish.
