DiY LED - Cree CXA3070

kkman

Member
I was about to put the order in for cxa3070 z2 ( 4 step ) chips when I got an offer for cxa3590 bd ( 2 step ) bin running at 36V. Does anyone have any input on this. Its not the 77V version so it looks more like the cxa3070 but run at more amps.
The price is going to be more than double, is it worth it. PS. Was going to run the cxa3070 at 700ma, what would I run the cxa3590 at, any ideas. I have tried getting ab bin cxa3070 but need to place a large order to get my hands on them, but the cxa3590 bd bins I can order 2.
 

SupraSPL

Well-Known Member
kkman, depending on the asking price, it sounds like the CXA3590 36V BD would be a better option than the CXA3070 Z2. ( I assume they are both 3000K?)

If you ran the CXA3070 Z2 @ 700mA it is about 25W and 42.5% efficient ($3.65/PAR W). The CXA3590 BD at 700mA is 24W and 47.5% efficienct ($4.28/PAR W). The CXA3590 BD at 1.4A would be 51W and 42% efficient ($2.28/PAR W).

That is assuming at cost of $38 for the Z2 and $49 for the BD. If you don't mind my asking, what is the price for the CXA3590 BD and is there a lot available?
 

kkman

Member
Hi PurpleBuz, are you saying the CC version will not work. I have already received 5 units. Is there any way around using the meanwell apc-12-350. Thanks for your help.

Super, nice talking to the man who put me on the path to my own DIY, thanks.

As to your question, the item is CXA3590-0000-000N00BD30H, I have found it offered at £59.24 (UK) for batches of 20 on mouser. My contact works for a lighting company and had ordered 50 units for a project. I have been offered between 2 - 10 for £70 each. Sounds a lot but as I am not a big spender, I cannot pay for a batch of 20 or so which is the minimum order on mouser. Also it would mean waiting for up to 8 weeks, whereas my contact has them now and I can pick them up over the weekend. PS. Am I correct that the 30H ( 2-step ) is better for us than the 30F ( 4-step ).

PS. I have been told that they will be receiving a second shipment in the next month so I could add more to my order, but at that sort of price I cannot afford it.

Just out of interest where did you get your figure of $49 I cannot find any prices close to that even if I wanted to order 100. If I could I could save my contact a lot of money on his next order which may mean I get mine as a freebee. :-)
 
Last edited:

PurpleBuz

Well-Known Member
Hi PurpleBuz, are you saying the CC version will not work. I have already received 5 units. Is there any way around using the meanwell apc-12-350. Thanks for your help.
.
I don't know if it will work, as everybody on this thread has commented, constant current drivers aren't usually used for dc brushless pc fans. I suppose if the amp\volts requirements of the fan match up just right with the driver, it could work.
 

SupraSPL

Well-Known Member
There is a 50 pack of 3000K BDs selling on Arrow/Verical for about $47 each but they are the high voltage version. I was basing my price on that info. It sounds like your guy want to charge almost triple that which means you would be better off using Vero29 or a CXCA3070 Z2 bin and run it soft.

https://www.verical.com/partID/840773?&utm_source=octopart&utm_medium=buynow&utm_campaign=octopart#landingPage=catalogItemView&searchTerm=840773&_i_=1&searchCriterion=mpnIDs&searchName=

If we were to use the high voltage version we would need to use the Meanwell HLG drivers. On the other hand it seems like all the ABs that were for sale on Arrow are finally gone. I have never done business with Dloomis514 so I cannot vouch for them but maybe someone else in the DIY community can. There is nothing illegal about selling and buying COBs so as long as you use a safe addy and a safe payment method I suppose there is no harm. One possible problem though, how do you verify that it is an AB and not a Z2? Apparently the new ones all look the same. I have Z2s, Z4s and ABs on hand and I have compared them using a basic light meter. The difference in output is easily detectable so maybe I can test one of Loomis' ABs to verify.
 

SupraSPL

Well-Known Member
In a perfect world we would get 4 step that was shifted toward the red spectrum. But a 4 step could also be shifted toward the green which we probably do not want. A 2 step would have less potential to be shifted either way and tend to cost more than a 4 step. Based on all that, I prefer the 4 step but when it comes down to it they will both do an awesome job.
 

kkman

Member
I just got off the phone with the guy, and you are correct. It seems the 36V is more expensive but the guy said for their application, each light needs to be controlled by its own driver (300ma) and the cost of the drivers (in the UK) would make up for the cost of the cob compared to the 77V version.

I managed to get him let me have 8 of them, which I will pickup on Saturday. The deal is I test out the lights with a meter and provide them with a report on the actual lumens/wattage. If I do a good job he will let me have it in exchage for a share of my first harvest.

By the way he said I will be getting 6 3590 bd bin @ 3000 and 2 3590 bb bin @ 2700. Apart from the light meter is there any other tool I should get to test the spectrum the lights put out? Lastly at the high level of the 2700 should I just go for that or stick with the 3000 version.

Again sorry for so many question but its my first go at this sort of thing and I feel way out of my depth.
 

kkman

Member
Super, for my setup I think I will be using 2 - 4 for the flower room and 1 for the veg. If I get them for free and manage to build the lights without blowing them up I would like to offer you the remaining lights.

This will be my way of thanking you for the great work you and a few others on this forum have done to inspire people like myself and I am sure you can put it to better use than I will.

PS. This may be a little premature as I need to get the lights frst. :=)
 

SupraSPL

Well-Known Member
Wow CXA3590 BB 2700K @ 36V, that is awesome! I suspect the mixture of 2700 and 3000K you have there would be perfect unless you are growing super stretchy sativas. Awesome deal for a portion of the harvest that is one cool contact you have there :leaf:

As far as testing their output, unfortunately it is a very expensive proposition to get accurate reliable measurements that can be used for quantitative comparison. For output it would require a calibrated PAR meter or for color temp and LER a spectroradiometer. It may be possible to rent one. Luckily, Cree tends to understate the performance so you can rely on the Cree specs to estimate the output reasonably well. To make a long story short, the 3000K BD bin is the second most efficient warm white LED every sold to my knowledge so you can expect great things :) That is a very generous offer if you get your hands on them I would love to put one to the test! I would be able to test the vF, current and dissipation very accurately if that helps fulfill your testing goals. I could also test its output side by side with a CXA3070 AB 3000K for a good qualitative comparison.
 

SupraSPL

Well-Known Member
Nothing major to update. Just finished trimming a batch, looks like a similar yield to the last few and I will have the numbers in a bout a week.

Here is the next batch at 5.5 weeks. Some new flavors in there. This batch will bring us up to 150 ladies sampled. Working hard to find the best. The Kandy Kush is still in the top 5.

(4) Cali Connection - Alien OG fems from seed (Tahoe OG X Alien Kush Alien Tech Pheno reversed)
(1) Cali Connection - 22 fem (Jack Herer)
(1) G13 Labs - Blue OG

not new:
(1) G13 Labs - Sour Jack
(1) DNA - Kosher Tangie
(1) Sin City - Alien Nightmare
(3) HSO EMDOGs


DSC07622a.jpg
 

indianajones

Well-Known Member
does anyone remember where the spectral graph of the 3500k vero was located?
thought i remembered a graph comparing it to 4000k but i can't remember what
thread had the graph.

next few builds will be-

6x vero 18
meanwell HLG-185C-1050A

mounted to big ol' passive LED flood light heatsinks rated for 200w.

@SupraSPL do you think I should let the HLG run at 190v or should I turn it down
to ~180v? I know it's going to change depending on the Vf, but I'm ordering enough
veros to balance the high and low Vf units.
 

SupraSPL

Well-Known Member
The closer to 190V you run it the more efficient it will run. It makes no difference which order you wire them in so no worries there.

The Vero 18 gives a great spread with no lenses or reflectors required, but at 1050mA (30W) the 3000K is only 33.6% efficient and cost $2/PAR W. Very good number but depending on what bin of CXA you can find, you can get up to 42% effficient for just over $2/PAR W. On the other hand the Vero 3000K does seem to act more like a 2700K which may be good if you are growing mostly indica doms.
 

churchhaze

Well-Known Member
I'm not sure about the efficiency exactly, but if you do want to run 6 vero 18 at 1050mA, I think you picked the perfect supply. That will give you a max average Vf of 31.67V on each, which is way more than enough. That's close enough!
 

churchhaze

Well-Known Member
Are you suggesting that it would be better for people trying to save money on startup cost to get a 3000k cxa3070 z4 and run it at a higher current, like 2.1A? I'm just trying to have a clear idea of your position, because I obviously do see the cxa line as better. Is it really so much better that cxa even wins in startup cost?

The closer to 190V you run it the more efficient it will run. It makes no difference which order you wire them in so no worries there.

The Vero 18 gives a great spread with no lenses or reflectors required, but at 1050mA (30W) the 3000K is only 33.6% efficient and cost $2/PAR W. Very good number but depending on what bin of CXA you can find, you can get up to 42% effficient for just over $2/PAR W. On the other hand the Vero 3000K does seem to act more like a 2700K which may be good if you are growing mostly indica doms.
 

Chronikool

Well-Known Member
(1) Cali Connection - 22 fem (Jack Herer)
(]
Im interested in the Cali connection Jack Herer take on this as i just started a Sensi Jack herer seed....Got a close up of it? (just to be a wank... ;) )

The pic does a poor job of showing depth and so it distorts the distance but ya some of them can get pretty close sometimes as little as 6-8" but they are running at 800mA and I try to bend the colas out of the way if they get too close.
Not running as many COB's as you...but i can get my mine to about 10cm (4ish inches) to the top of the canopy
 
Top