spectral analysis of Solarstorm and KIND K5

BOBBY_G

Well-Known Member
I took some spectrums of the 2014 Solarstorm 880 (5W LEDs) and 2015 Solarstorm 880 (3W Osram, and a much whiter spectrum than the 2014s), as well as Kind K5 XL750

The spectrometer im usng is an Ocean Optics USB2000, the intensity of the lights saturates it immediately, so these were taken off to the side with no other fixtures running

please use them to compare relative spectra only, as intensity is variable due to the placement issues described above. I tried to roughly match the peak heights for comparison purposes.

Notable:
Measurements of veg and bloom on each SS880 were taken from the same location for each picture. so the ONLY quantitative measurements you can take from here is 2014 veg vs 2014 bloom and 2015 veg vs 2015 bloom. there is no correlation of intensity between fixtures

as you can see, for a given fixture, that worked pretty well. The blue spectra on both of the 880 fixtures aligned right on top of each other for both veg and bloom settings.

only other notable is the 2014 SS880 is missing some green diodes so there might be a little more there in the 500-550 range, although id expect it to be low like the other fixtures
 
Last edited:

az2000

Well-Known Member
Much respect for owning a spectrometer! It would be interesting to see those spectra compared to Area51 or one of the COBs.

You know what I'd like to see.... the 18w Cree PAR38 spot/flood from Home Depot! And maybe the cool & warm version of the 9.5w A19 (not the plastic "4Flow," but the one fins around the base.). If you ever had the inclination to do that, that would be fantastic.
 

az2000

Well-Known Member
not sure why the par38s are so popular. not all that efficient and kind of expensive(1500 lumens for $25) .
For me, it's fewer to hang (at 18w) and they come with a reflector built in. And, 3000k. I did pretty good flowering a large/tall plant under them.

i buy these ones:
I looked for those the other day but, couldn't find them at my local Home Depot. I might buy some mailorder. I'd like to try some. I thought they could be useful for a new grower who doesn't want to make a longer-term investment. And, curious how their efficiency might rise with the plastic globe removed (although, I wouldn't recommend the electrocution hazard to a new grower.).

Offhand, if I need to string more bulbs like that, the 90-95'ish L/w of Cree A19 9.5w seems like a better investment (with 10-year warranty). And then, as I said, for more area, the higher-wattage PAR38s make the job easier.

I tried BR30s which are low L/w like those GE stick's and wasn't pleased. They seemed to have limited usefulness (only where space is tight). Those GE sticks seem like a contender for that role (although external reflectors add to the space requirement).

It would still be interesting to see the spectral diagram of these lights!
 

littlejacob

Well-Known Member
Bonjour
It is only 76 lm/w...even if it is cheap it is 2/2.5 time less lumen than most of diy or commercial cxa/b vero panels!
So in fact you will use at least 2 time more electricity to get the same results!
It is cheaper now...but 10 years of double electricity bill...I pay 10cts 1kw/h...I have 1000w so 10×12=1.20 € per day...36 € month...472 € year...
It is only vegg light ok! But I am sure I can do the same vegg with half of your w...
CU
 

az2000

Well-Known Member
Bonjour
It is only 76 lm/w...even if it is cheap it is 2/2.5 time less lumen than most of diy or commercial cxa/b vero panels!
I'm curious how those L/w improve with the frosted cover removed. Just scraping the rubber goo off the Cree A19 bulb improves performance about 10% (measured with a lumen meter).

For trailer-park growers like me, if those GE sticks rose to 90w/L, that could be useful (except for the shock hazard). Otherwise, if 76L/w is as good as that gets, I would look at the SMD5630 "corncob" lights on AliExpress.

I hate to recommend those because it's like buying from a flea market. They aren't accurately depicted, they draw only 1/2 to 1/4 the wattage they claim (i.e., you have to string many more than you anticipated in order to get the w/sq ft you planned for.). But, you're talking possibly 110 L/w.

IMO, the GE Stick would be best suited for a very small, enclosed space. And, that's what those SMD5630 bulbs will work best for too. But, they're much more efficient. And, about the same price ($3 each). I would go that direction (notwithstandng the distastefulness of dealing with the AliExpress crooks who sell them.).

But, I can see the GE stick being useful, especially to a new grower who might prefer picking p some things at Home Depot (instead of mail-order drama). When you're new, you don't care a lot about the details, you just want to see if you can grow. 76 L/w for $3.50 isn't bad.
 

Abiqua

Well-Known Member
Great Job OP
Yuck for the both of those lights though, where is the missing spectrum? ;)


Also the cree bulbs are known commodities with labeled bins...What are the GE again? Give me Radiometric not Photometric data comparisons :)

I think there is a reason why the GE's are probably cheaper.....
 

az2000

Well-Known Member
Also the cree bulbs are known commodities with labeled bins...What are the GE again?
When I looked at these a couple weeks ago I found this teardown on candlepower. I didn't see anything about which diode it uses. Maybe some of you guys recognize what it might be from the photos.

I like how they all face forward. There could be an implicit performance improvement from that (with the plastic globe removed), compared to the omni A19.
 

Alexander Supertramp

Well-Known Member
not sure why the par38s are so popular. not all that efficient and kind of expensive(1500 lumens for $25) . i buy these ones:

http://www.homedepot.com/p/GE-60W-Equivalent-Daylight-General-Purpose-LED-Bright-Stik-light-bulb-3-Pack-LED10S3-5K-96/205783755

at $760 lumens x3 = 2280 lumens for $11

leaves you money left over for

http://www.homedepot.com/p/HDX-75-Watt-Incandescent-Clamp-Light-HD-200PDQ/205139241
Its always great to save money. But unfortunately lumens do not add up with the addition of more lights. 1 760 lumen bulb or 10. Its still only 760 lumens.
 

Abiqua

Well-Known Member
When I looked at these a couple weeks ago I found this teardown on candlepower. I didn't see anything about which diode it uses. Maybe some of you guys recognize what it might be from the photos.

I like how they all face forward. There could be an implicit performance improvement from that (with the plastic globe removed), compared to the omni A19.
One of my favorites was Phillips bulbs before they went to their comical looking shape about a year ago? They utilized the same design and though Mr Flux analyzed the radiometric data for me and they were a tad less than Cree diodes, they still use Luxeon's :)
As for GE diodes, that is anybody's guess.....these are 1st gen cob's really, SMD's from strip lighting application, very small, but similarly constructed diodes, just way less material obviously....but the generic Home Depot and Walmart versions of led bulbs, both have these same "looking" diodes...

but still all interesting nonetheless....Not much progress in 3 years of bulb making eh? I have a post somewhere around here about the Walmart bulbs in 2013 with same design, pcb and everything...Makes you wonder a bit :peace:
 

Abiqua

Well-Known Member
and....on top of that last thing I noticed is that the GE's are using only 5 diodes, so a great chance of even lower efficiency and increased heat, since they probably have to be driven hard to "put out" 760 lumens....since we have no idea what efficiency is with no LER.
 

az2000

Well-Known Member
they probably have to be driven hard to "put out" 760 lumens....
That's an interesting thought.

With a lossy (plastic) globe, and no metal heat-sink'y material, it seems impressive that they can get 760 lumens from 10w. The 3-year warranty implies it's not driven too hard (it won't fail too quickly).

I'm thinking of that in context of the Cree BR30 which does only 650 lumens from 9.5w (and a metal heatsink). I think the BR30's loss is due to being essentially an A19 onmi with a reflector. Whereas, the GE Stik probably gets an advantage from being forward facing.

(Not necessarily an advantage for home lighting where the Stik converts the forward facing to diodes to omni. But, an advantage if you wanted to use a light directionally, compared to the BR30.).

I don't like the BR30. The Stik may not be any better. I'm just thinking out loud what the advantage might be, if any. If the Stik were a decent spectrum, I'd probably try it with the globe cut off.

I wish there were more specs about things like this.

I should probably spend this energy playing with more-efficient SMD5630 instead of discovering the merits between BR30 and Stik. :) (Seems pointless.). It's just that some people might be willing to try growing but don't want to go deep into lighting choices. If something has a degree of merit for those people (something they can pick up at the local store), that seems exciting to me. I mean, if it's better than CFL, it seems like it would have merit.
 
Last edited:

littlejacob

Well-Known Member
Bonjour
And why not V10 at 4 $ with a cheap driver to reach 120/140 lm/w depending on the driver? (250mA will be cool!) On cheap 50mm cpu cooler (3 $ on Aliexpress! )
CU
 

az2000

Well-Known Member
Bonjour
And why not V10 at 4 $ with a cheap driver to reach 120/140 lm/w depending on the driver? (250mA will be cool!) On cheap 50mm cpu cooler (3 $ on Aliexpress! )
CU
It's too bad someone doesn't make some bulb-alternatives like that. When I mentioned SMD above, I was thinking maybe I should make/sell some kind of more-efficient alternative to the retail options. I don't think there would be much demand for it. But, I'm not looking to start a full-time business.

I remember A51 was going to do XMLs(?) and expected it to have removable lenses and a dimmer. I think Jeff was considering this market.

I remember reading a post of his where he considered making replacement parts for the epi-whatever fixtures. I specifically remember he mentioned Apollo disks (the 15x3w circles), something which could be snapped in like that.

I think there could be a market for things like that. Oddball stuff I guess.
 

littlejacob

Well-Known Member
Its always great to save money. But unfortunately lumens do not add up with the addition of more lights. 1 760 lumen bulb or 10. Its still only 760 lumens.
Bonjour
I agree but with 10 light source you have the same intensity everywhere...that's why I think the new big cob (250/500w) are not good for growbox...maybe for big commercial grow without reflective walls!
CU
 

az2000

Well-Known Member
and....on top of that last thing I noticed is that the GE's are using only 5 diodes, so a great chance of even lower efficiency and increased heat, since they probably have to be driven hard to "put out" 760 lumens....since we have no idea what efficiency is with no LER.
I just saw this news: GE announces brighter 16w (100w equiv) Bright Stik. It's supposed to produce 1520 lumens. That's 95 L/w (with the plastic diffusion globe attached.).

That's better than the Cree A19 I use (about 90 L/w with the protective rubber coating removed).

If the diodes face forward like the 60w-equiv version, that could be a decent light compared to putting a Cree in a reflector to make it directional. Cut off the plastic diffusion cap. It pose somewhat less risk of electrocution compared to removing the globe on a Cree bulb (the protruding tower which invites contact. I never remove my globes.).

I wonder if the spectrum is decent (for a CRI 80, 2850k).

(I know DIY low-power COBs would be much more efficient. Like I said in a previous post, I'm thinking more about alternatives to CFL. Baby steps. Something the average grower could use off the shelf without getting immersed in digikey, etc. If this GE light faces forward, I think that would be considerably more efficient than the omni-Cree in a reflector. And since it would be slightly more realistic to remove the globe, another win. They're supposed to be $8 each. Not bad for a 100w-equiv LED bulb.).
 

az2000

Well-Known Member
I found a 3-pack of the 10w (60w equiv) GE Bright Stiks. I popped the plastic diffusor (warning: electrocution hazard). In a reflector it looks brighter than the Cree.

Using a lux-meter app on my smartphone I measured a Cree 9.5w 2800K (160 lux). And this Bright Stik 10w 2850k (215 lux). Nothing changed between those tests except replacing the bulb.

I know I should quit playing in the gutter and do COB. But, these look good to me. I'm going to use them at some point.

Popping the diffusor is easy. It's held in by three tabs, and partially plasti-welded to the base. Grasp the base and diffusor and bend like you're trying to break it in half. Move around the perimeter doing that. After 3-4 tries it will snap. (Store in an 8oz party cup, upside down, to protect the phosphors.).

I think this is the next big thing for me. :)
 
Last edited:

nogod_

Well-Known Member
There is a company called wicked grow lights doing that pretty much. but with "meh" parts and the website looks like Maddox made it in 1996. But the idea is pretty genius honestly.
Being able to change your spectrum by unscrewing a bulb is awesome.
And with a universal socket, upgrading bulbs would be a cinch from generation to generation.

It's too bad someone doesn't make some bulb-alternatives like that. When I mentioned SMD above, I was thinking maybe I should make/sell some kind of more-efficient alternative to the retail options. I don't think there would be much demand for it. But, I'm not looking to start a full-time business.

I remember A51 was going to do XMLs(?) and expected it to have removable lenses and a dimmer. I think Jeff was considering this market.

I remember reading a post of his where he considered making replacement parts for the epi-whatever fixtures. I specifically remember he mentioned Apollo disks (the 15x3w circles), something which could be snapped in like that.

I think there could be a market for things like that. Oddball stuff I guess.
 
Top