Got my GE BrightStiks

HydroRed

Well-Known Member
So today I'm 2 led bulbs shy of replacing literally EVERY bulb in my house.
Is that close enough on topic? :-P
I have to add, I only paid $72 for everything since she didnt look and rang everything up as 40W. :hump:
20160529_174948.jpg 20160529_174934.jpg
 

HydroRed

Well-Known Member
Utilitech. They had 2 different versions of the same bulb. I bought 2 60W replacements months ago which were 9.5W 830mA 800Lu 3000K.
When I popped the diffuser off I could see it was very well made and the inside was metal and wasnt just a small thin coin shaped aluminum punch out as a heatsink. It was a metal cap pressed into a hollow metal housing then covered in like a gritty plastic that was much more rigid. The new bulbs they have on sale now that I bought are Utilitech 60W replacements that are 9W 1010mA 750Lu 3000K. The new ones have a different amount of leds in it (15 in the new vs 12 in the old) and they are mounted in a ring pattern on a cheap aluminum thin coin shaped board and they seem to get a bit hotter than the older ones that run at a higher wattage, but lower mA. I would like to find a batch of the old style 9.5W ones for my light strip I put together.
NEW ONES:
9w 750lu model.jpg

OLD ONES:
9.5W 800lu model.jpg
 

BOBBY_G

Well-Known Member
nice thats the first 9W bulb ive seen rated over 100 lm/w. are those the blue and white boxes from home depot?
 

HydroRed

Well-Known Member
You got pics of the full light bar yet with all the outlets?
Not yet. I got so wrapped up in replacing them in the house, I forgot my original project haha
nice thats the first 9W bulb ive seen rated over 100 lm/w. are those the blue and white boxes from home depot?
I think you glanced at the mA instead of the lu in your reference. Unfortunately, still under 100lm/W but a lot cooler in my small 4x4 momma/clone room.
The pic of the lights/boxes from Lowes are in post #221 which include the 40's & 60's, and the dimmable 40's & 60's in the dark boxes.
 

BOBBY_G

Well-Known Member
Heres some completely semi-scientific tests i ran today with a kill-a-watt and the li-cor @ 12"

hard to tell much with single bulbs as the kill-a-watt only measures to whole watt and the readings were easy to vary +/- 2" by positioning the sensor and or adjust the position of the lightbulb slightly. I was looking to see if there are any clear 'leaders' and 'dogs', really

feit 5K
model stamped on lamp:CEA1500/5K/LED/10
stated specs: 1500 lumens, 5000K
availability- $5 each in 10 packs at costco
actual draw-13W
109 umol w/no reflector - ~ 8.4 umol/W
152 umol with angelina reflector held in place on bottom

GE"60W replacement" brightstik (http://www.gelighting.com/LightingWeb/na/images/LEDL019-GE-LED-Bright-Stik-Spec-Sheet_tcm201-99939.pdf)
model stamped on lamp:LED 10LS/DL
stated specs: 760 lumens, 5000K, 10W, 150 mA
availability- $3.33 each in 3 packs at home depot
actual draw-10W
56 umol w/no reflector - ~5.6 umol/W

GE "100W replacement" brightstik (http://www.gelighting.com/LightingWeb/na/images/LEDL019-GE-LED-Bright-Stik-Spec-Sheet_tcm201-99939.pdf)
actual model LED16LS/SW (possibly obsolete? datasheet above shows LED16LS2
stated specs: 1520 lumens, 2850k, 16W 168 mA
availability - $7.50 ea in a 2pack: http://www.galesburgelectric.com/ge-led16ls2-828-led-16-watt-bright-stik-2pak-35523/
actual draw - 14W
126 umol w/no reflector*
175 umol with angelina reflector held in place on bottom*
*note these were at about 11" (brightstiks are taller than others) and i got similar results (~125 umol) when i measured the Feit 5K at 11". I'm calling it a wash on the brightness of the two, with the 5000K having a very slight edge on efficiency as expected due to color difference

sylvania/osram "100 W replacement soft white"
model stamped on lamp:LED14A19/f/827/10YV 2700K.
stated specs: 1500 lumens, 2700K, 14W, 195 mA
availability- $6.70 each in 4 packs at walmart (LED14A19/f/827/10YV/RP4)
actual draw-14W
102 umol w/no reflector - ~7.3 umol/W
**about the same with "can opener" reflector- it was useless**
147 umol w/ angelina reflector held in place on bottom

philips "100 W replacement soft white" (http://www.usa.philips.com/c-p/046677455675/led-bulb)
stated specs: 1500 lumens, 2700K, 14.5W, 234 mA?
availability- home depot, i remember paying $9 but they are now $15? (https://www.homedepot.ca/en/home/p.led-145w--100w-a-line-a19-soft-white-non-dimmable-2700k.1000830763.html)
actual draw-13W
100 umol w/no reflector - ~7.7 umol/W
142 umol w/ angelina reflector held in place on bottom
 

BOBBY_G

Well-Known Member
moral of story:
1.daylight/cool white is slightly more efficient than warm white (duh)
2. brightness/efficiency was as expected per product specs - quick math on stated lumens per watt on the box seems valid even if you rip off the reflectors
3. "can opener" reflectors were useless though the angelinas worked pretty great
4. kill-a-watt on a single bulb <20W isnt all that accurate and or useful

future work:
-i hacked apart one of my 100W brightsticks and skewered the board upside down on another brightstick and lit em up. both lit up from one driver pushing both boards in parallel - pulled same 14 W. presumably this would cut the current in half to each board and could be a 10-20% efficiency bump? im gonna sacrifice a bunch of the 5K feit boards , mount em on a real heatsink and test them with a single factory cheapie driver pushing 1,2,3, and 4 boards each. also see the difference in cheapie driver vs a 90+% efficient meanwell. i expect better luminosity by more efficient cooling alone, yet alone reduced current

The acid test will be 4 of the 15W bulbs ($20 including driver) vs a 3590 at 60W ($50-55 including driver). waiting on my bjb holders for the citis and gonna do the cob tests all at once later in the week

time to buy a real benchtop power supply....
 
Last edited:

BOBBY_G

Well-Known Member
They get hot as fuck. It really sucks. But I doubt any other bulbs get that much less heat coming off them
these are about all the same from what ive seen, only advantage so far is the feit seems to have a metal housing as opposed to plastic
 

OneHitDone

Well-Known Member
moral of story:
1.daylight/cool white is slightly more efficient than warm white (duh)
2. brightness/efficiency was as expected per product specs - quick math on stated lumens per watt on the box seems valid even if you rip off the reflectors
3. "can opener" reflectors were useless though the angelinas worked pretty great
4. kill-a-watt on a single bulb <20W isnt all that accurate and or useful

future work:
-i hacked apart one of my 100W brightsticks and skewered the board upside down on another brightstick and lit em up. both lit up from one driver pushing both boards in parallel - pulled same 14 W. presumably this would cut the current in half to each board and could be a 10-20% efficiency bump? im gonna sacrifice a bunch of the 5K feit boards , mount em on a real heatsink and test them with a single factory cheapie driver pushing 1,2,3, and 4 boards each. also see the difference in cheapie driver vs a 90+% efficient meanwell. i expect better luminosity by more efficient cooling alone, yet alone reduced current

The acid test will be 4 of the 15W bulbs ($20 including driver) vs a 3590 at 60W ($50-55 including driver). waiting on my bjb holders for the citis and gonna do the cob tests all at once later in the week

time to buy a real benchtop power supply....
Why are you messing with this junk rather than starting from raw components?
 

BOBBY_G

Well-Known Member
Why are you messing with this junk rather than starting from raw components?
science brah. i love off the shelf stuff for the growers still using CFLs. also these make awesome cheap supplemental lighting when you gotta light up a clone tray or two or want to dick with your spectrum.

i got cobs out the ass over here. 3000/3500/4000/5000/6500k 3590s, 2700/5000 70 cri/5000 80 cri citizens, 135W and 275W cutter, shitload of photo red and far red monos, old ass ufos that need scavenged, etc etc..

and still have more gear/rooms than i can fill right now. idle shit all over the place

really my only three goals for the week are:
1. real world test of the citi 1825 series next to a 3590 to get a benchmark of its efficiency and value vs a well-understood control
2. real world test of the cutter cobs next to a 3590 to get a benchmark of their efficiency and value vs a well-understood control
3. mess with these $0.30/W off-the shelf A19s so we can get a rough idea where they fit in the grand scheme of things. are they 110 lm/W off the shelf? 120-130 lm/W when you rip the diffusers off? etc.

takin it for the team, trying to fill in the knowledge by doing things nobody else is. do you really want to see another boring 3590 rig built?

id have more cob rigs if id stop fucking breaking 4-40 taps grrrrr.
 
Last edited:

BOBBY_G

Well-Known Member

Airwalker16

Well-Known Member
science brah. i love off the shelf stuff for the growers still using CFLs. also these make awesome cheap supplemental lighting when you gotta light up a clone tray or two or want to dick with your spectrum.

i got cobs out the ass over here. 3000/3500/4000/5000/6500k 3590s, 2700/5000 70 cri/5000 80 cri citizens, 135W and 275W cutter, shitload of photo red and far red monos, old ass ufos that need scavenged, etc etc..

and still have more gear/rooms than i can fill right now. idle shit all over the place

really my only three goals for the week are:
1. real world test of the citi 1825 series next to a 3590 to get a benchmark of its efficiency and value vs a well-understood control
2. real world test of the cutter cobs next to a 3590 to get a benchmark of their efficiency and value vs a well-understood control
3. mess with these $0.30/W off-the shelf A19s so we can get a rough idea where they fit in the grand scheme of things. are they 110 lm/W off the shelf? 120-130 lm/W when you rip the diffusers off? etc.

takin it for the team, trying to fill in the knowledge by doing things nobody else is. do you really want to see another boring 3590 rig built?

id have more cob rigs if id stop fucking breaking 4-40 taps grrrrr.
I might buy those Utilitechs ($2.98/2) to take apart, clip the wires off the metal mogul threads, mount them on a heatsink and wire them all up in series to a power cord to rebuild my hoods. That's cheaper than buying another driver and some say. ...2540's or so. Just use a new white tub as the reflector, cut the bottom out and screw it to the heatsink. I imagine these leds just need direct ac power unlike chips that need a driver, right?
Or are they all equipped with some kind of mini driver under the metal "heatsink" you always see them mounted on?
 

BOBBY_G

Well-Known Member
I might buy those Utilitechs ($2.98/2) to take apart, clip the wires off the metal mogul threads, mount them on a heatsink and wire them all up in series to a power cord to rebuild my hoods. That's cheaper than buying another driver and one sat 2540's or so. Just use a new white tub as the reflector, cut the bottom out and screw it to the heatsink. I imagine these leds just need direct ac power unlike chips that need a driver, right?
a lot of the boards say 94V, pretty sure all diodes run on DC but i could be wrong. i think the fact that a lot of the boards are designed to run on 94V is its easy to get there from the usual 115-120V AC

lemme blow some up on the bench and ill let you know lol

heres a brightstik driver... lots going on in there... looks like a simple AC>semi-regulated DC power circuit. i cant diss it those are some fat ass caps. im sure its pushin 80%+ eff.



 
Last edited:

BOBBY_G

Well-Known Member
ill bet another simple measure of these different bulbs is just by count of/ size of SMDs

the more efficient 14-15 watters all have 24-32 chips. the inefficient 10W one i have has 7.....obv the SMDs are gonna run more efficiently at 1/2W each vs 1.5W each.
 

Airwalker16

Well-Known Member
a lot of the boards say 94V, pretty sure all diodes run on DC but i could be wrong. i think the fact that a lot of the boards are designed to run on 94V is its easy to get there from the usual 115-120V AC

lemme blow some up on the bench and ill let you know lol

heres a brightstik driver... lots going on in there... looks like a simple AC>semi-regulated DC power circuit. i cant diss it those are some fat ass caps. im sure its pushin 80%+ eff.



Damn looks like I can't do it then.haha I will be building some nrlew and improved hoods like you've seen of mine before, but with a heatsink on top and 2540's rather than a baking pan and moguls with Y splitters.
 
Top