LEDs in GE grow light

bobski

Newly Enlightened
Joined
Mar 7, 2007
Messages
102
Location
Delaware
Can anyone ID these LEDs?
Mw2cDxp

Mw2cDxp.jpg

ETjUZgv.jpg

The center 4 are deep red (640-665nm using a cheapie visual spectrometer) and the rest warm white (450nm driver band). It's hard to tell through the magnifier, but I get the impression the LEDs' base plates are glass.
It's a GE PAR38 grow light, PN 93101232. Rated 32W at the socket, hefty cast aluminum body with fins, thermal shmoo between the body and LED board, large TIR optic.
 

chillinn

Flashlight Enthusiast
Joined
Jul 19, 2014
Messages
2,527
Location
Mobjack Bay
First of all, I cannot identify the LEDs, but wanted to ask how confident you were certain about the frequency of the red LEDs. 640-665nm would peak around 652.5nm, and the ideal red wavelength to encourage photosynthesis, stem, leaf, and general vegetative growth, is 660nm, which is an easy target for LED. There are at least a few manufacturers of 660nm deep or "photo red" (which gets its name for photosynthesis, not as many believe for a light that will not expose photosensitive paper in a dark room, which is incidental). However, those made by Cree are by far the most prevalent (Google "photo red LED" and nearly all the results will be Cree), and Cree makes (or made) at least three varieties, which may be generational, XLamp XP-E, XLamp XP-E2, and XLamp XP-G3. I am no expert! Just trying to help.
 
Last edited:

bobski

Newly Enlightened
Joined
Mar 7, 2007
Messages
102
Location
Delaware
wanted to ask how confident you were certain about the frequency of the red LEDs. 640-665nm would peak around 652.5nm, and the ideal red wavelength to encourage photosynthesis, stem, leaf, and general vegetative growth, is 660nm, which is an easy target for LED.
Not particularly confident. The spectrometer I used is a black plastic box with a hole and diffraction grating at one end, and a sample light slit and nm scale at the other. You look through the diffraction grating, point the slit at your light source and get a rainbow overlaid on the scale. The red emitters have a strong band in the range I listed, but I can't quantify intensity well enough to tell where the peak is within that range. That's why I'm curious what LEDs these are - I generally trust spec sheets more than my poorly calibrated instrumentation.
Yes, I'm aware why they use deep red LEDs in grow lights - the top image result for "chlorophyll absorption spectrum":
absorption-spectrum_med.jpeg

So chlorophyll a soaks up that deep red spectrum, but chlorophyll b likes the more orange-y red range. The color bar under the graph is a bit off - 650 is a saturated red, not orange.
 
Joined
Apr 13, 2020
Messages
336
Unless you are replacing the LEDs, who makes them is really not very critical.

They could be Cree, Lumileds, OSRAM, Samsung (LH351 series), Luminus (don't think so), lots of Asian suppliers, etc.


There are not many vendors of deep red die though. Lumileds, OSRAM, Epistar, can't remember who else. Cree does not make their red die, it is Epistar. Cree just packages those.
 

Dave_H

Flashlight Enthusiast
Joined
Nov 3, 2009
Messages
1,345
Location
Ottawa Ont. Canada
Out of curiosity I picked up a cheap LED plant-grow bulb, A19 8W 800 lumens, Noma brand. Spectrum
shown has a fairly small sharp blue peak (454nm) relative to larger/broader spectrum peaking in orange/red
region (612nm). CRI and CCT are not given, makes sense for this application.

The light is definitely yellower than the average 2700-3000K bulb, but not too orangey like 2200K.

There is a warning on the box about not being suitable for general illumination, and may require special
safeguards. As the tint seems nice for general use, I wonder what the risk might be, any ideas? Or is
it just some kind of CYA?


Dave
 

Dave_H

Flashlight Enthusiast
Joined
Nov 3, 2009
Messages
1,345
Location
Ottawa Ont. Canada
What strikes me about the colour of my low-cost bulb (plant growth aside, not
my primary usage) is the resemblance to light from a good old incandescent bulb;
white with slight orange tint.

Seems like I got a deal, regular price of other E26 A19 bulbs like this is typically
$15-$20 or more. Other than lower volumes and that these are "specialty" bulbs, not
sure why so expensive, just for a slightly different mix of LED colours and spectrum.

Meantime I picked up another low-cost A19 grow bulb, Globe which has somewhat different
spectrum shown on package, more skewed towards red, and higher blue spike. This one
emits pink light. Similar bulbs for the same purpose, but visibly much different.

Unlike the OP I don't plan to open either bulb at this point, would expect to
find a mix of whites and reds, though at 8-9W total consumption not too beefy.

Dave
 
Top