I recently purchased (2) 300 watt LED grow lights from a retailer in China.


These lights have 100/3w LED chips divided amongst 4 circuits at 25 chips per circuit (4 separate drivers contained within). The drivers are labeled as 55-100 volts DC output @ 500-550 mA.

As shipped the color spectrum breakdown is as follows, 72 red, 16 blue, 6 white and 6 infrared, each and every LED chip has its own resistor.

The lights were designed for Marijuana growers, hence the over abundance of red LED chips, apparently red is the desired spectrum for flowering Marijuana.

I am not a Marijuana grower and the lights are somewhat ineffective for starting vegetable sprouts, thus as they currently are structured they are of little use to me. The project here is to modify these fixtures to a more vegetable friendly white/blue spectrum, at issue is the difference between the voltage/current requirements of the red and white LED chips.

I attempted to change out several reds with whites and found that the threshold for replacement was 3 swaps, anymore and only the reds in the circuit light bright enough to be useful, the whites appear as unlit or lit dimly. When only 3 reds are replaced with whites the circuit lights all chips brightly.

The following are the detail characteristics of the different chips.

Red, 2.3-3.0 driven volts @ 550 mA
White, 3.2-3.8 driven volts @ 700 mA

The resistors are micro type surface mount and it is difficult to identify the properties of each.

So, I am now stuck with only being able to replace 12 total out of the 72 reds? Should I call it a day and stop where I am at, or is there a cost effective solution to get more red out and get more white in?

I am at an impasse.

 

First of all take the specs with a grain (or a ton) of salt: 300 W output and only takes 190 W from the power source while wasting some power in the resistors. Free energy!

Resistors are easy to measure and easy to replace but unless you can find or make a schematics it is hard to give exact instructions. Most likely the drivers are voltage sources and resistors are limiting the current. Not the best design. LEDs require current regulation. Ideally one would connect all the white LEDs in series without resistors and add a current regulator. For example if you have a 50 V DC source you can have a string of 12 white LEDs (assuming 3.8V drop each) and a current regulator for 700 mA.

 

You need royal blue LEDs for startig up seedlings. Regular blue one's also work it is not as critical as with the red one's. But lower efficiency.

You can use white, but seedlings will have abnormal growth.

 

Resistors are easy to measure and easy to replace but unless you can find or mtake a schematics it is hard to give exact instructions.

It appears to be two lines run parallel with twenty-five sub circuits, the subs consist of a resistor and diode. Eighteen of the subs are red diodes, four are blues, the last three are a mix of white and infrared, the resistors appear to be static, only through isolating and measuring resistance could their true values be found.

You can use white, but seedlings will have abnormal growth.

Abnormal in what way?

 

Chlorophyll as two absorption peaks. One red and one blue (royal blue or violet). If you are growing seeds, the color of the light doesn't matter - the seed is looking for heat not light) until it is above ground and appears green when exposed to white light.

The green color is what is left from white light once blue and red ate absorbed by the leaf. If lit with only red and blue light, the leaf will appear eerily dark - even black - if no ambient light is present.

 

The LEDs in series with the resistors are pretty much an open circuit below the threshold voltage - at least 2.x volts. So a 1.5 V battery in series with an ammeter can be used to measure the resistors like they were not in the circuit.
The described topology suggests that the rail have constant voltage, say 4 volts and individual resistors are used to limit the current. From the described problem it appears that the "constant voltage" is not very constant at all: using more white LEDs taking more current than red ones causes the rails voltage to drop. Easy to measure.

Probably the simple thing to try is to measure the rails voltage with stock LEDs and find a better power supply. For example if it is anywhere close to 5 volts, an old PC power supply could probably be used. Now the resistors must be different for red vs. while LEDs. Suppose we have 5 V. If a red LED drops 3 volts, 2 volts must be dropped by the resistor. For 550 mA that will be 3.6 ohms and a full 1.1 watt will be heating that resistor. If a white LED drops 3.8 V, it leaves 1.2 V and at 700 mA that will need a 1.7 ohm resistor to waste 0.84 watts.

 

It appears to be two lines run parallel with twenty-five sub circuits, the subs consist of a resistor and diode. Eighteen of the subs are red diodes, four are blues, the last three are a mix of white and infrared, the resistors appear to be static, only through isolating and measuring resistance could their true values be found.

Abnormal in what way?

they will be stretching too much + weak roots.

i did some experiments with leds to grow plants.

as said with white they stretch, but that is not all. after a while, leaves develope abnormally, verry thin. to some degree this was also caused by wrong type of soil, peat instead of clay.

i never used resistors. I made panels with LLEDs in series, and then used these cheap buck converters from china. They have a 10 turns adjuster.

With an Amp meter, I did set them to a voltage whhere the right current would flow (waiting until they turn hot).

With good cooling, it is not so important to stay within the limit.

Cree LEDs however are more sensible, I had one chip burning out.

I am not growing anymore now, spent a lot of money + efforts but then dismantled everything.

One time I ha a grow box with white LED and radish plants. They stretched a lot + the red radish roots were pretty much underdeveloped.

 

Just put the fixture back to how it was. It will grow any plant just fine.
"pot" isn't some magical plant that grows on some remote planet with only red light..
it will grow right next to a tomato plant or whatever..

As stated above the best light for any plant is in the blue/red spectrums. Higher red counts are better for the flowering stage though but will work just fine for the vegetative state. Plants grown under red/blue do look almost black as there is very little "green" in the light and color comes from light being reflected or not absorbed.

If you are having problems starting seedlings I'd bet its caused by something else.


And if you wanted to convince someone you weren't growing the pot I would have picked a different screen name

 

Just put the fixture back to how it was. It will grow any plant just fine.

+1
At most I would supplement the light from this panel by adding another panel with the LEDs you think you need, not by modding the pre-built device.

“When you can measure what you are speaking about, and express it in numbers, you know something about it, when you cannot express it in numbers, your knowledge is of a meager and unsatisfactory kind; it may be the beginning of knowledge, but you have scarely, in your thoughts advanced to the stage of science.” - Lord Kelvin

My point is, you're attempting to alter the spectrum of your light source but I'll bet you cannot show the current spectrum versus the desired spectrum, and show that there is a gap to fill.

 

Chlorophyll as two absorption peaks. One red and one blue (royal blue or violet). If you are growing seeds, the color of the light doesn't matter - the seed is looking for heat not light) until it is above ground and appears green when exposed to white light.

The green color is what is left from white light once blue and red ate absorbed by the leaf. If lit with only red and blue light, the leaf will appear eerily dark - even black - if no ambient light is present.

This makes sense, I thank you for your easy to understand explanation.

Just put the fixture back to how it was. It will grow any plant just fine.
"pot" isn't some magical plant that grows on some remote planet with only red light..
it will grow right next to a tomato plant or whatever..

As stated above the best light for any plant is in the blue/red spectrums. Higher red counts are better for the flowering stage though but will work just fine for the vegetative state. Plants grown under red/blue do look almost black as there is very little "green" in the light and color comes from light being reflected or not absorbed.

Again, this too makes good sense, and I thank you. I will take your advice and reconfigure the light back to its stock configuration.

And if you wanted to convince someone you weren't growing the pot I would have picked a different screen name

Actually, the screen name represents my knowledge of electronics, I try to understand how it all works but it is very complicated to me.

+1
At most I would supplement the light from this panel by adding another panel with the LEDs you think you need, not by modding the pre-built device. My point is, you're attempting to alter the spectrum of your light source but I'll bet you cannot show the current spectrum versus the desired spectrum, and show that there is a gap to fill.

Thank you for your input, I understand your point, if it ain't broke don't try to fix it.

After reading your replies I did further research, according to others white light from LED chips is wasteful when used for gardening because white light is composed of red/blue/green and plants only use the red/blue while reflecting the green, many consider this wasted energy, these findings mimic what you folks stated in your replies.

Thank you all for helping me to better understand.

 

Be very careful, Chinese products do not generally meet UL specs. Likely as not these LEDs are powered directly from the main power line, which translates into a major shock hazard if you are modifying it.

When people are designing LED circuit we do not allow LEDs or power supplies connected to mains without a transformer. It is directly against our ToS (Terms of Service).

 

Be very careful, Chinese products do not generally meet UL specs. Likely as not these LEDs are powered directly from the main power line, which translates into a major shock hazard if you are modifying it.

.

Except him stating power supply states 50V to 100V DC 500-550mA output would lead me to believe its using a "constant current" driver.

 

Except him stating power supply states 50V to 100V DC 500-550mA output would lead me to believe its using a "constant current" driver.

The fixture incorporates (4) constant current drivers, albeit they are cheaply made, consisting of a circuit board and several components covered by a removable plastic bonnet. The drivers do not have a UL rating stamp or sticker on them. All components are grounded to the case, as is the power cord.

I did other research and found that distance from fixture to plant canopy is critical to avoid light bleaching, in effect to much light, something I had always thought to be impossible. It seems that there are many different opinions on the subject of just how close to the canopy the fixture should be, dependent which type of chip is used, and what the particular viewing angle is used and what type of plant is grown, and what medium is used for the plant, and what type of feeding regiment is used....

It is all very confusing, but the consensus is that the fixture should be somewhere between 12 and 18 inches from the top of the canopy for a starting reference point, tweaking is required in each individual instance to obtain the correct balance for optimal growing.

Bottom line, it's trial and error.