LED driver selection for grow light(s)

[bsaylors]

Would like validation of my understanding of driver need and sizing as this is my first time trying to build a DIY LED light array. I’m trying to setup an adjustable light for growing indoor vegetables. Best guess or starting point of four lights in series. Then tweek based on PAR/DLI values.

Below are numbers I’ve pulled from the datasheets.

Light(s):

Samsung SI-B8T502560WW, DigiKey part number 1510-SI-B8T502560WW-ND

Operating Current mA: min -, typ 1200, max 1600

Operating Voltage Vdc: min – 39.5, typ 42.0, max 45.0

Power Consumption W: min 47.4, typ 50.4, max 54

Possible drivers:

Meanwell pops up frequently when searching, so I’ve used theirs for this example. Other driver options are welcome.

Meanwell HLG-240H-C1400AB, DigiKey part number 1866-HLG-240H-C1400AB-ND

Rated Power: 250.6W

Constant Current Region: 89 – 179V

Current Adj. Range: 700 – 1400mA

Meanwell HLG-320H-C1400B, DigiKey part number 1866-2500-ND

Rated Power: 250.6W

Constant Current Region: 89 – 179V

Current Adj. Range: 700 – 1400mA

Meanwell XLG-320-L-AB , DigiKey part number 1866-XLG-320-L-AB-ND

Rated Power: 315W

Constant Current Region: 150 – 300V

Current Adj. Range: 500 – 1400mA

From searches it looks wiring in series is the way go. Would like to have dimmer option.

If you could verify or clarify my understanding.

Rated Power / Constant Current Region: Each light will consume between 39.5 – 45 Vdc. So looking at the first driver, with the Constant Current Region (89-179V) it would ideal for 2 – 3 lights. Adding a 4th light could be just outside the upper range (4 x 45V=180V) and having only one light would be below the region?

Is it over simplistic to assume if the driver has a rating of 250W (1400mA x 179V = 250.6W) and each LED strip requires 50W. A multimeter would see a decrease in voltage leaving each LED strip?

Current Adj Range: Assuming going over the Max operating current would damage the strips and the Current ADJ. Range of all the above drivers max out between the typical and max current of the LED strip would make them all candidates for the selected light?

Current Adj Range: The first two show a 700 – 1400mA and the third shows 500 – 1400mA range. What is the significance of the first number being 700 on the HCG and 500 on the XLG?

DC 0/1-10V dimmer: Per the Meanwell datasheet “Output constant current level can be adjusted by applying one of the three methodologies between DIM+ and DIM-:” Back to the previous question, would this dim the lights all the way to zero or just to the bottom of the Current Adj Range? Looking through the TechForum I think the 500-1400 and 700-1400 are the sweet spots of the driver.

As a follow on, IF the lights are dimmed and only running at ~60% to reach the max PAR/DLI needed , probably not best practice, is it safe to assume additional strips could be added to the same driver? Which circles back to the multimeter question of being able to see input / output voltage at each strip and being able to determine how much voltage is being consumed by each strip. Would this be a bad idea or a common practice?

TIA

[rick_1976]

Greetings,

@bsaylors

Rated Power / Constant Current Region: Each light will consume between 39.5 – 45 Vdc. So looking at the first driver, with the Constant Current Region (89-179V) it would ideal for 2 – 3 lights. Adding a 4th light could be just outside the upper range (4 x 45V=180V) and having only one light would be below the region?

A few things to note:

• A device may exhibit voltage characteristics anywhere within the stated range: for driver selection purposes, one should ideally consider worst-case scenarios at either end of the range.
• The stated voltage characteristics apply only at the indicated current and temperature test conditions. Increasing drive current will push the voltage numbers up somewhat, increasing temperature will push them down.

• Given these considerations, 3 is the only “good” number of the mentioned LED arrays to use with the driver.

@bsaylors

Is it over simplistic to assume if the driver has a rating of 250W (1400mA x 179V = 250.6W) and each LED strip requires 50W. A multimeter would see a decrease in voltage leaving each LED strip?

The driver power rating is perhaps a distraction to the voltage question, but yes, a voltage drop will appear across every element in a series-connected array.

@bsaylors

Current Adj Range: Assuming going over the Max operating current would damage the strips and the Current ADJ. Range of all the above drivers max out between the typical and max current of the LED strip would make them all candidates for the selected light?

First, I’d skip the XLG320-L. It’s a bit weird in its characteristics and doesn’t want to run at less than 150v.

LED devices generally should not be run beyond their maximum rated current. The motivating factor behind this however is temperature, and any operating condition that results in the internals of the individual LED devices getting much beyond 85°C will significantly decrease device service life, regardless of drive current. That said, if the devices are kept sufficiently cool, operating at 1.4A is within reason.

@bsaylors

Current Adj Range: The first two show a 700 – 1400mA and the third shows 500 – 1400mA range. What is the significance of the first number being 700 on the HCG and 500 on the XLG?

These indicate the lower limit of the adjustment range.

@bsaylors

DC 0/1-10V dimmer: Per the Meanwell datasheet “Output constant current level can be adjusted by applying one of the three methodologies between DIM+ and DIM-:” Back to the previous question, would this dim the lights all the way to zero or just to the bottom of the Current Adj Range? Looking through the TechForum I think the 500-1400 and 700-1400 are the sweet spots of the driver.

For devices with both an internal adjustment knob and a dimming input, the knob typically adjusts the maximum value, while the dimming input controls the fraction of that maximum value that is applied. Note that for proper operation, the forward voltage of the LED string should remain within the specified voltage limits, regardless of the output current setpoint.

@bsaylors

As a follow on, IF the lights are dimmed and only running at ~60% to reach the max PAR/DLI needed , probably not best practice, is it safe to assume additional strips could be added to the same driver? Which circles back to the multimeter question of being able to see input / output voltage at each strip and being able to determine how much voltage is being consumed by each strip. Would this be a bad idea or a common practice?

Operation at reduced current levels will generally reduce device temperature and extend service life–nothing wrong with that, aside from buying more hardware than needed.

While reduced drive current does result in lower forward voltage, the effect is highly non-linear and generally does not afford opportunity to add additional devices to a series string. Typical practice is to select an LED load targeting the mid-range of the driver voltage specification (or pick a driver to match the LED…) and then verify that shifts in LED forward voltage to to changes in operating current, temperature, etc. do not cause the forward voltage of the LED load to shift outside the driver’s specified operating range.

All this said, I would strongly recommend against the use of drivers capable of delivering more than 60V or so in a DIY context, due to significant shock hazards that exist above that level.

[bsaylors]

Rick, Thanks so much for the reply.

To repeat back, to make sure I understand. Count on each light strip pulling the ~42 - 45Vdc regardless of anticipated dimming. The dimmer function won’t result in a linear drop in voltage.
Also, running towards the upper limit of the driver is not a good idea, as it will create more heat and shorten its life.
Sorry for the copy/paste problem when listing the 240H and 320H drivers. So when you say try to select a load targeting the mid-range of the driver. Looking at the voltage ranges the 240H is 89 - 179V and the 320H is 114 - 229.which would make the midrange ( (max -min) /2) + min)). 134V for the 240H and 171.5V for the 320H. Putting the 240H dead on for the mid-range for three strips @ 45V each (3 x 45 = 135) and the 320H looks to be a good fit for four strips (4 x 45 = 180).
Lastly, point taken about the voltage. Going way back in the time machine 40 years. The phone company would apply sealing current to problematic circuits, which I grabbed wires for one of these circuits thinking had been disconnected and fingers became stuck to the wires for a few unnerving minutes.

[rick_1976]

@bsaylors

Count on each light strip pulling the ~42 - 45Vdc regardless of anticipated dimming.

I’d use the 39-45v figures, and maybe expand that by 2V at either end for good measure.
Looking at the characteristics for the individual LEDs on the panel can show why. At the nominal test current of 65mA (red) the V(f) of a single LED is about 2.75v. Increasing the applied voltage by 0.15v or so (about 6%) roughly triples current flow (green) while moving that same amount in the other direction results in nearly zero current flow (purple).

In addition to that, the entire curve can shift left or right by roughly that same 0.15V as a result of manufacturing variations: The voltage that barely gets a glow from one LED off the assembly line might fry the next one.

@bsaylors

running towards the upper limit of the driver is not a good idea,

That’d typically be more true of the LEDs than the driver, but it’s not false for either.