Large Scale Central

LED Drivers

Electronics
#1

I am a little out of my depth here, so I hope one of our electronic experts can offer some advice.

These guys are LED drivers designed to maintain constant brightness on some high-power LED spotlights, (which are on the other side of this clump of rock.) The spotlights are very, very bright. In the store, under store ceiling spots, it wasn’t apparent, but at home we keep it not so bright, so they are overpowering.

My usual solution would be to (a) install a dimmer, or (b) run 2 LEDs in series instead of parallel [i.e. use 1/2 as many drivers.] It appears the Drivers will just accept the 2 LEDs and adjust their output accordingly. So here’s a few questions I am pondering.

  • can I just reduce the AC power feed to get a lower DC output? As one does on a light dimmer?

  • is there an LED driver that also incorporates a dimmer?

Any advice welcomed.

#2

Since I’m as qualified as you, I’d try painting the leds with washes until I dimmed them to suit me.

#3

John Caughey said:

Since I’m as qualified as you, I’d try painting the leds with washes until I dimmed them to suit me.

John I just color mine with yellow highlighter and it take the white off, Bill

#4

Pete Thornton said:

I am a little out of my depth here, so I hope one of our electronic experts can offer some advice.

Pete, not sure you can dim them, they make special LED’s that can be dimmed but usually dimmers reduce voltage and being as LED’s ARE current driven I don’t think it will work, on my small LED’s light scale house lighting I just use a yellow highlighter pen and it works great but not knowing shape and size of yours it’s hard to say. Krylon makes glass paint that is clear but tinted in several colors, next I would ask like some one like Josehf Murchison with the www.instructables.com, Bill

These guys are LED drivers designed to maintain constant brightness on some high-power LED spotlights, (which are on the other side of this clump of rock.) The spotlights are very, very bright. In the store, under store ceiling spots, it wasn’t apparent, but at home we keep it not so bright, so they are overpowering.

My usual solution would be to (a) install a dimmer, or (b) run 2 LEDs in series instead of parallel [i.e. use 1/2 as many drivers.] It appears the Drivers will just accept the 2 LEDs and adjust their output accordingly. So here’s a few questions I am pondering.

  • can I just reduce the AC power feed to get a lower DC output? As one does on a light dimmer?

  • is there an LED driver that also incorporates a dimmer?

Any advice welcomed.

#5

Since LEDs have a constant voltage drop, dimming is generally accomplished with a PWM circuit. There are dimmers that can go on the DC side and chop it up, but I don’t know what kind of current you need. Something of this nature: https://www.amazon.com/SUPERNIGHT-Aluminium-Single-Channel-Controller/dp/B00JQ437YM/

Dimming the AC side won’t accomplish that unless the power supply is designed for it.

#6

I have been installing LED down lights in my train room ceiling that came with those drivers. Knowing enough to be dangerous I measured the voltage being drawn under load (mine were 24v) and then used a single large 5A 24V supply to drive them all through one of the dimmers like Eric posted above. That works pretty good except at the bottom of the scale they pulse. I’m told that’s because the PWM frequency in these cheap dimmers is too low. To solve that, for a ‘very dim’ setting I switch in an DC-DC voltage converter set to a level that drives them where I like. If you want a single fixed brightness, one of these cheap DC-DC Buck converters and a power supply of proper voltage and current might do it.

#7

dimming is generally accomplished with a PWM circuit.

Eric, thanks. I wondered about that. I have a little PWM circuit in one of my locos for manual control - probably a lower current version of the one you suggest.

and then used a single large 5A 24V supply to drive them all through one of the dimmers like Eric posted above.

Jon, the LED drivers are apparently designed to control the current as the bright LEDs get warm, thus preventing “thermal runaway”. Your dimmer is probably controlling the warmth, so you aren’t in danger of them overheating. [LEDs overheating? Never heard of it!]

In either case, I worry that the dimmer PWM will confuse the LED driver current detection?

#8

Pete - This is how I think it works. I’m sure Greg will jump on me if I don’t have it right. I put my meter on the output of one of those LED drivers and measured both voltage and current. The current reading I made note of so I know how many I can drive from my supply. The voltage was noted as the operating voltage which I matched with a proper supply. When the LEDs are operating at their optimum current; the voltage will be a fixed number. So long as I drive them with the same voltage, they will not draw more current than they drew from the LED driver.

if that photo was of your setup you would need one of those PWM dimmers for each LED driver. That was my situation as well. I also did not want to run a lot of 110V wire and all the hardware it requires all over my basement ceiling. By operating low voltage, I only need drill a hole in the drywall for the lamp and snake some light wire back to the central power supply and it’s all to code.

Speaking of code - If that is your setup in the picture, I would recommend a metal box for your drivers and 110V connections. It looks like you are mounting on a piece of wood. If your 110V connections fail for any reason, that wood might ignite !

#9

Jon Radder said:

Speaking of code - If that is your setup in the picture, I would recommend a metal box for your drivers and 110V connections. It looks like you are mounting on a piece of wood. If your 110V connections fail for any reason, that wood might ignite !

Pretty much

#10

Perhaps I missed it but what are we lighting up other then who ever accidentally mis uses that 110 connection?

Honestly Bills suggestion of the sharpie works very well …so does sanding the LED …but I don’t think I remember seeing what we were lighting up ?

#11

" Rooster " said:

Perhaps I missed it but what are we lighting up other then who ever accidentally mis uses that 110 connection?

Honestly Bills suggestion of the sharpie works very well …so does sanding the LED …but I don’t think I remember seeing what we were lighting up ?

The object being lit wasn’t really relevant, but as you asked, it’s a glass sea life sculpture. Here it is:

#12

" Rooster " said:

Jon Radder said:

Speaking of code - If that is your setup in the picture, I would recommend a metal box for your drivers and 110V connections. It looks like you are mounting on a piece of wood. If your 110V connections fail for any reason, that wood might ignite !

Pretty much

The lighting was installed by an electrician, not by me.

#13

Jon Radder said:

Pete - This is how I think it works. I’m sure Greg will jump on me if I don’t have it right. I put my meter on the output of one of those LED drivers and measured both voltage and current. The current reading I made note of so I know how many I can drive from my supply. The voltage was noted as the operating voltage which I matched with a proper supply. When the LEDs are operating at their optimum current; the voltage will be a fixed number. So long as I drive them with the same voltage, they will not draw more current than they drew from the LED driver.

Jon, the problem is that if the LEDs are sufficiently powerful that they get warm, the voltage drops and then the current changes. I got my ‘thermal runaway’ explanation from googling LED drivers.

#14

OK - That makes electrical sense. If heat causes the resistance in the LED to change and the voltage does not change the current will - Ohms law. I assume the drivers respond to that and lower the voltage. I’ve not had that problem even when not dimmed mine do not get overly warm. They did come with a big honkin’ heat sink. Maybe that is keeping them from overheating. You can almost see the heat sink in this pic from Amazon…

The entire back is an aluminum heat sink.