Page 1 of 1

Small electronics/ LED brightness question.

PostPosted: Thu May 18, 2017 10:34 am
by Barbarianna
I have a red 12v led light. I need it to be about 50% as bright as it currently is. Only that- I don't need to be able to switch it from one brightness to the other, or dim it by varying degrees. Just 50% of current brightness, permanently.

What do I add to the circuit to make this happen? The specs say that it draws about 0.2A, running at 2W.

Re: Small electronics/ LED brightness question.

PostPosted: Thu May 18, 2017 10:37 am
by Strawman
Put a partially opaque cover over it that allows 50% of the light through.

Re: Small electronics/ LED brightness question.

PostPosted: Thu May 18, 2017 10:42 am
by Disastrous
I was going to suggest similar. A photographer's filter that allows exactly 50% of light through as a cover??

Re: Small electronics/ LED brightness question.

PostPosted: Thu May 18, 2017 10:45 am
by Barbarianna
I have thought of that but I can't. The shape and material of the LED prevents a cover from being put on it. Soft gel-like material, in a ring shape, with another fixing protruding from the middle. It would be a mega pain, and look dodge to boot, to make a nice-looking cover for it.

Re: Small electronics/ LED brightness question.

PostPosted: Thu May 18, 2017 10:50 am
by Il Duce
What about a dab of thinned-down Airfix paint?

Re: Small electronics/ LED brightness question.

PostPosted: Thu May 18, 2017 10:54 am
by Dirk
I don't think it can be answered electrically on what you have said Barbs. I think (but may be wrong) that some LEDs can be dimmed, others cannot. And those that can it depends on the type.

You could try sticking a resistor in, but it may not work and you can't say how big based on what you have said.

Pigeon would probably be the one to know though

Re: Small electronics/ LED brightness question.

PostPosted: Thu May 18, 2017 11:17 am
by Barbarianna
Yes, I will await Pigeon's response to this late into tomorrow morning when his vampiric musings emerge.

The paint thing could work ID, but IME this might result in a very uneven coating/dimming, which can't be wiped off this particular light surface without risking damage.

Re: Small electronics/ LED brightness question.

PostPosted: Thu May 18, 2017 3:55 pm
by Bodhi
What about buying an in-line dimmer for LED lights.

Re: Small electronics/ LED brightness question.

PostPosted: Thu May 18, 2017 6:49 pm
by CJ+
I thunk dimming LEDs was done by modulating the line voltage at high frequency or summat.

Re: Small electronics/ LED brightness question.

PostPosted: Thu May 18, 2017 6:55 pm
by Pigeon
It depends what sort of ballast it's got. If it's just plain resistive, then no problem. If it's a linear regulator, still possible but more fiddling needed. If it's a switching regulator, things get not simple.

Try running it off four AA cells for experiment. If that does the trick, then it's probably a plain resistive ballast, which would mean you can achieve the same result running it off 12V with a 47 ohm 0.5 watt series resistor.

If it doesn't, it may have a linear regulator, in which case you can still do it by using a bigger resistor, to deprive the regulator of its freedom of action. Something between 47 ohms and 100 ohms, but you'll need to experiment to find out exactly what.

If it has a switching regulator, the resistor trick may or may not work depending on the topology. It might be possible with probably more fiddling about to find the right value, or it might just stop working altogether before it gets dim enough.

Can you see the ballast? Some sort of "thing" between the incoming wires and the actual LEDs. It might be a resistor, it might be something electronic with just tiddly little bits on (linear regulator), or it might be something electronic with some kind of wee coil on it (switching regulator). Also, how many LEDs has it got? I'm assuming just one big one.

Re: Small electronics/ LED brightness question.

PostPosted: Thu May 18, 2017 6:59 pm
by Pigeon
CJ+ wrote:I thunk dimming LEDs was done by modulating the line voltage at high frequency or summat.


It's done by varying the average current. What you describe is basically the principle of a switching regulator, but there are other methods; which is used depends on whether the manufacturer was going for efficiency or cheapness, and also what kind of supply it's supposed to be run from.

Re: Small electronics/ LED brightness question.

PostPosted: Fri May 19, 2017 12:01 am
by Barbarianna
Pigeon- thanks. Your solutions are remarkably similar, though far better resolved, than my guesswork.

I have ordered a 100w resistor to put in line.

My reasoning was murky, but had to do with reasoning was sonmething to do with allowing for forward voltage, then power calculation as a function of current and resistance, and plugging the result into V = IR, to give a result of c. 100-150 ohms for an in-line resistor. The fudge would be to not drop the current so low as to deactivate the led completely, so i went for the 100. I do know there is a different version of the same bulb that is dimmer but with the same basic design, so i am hoping this will allow mine to run more dimly.

Re: Small electronics/ LED brightness question.

PostPosted: Fri May 19, 2017 2:35 am
by Pigeon
I take it you mean 100 ohms, not 100W :)

Did you, then, try it on four AA cells, and find it was no less bright?

Way it works is:

Assuming it has a resistive ballast.
Voltage drop across LED = 2V (assuming single emitter or parallel emitters) (red LEDs are always 2V). You knock this off the 12V supply voltage to do the calculations, because it is constant and does not vary (significantly) with current.
Voltage drop across ballast resistor = 10V (this is the important bit of the voltage)
Current = 0.2A
=> ballast resistor that it has already = 50 ohms

To halve current, double ballast resistor; this means, add another resistor of the same value (ie. 50 ohms) in series. 47 ohms is the nearest E12 preferred value to 50 ohms. (51 ohms in the E24 series is closer, but E12 values are more common, and it's not that critical.)

If it has a linear regulator ballast, which is a thing that acts like a resistor that gets bigger the more voltage you put on it, then that will compensate for you adding external series resistors by reducing its own resistance, until it gets to zero and runs out of range. So you need to make it turn itself all the way down to zero and supply all the necessary resistance externally yourself. That means adding a 100 ohm external ballast resistor.

If it has a switching ballast, you basically need to do the same thing, but it may not work. A linear regulator is more or less guaranteed to turn itself down to zero if you take away all its headroom, but a switching one may, or may not, or may start doing something weird, depending on its topology. Impossible to predict without seeing its actual circuit.

Re: Small electronics/ LED brightness question.

PostPosted: Fri May 19, 2017 9:10 am
by Barbarianna
Ah yes of course, the topology. I always omit to consider the topology.
Much to my chagrin at meetings of the radio-enthusisats' circle in Belarus Ministry of, er, Sound!

Re: Small electronics/ LED brightness question.

PostPosted: Mon May 22, 2017 7:54 pm
by Barbarianna
Hotly anticipated update!

The 100ohm resistor has only been and gone and done it! A lovely, neatly dimmed light now graces uts way from the bulb, rather than the bedevilled stellar explosion from previous days. We are saved!

Return to your stations, all is well.

Re: Small electronics/ LED brightness question.

PostPosted: Mon May 22, 2017 7:58 pm
by Hooli
Hurrah for dullness.

Re: Small electronics/ LED brightness question.

PostPosted: Mon May 22, 2017 9:16 pm
by Pigeon
Oh goodie!