LED Lighting (350 mA). Is there any standard for # of LEDs is a string?

[Benta]

The title says it best.

My idea is reusing laptop PSUs (normally 19 V, 90 W = 4.7 A) for lighting. The landfills are full of them, and I think it's a recycling project worth considering.
But to keep it simple, 19 V would only supply 4-LED strings.
Is this a normal length? Is there even some kind of standard?

Thanks.

[tom66]

Not as far as I'm aware.  Most LED power supplies are constant current, with a maximum output voltage under SELV limits (~100VDC is as high as I've seen them.)

If you want to drive LEDs with a laptop supply, you can, but be aware that it will require a power resistor to balance the current between each string.  That resistor might well drop 5V @ 350mA = 1.75W.  So not insignificant power to dissipate.

[Benta]

Thank for your reply.
No, not resistor-drop, but active regulation and splitting. I'm fully aware of how to drive LEDs.

[T3sl4co1l]

Good LEDs are down around 3.0V; should be able to run at least five, preferably six at exactly 19V.  Even at 3.6V, five should be easy.  If they're over some distance, one less would be acceptable of course.  Four may be about ideal if using a CC buck converter to drive them.

Yeah, not aware of a standard either, but on the other hand, laptop supplies are surprisingly well agreed on something around 19V.  Sounds like you've got your standard!

Tim

[Zero999]

Good LEDs are down around 3.0V; should be able to run at least five, preferably six at exactly 19V.  Even at 3.6V, five should be easy.  If they're over some distance, one less would be acceptable of course.  Four may be about ideal if using a CC buck converter to drive them.

Yeah, not aware of a standard either, but on the other hand, laptop supplies are surprisingly well agreed on something around 19V.  Sounds like you've got your standard!

Tim

I've previously run six old LEDs off a 19V constant voltage power supply, but they had forward voltage of more than 3V at the rated current, so were been under-run.

Six modern LEDs, run at 80% of the rating, with the standard op-amp+reference+transistor+current sense resistor, is feasible off 19V.

You don't need anything fancy. The crappy old LM358, a generic NPN transistor, such as the BC337 and a reference, such as the TL431, with a potential divider to make the sense voltage in the hundreds of mV.

[T3sl4co1l]

A four-transistor circuit will also do:



But this needs about as many components / as much cost, and you may find the simplicity of an off-the-shelf LED regulator is worth the purchase.

Ed: to expand on this a little further, it'll take about 5 components (discounting the reference: pull-up, zener/431, divider) to use the op-amp, about 10 here -- to build two channels, while making use of dual components, except the power stuff.  And not much power is really needed, so a dual output transistor would be fine here too.  Say '4401 BJT, or preferably low-Vce(sat) type, or NMOS as shown.

I suspect the cap can be optimized out by tuning tail current (increasing R2, R3) to achieve compensation.  And D1 doesn't really matter, as long as Vgs(max) < VCC, or a BJT is used.

Hopefully, the same is true of the opamp case; perhaps shopping around for suitably slow, or enough phase margin, op-amps would be worthwhile here, eliminating compensation components.

Heh, since supply is regulated, well, no big deal using that for reference, just a bare divider will do in this case.

Foldback would be nice, to limit power dissipation in case of shorted LEDs; should be easy enough to add two resistors and call that done.  May need to use SOT-89 or larger transistors to ensure adequate dissipation at worst case (say, half nominal load current at a couple volts?).  Could also use a thermistor for added thermal protection, with the downside that it can't be supplied from a resistor pack (dual/quad).

Tim

[jmelson]

The title says it best.

My idea is reusing laptop PSUs (normally 19 V, 90 W = 4.7 A) for lighting. The landfills are full of them, and I think it's a recycling project worth considering.
But to keep it simple, 19 V would only supply 4-LED strings.
Is this a normal length? Is there even some kind of standard?

Thanks.

For ~ 90 W, you'd need a boost regulator, there are tons available.  I have built up strings of 20 lighting LEDs running at 300-350 mA.
They run at about 65-68 V for the string.  So, that's only about 20 W/ string.  You could put 4 strings in parallel with individual boost regulators or just a small resistor to balance current between strings.  Watch out, though, a bank of 80 1 W LEDs puts out a LOT of light.  We have 3 strings of 20 and they light up our whole large kitchen quite well.
Jon

[tooki]

But to keep it simple, 19 V would only supply 4-LED strings.

How do you arrive at this number?!?

LEDs made for lighting come in various voltages, and some of the most powerful ones do not have the highest voltages. For example, the Samsung LH181A LEDs I used in a recent project are rated from 2.82V at 350mA to at 1.5A at just 3.15V. So you could almost run 7 of them with 19V, and given that many “19V” laptop chargers are actually a bit higher (19.2-19.5V), 7 would be no problem.

19V/4 is 4.75V per LED, which is far too high for almost every visible-light LED. You have to get up to UV-B LEDs to require such a high voltage. (Yes, some very high-power blue LEDs use voltages in this neighborhood, but it’s rare.)

[bdunham7]

Is this a normal length? Is there even some kind of standard?

I recently made a bench light running 9 in series from a 30 V supply and in the process of making the PSU I learned a few practical things.  First, the 'normal' way of doing LEDs in a standalone situation seems to be to use as long a string as practical.  Second, for the white-ish LEDs that I tested, the voltage at the rated current my vary  a bit, perhaps 3.0 to 3.3V, but the knee of diminishing luminosity/current is in the 3.0V range.  I didn't test that many so perhaps this isn't a general rule, but I didn't find a good reason to go much over 3.0V when making my lights.  So you should be able to run 5 lights using an LM317 or similarly crude current regulator, six if you do a bit better.

If you are reusing a laptop brick and are willing to chop it open, it might not be that hard to tap into the voltage feedback circuit and convert it to a voltage-limited constant current supply. 

[nctnico]

The title says it best.

My idea is reusing laptop PSUs (normally 19 V, 90 W = 4.7 A) for lighting. The landfills are full of them, and I think it's a recycling project worth considering.
But to keep it simple, 19 V would only supply 4-LED strings.
Is this a normal length? Is there even some kind of standard?

AFAIK not. From what I've seen many LED lamps made from a string of LEDs require 50V to 80V to drive. A long string reduces wiring costs and lowers current (=less fire hazzard). You'd need a constant current boost converter.

[Benta]

OK. no standard for power LED strings. I can live with that, but it's a bit of a nuisance.
I'm going completely KISS here, so no step-up converters or other things. This has to be as simple and cheap as possible.
Also, when I say "LED string", I don't mean metre-long things with 100s of LEDs. I'm talking about white power LEDs for lighting.
At this time, it seems the electrical optimum is 4 LEDs per string at 19 V supply. The laptop supply could then drive up to 13 strings (12 is probably a better number) that can be mounted optically separately or in groups.
Each string will have its own simple controller.
The simplest I've come up with is this using Osram OSLON SSL 150 whites, and it works like a dream, spot-on at 350 mA:

[Zero999]

How many strings are you planning to have per luminaire?

Consider having a lower reference voltage, then you can have an extra LED or two and cut down power dissipation in the transistor.

[mariush]

You could use a basic linear driver like  BCR421 which can do up to 350mA (absolute max rating somewhere around 500mA) using an external resistor (defaults to 10mA without one) : https://www.digikey.com/en/products/detail/diodes-incorporated/BCR421UFDQ-7/7930771

32 cents if you get 100, 20 cents if you buy 1000 ... datasheet here : https://www.diodes.com/assets/Datasheets/BCR420UFDQ-BCR421UFDQ.pdf

Note that the enable pin can only handle up to 18v but it can also be used as pwm. The IC has a voltage drop of around 1v ... if you allow for 1.5v drop you're looking at 18v - 1.5v = 16.5v / 3.2v = 5 leds in series.

If you go with a step-down led driver, you could have groups of 3 leds in parallel, so your step-down led driver would support a wider range of laptop adapters ex 12v ...18.5v or even only 16.5v .. 18.5v and output  8.5...10v (3 leds x 2.8v...3.3v in series)


for example AL8843 : https://www.digikey.com/en/products/detail/diodes-incorporated/AL8843QSP-13/10668306

it can do up to 3A , so you could safely do  8 groups of leds in parallel as 8x 0.35 = 2.8A  ... so then have 3-5 leds in series, then 8 such series in parallel, for a total of 24-40 leds controlled by one driver.

[nctnico]

LEDs strings in parallel don't work well. You'll see differences in brightness. I've seen current mismatches up to 25%. For the best results, you need a current source per string (even if it is a simple dropping resistor).

[mariush]

You can bin the leds so that the forward voltage of every series ends up very close to the forward voltage of other series that are in parallel. Super easy to make a test fixture of some sort to measure the forward voltage and brightness at various current levels.

They do this parallel thing with led strips for monitor backlights and in other places and nobody complains about differences in brightness.

[tom66]

I don't know of many monitors that don't have independent current balancing for each parallel string, but even then the most common approach is all LEDs in series.  Yes that does mean if one fails, the whole panel goes dead but hey, e-waste is not their problem.

[Benta]

How many strings are you planning to have per luminaire?

Consider having a lower reference voltage, then you can have an extra LED or two and cut down power dissipation in the transistor.

Not decided yet. I'll need to experiment a bit with the subjective optical power, which also has to with materials, optics and enclosures.

I though about using a lower reference voltage, but the simplicity and incredibly good current stability of the TLV431 solution led me to that choice.
Also, the output of the laptop power supply is marked as pulsating DC, so I'll need a bit of headroom to be sure of correct operation.

Thanks for Your inputs All.

[Benta]

You could use a basic linear driver like  BCR421 which can do up to 350mA (absolute max rating somewhere around 500mA) using an external resistor (defaults to 10mA without one)
32 cents if you get 100, 20 cents if you buy 1000

Nice little device, but not very precise. With the LEDs being 1.50 Euro a pop (=6 Euro), I'd rather be certain of my LED current.

But Thanks.

[Zero999]

How many strings are you planning to have per luminaire?

Consider having a lower reference voltage, then you can have an extra LED or two and cut down power dissipation in the transistor.

Not decided yet. I'll need to experiment a bit with the subjective optical power, which also has to with materials, optics and enclosures.

I though about using a lower reference voltage, but the simplicity and incredibly good current stability of the TLV431 solution led me to that choice.

I was thinking use the TLV431 as a reference for an op-amp. If you used a quad, such as the LM324, it would actually work out cheaper than four TLV431s.

Also, the output of the laptop power supply is marked as pulsating DC, so I'll need a bit of headroom to be sure of correct operation.

Thanks for Your inputs All.

That's odd. Does the label/manual say what frequency it's pulsating at? What does the waveform look like on an oscilloscope? I think it's likely you're mistaken. The symbol for steady DC is a solid line over a dashed line.
⎓

Oh and just to be pedantic, there's no such thing as pulsating DC. It's really AC with a DC bias, but I know what you meant.

[Benta]

I was thinking use the TLV431 as a reference for an op-amp. If you used a quad, such as the LM324, it would actually work out cheaper than four TLV431s.
-------
That's odd. Does the label/manual say what frequency it's pulsating at? What does the waveform look like on an oscilloscope? I think it's likely you're mistaken. The symbol for steady DC is a solid line over a dashed line.
⎓

To your first point: that would defeat the purpose here. The idea is to have the controllers as POL regulators, one regulator per string of four LEDs. This allows me to place the strings in any physical configuration. Consider it as a "19 V, 350 mA, 4-Led module".
Imagine a "splitting box" with twelwe 19 V outputs powered from one laptop supply. The LED strings will do the rest themselves.

The voltage headroom on my design also allows me more freedom in cabling (losses).

Concerning the "pulsating DC": of course you're right, bad choice of words on my side, sorry. I interpreted it as "mediocrely regulated DC".

-----------------------

Something that might interest you (you seem to like playing with circuits).
The TLV431/Darlington power BJT is dangerous, which I only found out after having serious problems with my first circuit. An emitter follower (which this circuit basically is) can oscillate with significant amplitude at >1 MHz frequencies. In a discrete circuit it's usually not a problem, but add in a TLV431 which has a lot of gain, and it gets very hairy.
The TLV431/BD679 combination proved impossible to tame, which made me go back to the BD437 plus a relatively large base resistor. This is stable and works reliably. R1 as 1 kohm and R2 as 5.6 kohm works even better.
You live and learn


PS: (added later). Do you remember the "backlighting" thread where temperature stability was brought into the discussion?
Well, look at this:

[T3sl4co1l]

Try a resistor from Rshunt to REF, and a C or R+C from REF to A.

Tim

[Benta]

Try a resistor from Rshunt to REF, and a C or R+C from REF to A.

Tim

Sorry, but throwing capacitors at the problem does not help (I already tried all that in all permutations).
I suggest you read the following (scroll down to "Advanced Amplifier Topics and Design Examples"):
https://www.sciencedirect.com/topics/engineering/emitter-follower

[T3sl4co1l]

No not the emitter follower, the TLV431.

Follower stability is just, add base resistance.

Tim

[Benta]

Seems you're in stenographic/shorthand mode today. Not really intelligible.
Perhaps tomorrow.

[T3sl4co1l]

I mean if you want to discuss, you're welcome to, but...

but add in a TLV431 which has a lot of gain, and it gets very hairy.
The TLV431/BD679 combination proved impossible to tame, which made me go back to the BD437 plus a relatively large base resistor. This is stable and works reliably.

Try a resistor from Rshunt to REF, and a C or R+C from REF to A.

You already discovered the base resistor, that's fine.
You seemed to indicate reducing TLV431 gain might be helpful...
...That's the way to do it.
It's not clear if your "impossible" search included such.  So it's worth mentioning.

Perhaps a diagram is missing; I take words for granted much too often.

This for example stabilizes a '431 into a C load:



You already have the R3, more or less.  R2 is missing.  C1 can be R+C for additional phase margin.

Some damping (R+C) at the output (C-E or C-GND) can also help, when the load is squirrelly.  LEDs wouldn't usually be long enough I think, but a high fT transistor might be able to anyway.

I could further describe how to solve for such values, but I'm guessing that isn't necessary here.

Not sure how much more detail to add.  Emitter ferrite bead?

Tim