LED drivers / concerns or opinions?

[blasterman]

Greetings everybody. For the past couple of years I've been building really cool high powered LED fixtures (reef lights, DJ lights, whatever), but have been frustrated when dealing with lasers. Commercial fixed lasers are either uber expensive, or total junk. There's nothing in the middle. Quite awhile ago I thought about using LED drivers with laser modules, but I was instantly told it wasn't possible. Now I'm reading that the big 445's are being powered by fairly generic current regulated LED drivers.

Ok, that changes things - big time. If LED boost drivers are stable enough to do this, then I don't see why LuxDrive Bucks couldn't do the same - or am I missing something? These in turn are easily driven by low voltage AC input power supplies, which is why I'm asking. My next question is if this works with the big 445's would it also work with >100mW Greens and the popular 200mW red?

I have a ton of projects waiting and a lot ofvery cool stuff to build if indeed the LED driver approach works, so any advice welcome.

 

[jib77]

Because 445's are TOUGH. They take the ripple and transients inherent with LED drivers and don't complain, where as any other normal diode would fry under those conditions.

 

[blasterman]

Do explain how a cheap Chinese driver mounted on the end of a laser pointer and powered by a non-linear power source as a battery is not prone to ripple and transients while a state of the art regulated LED supply is? :beer:

I've measured some of my MeanWell LED supplies to the best precision of any multimeter I can get in terms of stability.

I am aware that the bigger the laser diode is the more tolerant it is of voltage spikes, and obviously a battery is physically capped in terms of the voltage it can produce. However, transient current inrush/voltage from LED supplies is easily negated with the addition of more diodes in series or a basic thermistor.

 

[Cyparagon]

Good question. I have yet to see someone put one of these on an oscilloscope. The only evidence is anecdotal. "It didn't die."

 

[blasterman]

I hear ya'; "didn't let the magic smoke out - so it works". :thinking:

As per the other thread, I suppose I should just get a cheap 100mW green pointer, tear the driver out and fire it up with a 350mA Buck with a 12volt source on the other end and see what happens. Maybe throw some 3watt LEDs in series to gobble up any transients. 100mW Greens are about ~350mA, correct? *If* that works, then the door is open.

I have a $8,000 Q-beam in the rafters right now running at around 300mW, and after several years I'm starting to worry about the power supply. I'd much rather replace it with a bunch of smaller lasers anyways, not that I care because I have a variance. I ain't shooting holograms, so TEM mode or coherence isn't an issue.

 

[jib77]

Do explain how a cheap Chinese driver mounted on the end of a laser pointer and powered by a non-linear power source as a battery is not prone to ripple and transients while a state of the art regulated LED supply is? :beer:

I've measured some of my MeanWell LED supplies to the best precision of any multimeter I can get in terms of stability.

I am aware that the bigger the laser diode is the more tolerant it is of voltage spikes, and obviously a battery is physically capped in terms of the voltage it can produce. However, transient current inrush/voltage from LED supplies is easily negated with the addition of more diodes in series or a basic thermistor.

Because linear ICs are excellent at ripple rejection and the input and output caps take care of the transients. The understanding I came away from reading this forum is this, don't use a LED driver unless you add output filtration. Not that they are forbidden. So if you can scope them and they seem fine, go for it. But remember most people here do not have access to an o-scope, so better for them to go with a known proven LD driver than to blow a pricey diode.

 

[blasterman]

I hear what you're saying, but 'proven LD driver ' is a bit of an ambiguous term. This can refer to a $5.00 driver in parts, or $5,000, and I'm trying to get at the functional difference between LD driver and LED driver and how each one differs other than the former claiming to do a better job at voltage regulation. Both use the same parts in base form. I'd simply feel better using a LED driver and adding something additional to get where I need.

I understand the increased sensitivity of laser diodes over LEDs to transient voltage, but I'm not sure how/where this is a problem with LED drivers and not LD drivers. Lasers can be modulated, and quite rapidly, and how exactly does a laser diode distinguish from a modulated source from a voltage ripple? This would lead me to conclude it's excessive transient spikes that kill the diode and not ripples, but I'm guessing. Again, good LED drivers are pretty darn stable and inexpensive, but they all tend to have an issue with current inrush when fired up, and this has cost me plenty of LEDs. It's easy to compensate for, but want to be more certain this is the gremlin I'm trying to avoid.

 

[rhd]

Blasterman - I agree with your sentiment whole-heartedly.

The two most inconsistant drivers I've ever used were in fact a Microboost and a Flexdrive. The former would fluctuate wildly (and rapidly) in current supply (definitely not a constant current source). So, next time around, I ordered a flexdrive pre-set to 1A. About 2 months later I actually received it, and it didn't work at all. With a pre-set driver, there's not much you can do to screw it up. Solder on the leads, hook up a test-load, battery, and give it a metering... Well, my flexdrive gave literally nothing as an output, and I checked my lead solders with a magnifying glass - no issues there.

"Proven LD driver" is not only ambiguous, but meaningless in my mind too - at least at this level. I'm sure $5,000 laser drivers are "proven", but I don't buy into the notion that the Dr. Lava drivers are "proven". Indeed they have thoroughly disappointed me. The only "proven" LD driver that I've been impressed with - is the Groove 2 Driver - and that's probably because it's just an Lm1117 chip

Conversely, in my experience, I have a few 445s and 405s running on $5 boost LED drivers (the NJG-18s) and they'll take a real torturing. Having said that, I usually add a big cap and a smaller ceramic cap to the output. But that's not a big deal to do. So I'm with you - I think there are plenty of alternative driver options (including LED boost drivers) that work just as well - and IMO better than the Dr. Lava drivers.

 

[LarryDFW]

I understand the increased sensitivity of laser diodes over LEDs to transient voltage, but I'm not sure how/where this is a problem with LED drivers and not LD drivers. Lasers can be modulated, and quite rapidly, and how exactly does a laser diode distinguish from a modulated source from a voltage ripple? This would lead me to conclude it's excessive transient spikes that kill the diode and not ripples, but I'm guessing. Again, good LED drivers are pretty darn stable and inexpensive, but they all tend to have an issue with current inrush when fired up, and this has cost me plenty of LEDs. It's easy to compensate for, but want to be more certain this is the gremlin I'm trying to avoid.

Laser are sensitive to optical damage caused by brief transients.

You will only see these with a fast scope.

They are also more sensitive to static discharge.

A "transorb" will protect from the static discharge.

MeanWell is a good supply builder,

but even their supplies would need to be examined.

LarryDFW

 

[blasterman]

Hey guys, thanks for all the tips.

I have a 445 on order from DangLasers, and am going to try it out on various reputable LED drivers ranging from 350mA to an adjustable MeanWell 48D that goes up to 1.4amps. I'm talking to some electronic geeks more versed in the minor nuances of current regulated supplies and getting their advice on what if anything I should place in series with the out-put side to smooth out any power up transients.

The advantage with higher quality AC based LED supplies is they have excellent 'run mode' stability. Better than any battery powered boost or buck type driver I've tested. The disadvantage with AC based supplies is they throw a wallop of a current inrush when turned on, and this is what I want to guard against. I'll report back when things are tested (or I have a dead diode).

 

[rhd]

I've got a few of these:
$5.35 - 17mm 1000mA 5-Mode with Mode-Memory LEDDriver Circuit Board for Cree XR-E Emitters (0.9~4.5V Input) - Flashlight Parts and Tools
and these:
$3.10 - AK-007 0.8~1.5V 3-Mode Circuit Board for Flashlights (11.9mm) - Flashlight Parts and Tools

Anyone have any thoughts or experience with either as an LD driver?

 

[blasterman]

AC powred LED driver functionality has been confirmed baby!!!!!!!

I got a 445 from Danglasers today and just finished wiring it up.

I used two AC powered LED drivers. The first was a 24volt / 350mA LightTech driver. The second was a 48volt 700mA MeanWell. The MeanWell is a brute - it will pop an entire series of 3Watt Crees if connected hot, and I've even torched a few 3watt Rebels with it fired up cold. You can see it throw a small arc when it makes contact hot and will vaporize solder. Thermal inrush is no joke with a 48volt LED driver.

To get around this problem and knowing the extreme sensitivity of LED diodes I tried another widget I've been dying to cofirm. I wired a 40 -.2ohm 8 watt Thermistor is series with the diode and driver. The 350mA driver fired up safely, and then the big MeanWell with the thermistor in series. There's no question in my mind the laser diode would have fried in a nanosecond on the MeanWell without the thermistor.

Didn't run it for long because it's not sinked, but I'll run it longer tomorrow when that's accomplished.

Oh yeah....this is my first experience with 445. Those things frikken hurt when you pass your hand in front at 700mA. I mean--hurts a lot.




I added a 50~.2ohm 8watt Thermistor