WARNING: Do Not Use

Hey everyone. I've been playing around with this driver for a while now, and I have finally realized why it has managed to kill off THREE of my 445nm diodes....

This is the offending driver, by the way: 18V 5W Cree Circuit Board for Flashlights (16.8mm*5.5mm) - Free Shipping - DealExtreme

It's a great driver... when you get it to work properly. The problem is, the negative lead is situated rather closely to the edge of the board. And, due to the nature of this type of board, it requires being very close to the negative feed. Which is bad.

Why is this bad? It's a buck driver, and if direct negative feed it, it manages to pump out ALL the current it can. When there is even a slight contact between that negative lead and the case negative, your diode is fried.

So please heed this advice - I know I should have.

i used a driver like that for a while.
Why nut just use large heatsink around the board to make sure the negative never makes contract with heatsink or host?

Ever heard of hot glue or heatshrink? These drivers are meant to be put into a host with voltage running through the host.

I don't really understand why the negative output would dump all current if the negative input is shorted to it. The positive output is the "filtered" one, while having a common ground (as far as I know).

Have you run the driver through a testload? I think you will find that either the current is way too high or it has a dirty output or it was user error. Were you discharging the caps etc... every time you changed from battery to diode. Was the driver ever run without a load? This could have destroyed it and output "all" the current as you said.

I don't think its the drivers fault... Unless it is proven to have dirty output... till then my bet is on user error.

THe Negative path is not continuous.

So the - lead that runs from the driver to the Diode - cannot make contact with the - input of the Battery.

If it does you will get ~2.3A to the diode or something like that. I cant remember and will kill it like in your case. THis is why you cannot use this driver for red laser. Since they are case Negative.


If you look at any other driver the - is always close to the end of the board. Even the Micro boost which the - does not run through is close to the Edge.

You just have to be careful.

Use a testload next time. Thats what there for.


Sorry to here your Loose of 3 445nm diode. Ouch..

totally untrue.. I have bought dozens of these drivers, I am even using it in my 445nm handheld.. no way you can mess that up.. I don't know how you did it but this driver it's pretty good IMO.

by changing the resistor you change the output current, neither positive nor negative are continuous, how the hell did you kill those diodes with this???
I just don't get it..

so I don't see why I must NOT use this driver., it's actually better than most on DX!!!

So USE it PROPERLY and no harm done :beer:
check your circuit and see where your error is

I use this Driver for My C6 445nm Laser. Still alive.

Maybe you got The Crap ones. VERY sorry about your Diodes.

Isn't that the driver that is used in the Survival Laser?

Never had problems with this driver. I still use it in every 445nm build I make. Perfect. You just need to be a little more careful
@rhd, prolly.

It is *not* user error. I can assure you that much. The point of the matter is that if the negative lead and the negative input short, you are dumping as much current into the diode as it can take, which, in my diode's case, was around 3A or so.

I tested it with a testload and I noticed this effect as well. I tested it with multiple of these drivers, and I noticed this effect as well. It's an actual fault of the driver's design.

A lot of people use these drivers, and you can go ahead and use them - I am just saying, it's not necessarily the safest driver. If you are using hot glue or something to specifically isolate the negative lead you should be fine, but the point of this thread is to warn those who otherwise may not know.

In my set up, the driver was slightly larger than pill size, so the driver was sitting in at a slight angle, and because the contact with the outer pill wasn't perfect, the uneven pressure from the batteries pushing on it managed to get that contact slightly with the host. I even tested this and noticed it with a test load (I thought I had managed to prevent it for this diode).

Anyway, if you are going to use it, this just serves as a warning: BE CAREFUL.

Well technically it is user error if you didn't isolate it properly. But I appreciate the heads up. It wasn't meant to be an attack on you or anything... but hot glue and a small spacer would have saved you $50 is all I'm saying.

Not really... it's impossible to isolate it with the negative lead that close to the edge of the board, where contact is SUPPOSED to be made. It's a design flaw, really, it is. Hot glue *may* have done the trick, but even then, with the solder a mil from the edge of the board, even a little knock can create contact.

wolfman can you take a picture of your driver pls.. i am looking at some of mine right now I don't see how an accidental contact can occur unless your driver isn't held in place and it's woobling around your host...

no way a little knock can short the input and output.. it's really not a design issue here..

if you hold in place your driver that it cannot touch the negative path you will never have problems with that .. really!

please do take a picture I want to see where do you think the design is not right..

Shouldn't need a picture. Should be pretty easy to see that the negative lead is obscenely close to the edge of the board.

Maybe try a different driver with that host? Or a different host with the driver.

If you placed a small plastic spacer over the board and hot glued the spacer in place as well as the contact for the negative output you would not be able to make contact with the host.

Plus that R22 resistor is closer to the edge than the negative ouput... just saying

ah I see what he means.. but then again just securely hold your driver in place and no problems at all just saying ...

oh and as I read your posts again and again.. are you trying to make contact from the components side??
use the bottom circle for the negative input not the upper side... :beer: sorry if I misunderstood that part from your posts.