Forgot your password?
User Journal


Journal by Khyber

Well, Soylent decided to rub me the wrong way.

Anyone want 26 vulnerabilites, POC, and sample code to what their site runs on, with full VM pre-loaded with POC and code snippets for testing?

User Journal

Journal: Flashlight Modification - From CFL to LED

Journal by Khyber

My old Coleman 5324 CFL lantern had the bulb die (ballast still good.)

However, I did not want to go back to the horribly inefficient CFL it used.

So I slammed in an MK-R LED with a heat sink.

This weekend, I'll be making a test video from the top of the nearby mountain. I'm going to try to spotlight my camera at my apartment while atop that mountain, and compare it to other light sources, like the LED on the iPhone 4S.

User Journal

Journal: Lighting LEDs directly from AC easily achieved

Journal by Khyber

I figured out a simple way to make DC-operated LEDs work from AC. Most people suggest using a bridge rectifier. Why not make the LEDs the rectifier itself?

So I salvaged some LEDs from old damaged panels, and used them to create a rectifier. However, I used a 12VAC 200mA max wall wart, not 120VAC from the wall (I did not want to wire up four 120V strings of LEDs plus resistors and a smoothing cap for flicker reduction) so the circuit could be made incredibly basic. Essentially, the LEDs themselves acting as a rectifier eat up all of the voltage. Reverse voltage breakdown is not a problem, as voltage seems to add both ways. 12V total Vf for the LEDs, but as each LED has Vr of -5V, the total Vr is -20V. Can't overpower that with -12V. I've had my prototype running an hour straight on a heat sink, and so far not a single issue.

Too bad this is too obvious and simple to be patented. But have fun with it!

The flicker is certainly there, but for horticultural applications, as long as the photon output is acceptable, it will work. This should simplify LED design drastically.

Have video and have some some neat pictures.

User Journal

Journal: Google Took My Money 1

Journal by Khyber

I've been using Google Helpouts for a week, now. I haven't done much, only two clients, but I was supposed to be paid. I was under the impression that I didn't need a bank account to use Google Wallet. The wording on their site is 'Recommended.'

Well, my customers have had money taken from them, and not delivered to me. In order to get this money, I *HAVE* to have a bank account because I am now a service provider. None of this is mentioned when I went to update and verify my Google Wallet information. I've been on the phone with Google, they've been unable to do anything, so they say. Google, acting as the middleman, has essentially taken my money and says they can't do anything about it until I get a bank account. That's not possible until I correct an outstanding balance. I was hoping to use Google Wallet to be able to save up that money and then pay off the balance then and there. Nope.

So I've given out services and Google essentially refuses to pay me what I'm owed until I go through a third party.

Help me spread awareness, if you can, by giving me a +1 at

Maybe Google will take notice and fix what logically is a violation of the anti-tying provisions of the Magnuson-Moss Warranty Act.

User Journal

Journal: Free LEDs!

Journal by Khyber

Just go look. You can get one free sample of any of those lights, but you can only pick ONE.

Dunno about you guys, but I picked the Bay Light.

It might help more to work in a lighting company. I am unsure if they have a choosing criteria for this.

User Journal

Journal: Going to a Low Tech Hobby 2

Journal by Khyber

A few days ago, I went for a walk down the local riverbed. To my surprise, I wandered upon a large cluster of grape vine, bearing wild grapes, tons of tiny ones.

Well, they're about 70/30 seed/fruit. Go figure. A little sample, and the flavor was a mix of tart and sweet. These would make a good wine, I figured.

I decided I'd go ahead and make a slightly more refined version of what the Romans and Greeks might have had. After the grapes were washed with a light bleach/water solution, rinsed thoroughly, and mashed, I had about 1.5 gallons of juice, to which another 1.25 gallons of water was added, plus 3 pounds of raw cane sugar, and 2 teaspoons of proofed bread yeast. All of this was put into a 3 gallon carboy with a rubber stopper and typical airlock. Right now, I'm getting a bubble roughly every 2 seconds from the airlock.

I'm expecting a wine with not a whole lot of alcohol, and a bit of sediment. There's not much point in doing a bunch of fermentations and filtrations on my first go-around. The smell right now is very light and quite pleasant. Almost like a perfected Boones Farm wine.

Have some HD video.

This should be ready in about 4 weeks, I'm not expecting a huge return on alcohol content given the yeast is past the best by date

User Journal

Journal: Taking A&D Electronics Down a Peg

Journal by Khyber

Well, for a very, very short time, I had a gig as an LCD panel repair tech. Until I learned that I was not going to get paid. So I quit, and straight to the labor board I went.

Looks like I stand to get about $3,200 max from them for their illegal practices.

I bet ChiMei, the company contracting A&D to do their panel repair, would *LOVE* to know one of their chosen repair depots isn't paying workers.

User Journal

Journal: Review: SinkPAD Direct Thermal Pathway PCB 1

Journal by Khyber

I've spent the last month testing and demolishing a bunch of new-design thermal PCBs from SinkPAD, and have come to the conclusion that this is a solid solution versus traditional dielectric PCBs commonly used for mounting LEDs. There are many pros, and a few cons, and I will address these in my review of their product, written below.

We will start with the good stuff first, the Pros. The SinkPAD Direct Thermal Pathway PCB is made from solid aluminum or copper (as of the time of this writing I had yet to test any copper PCB), so it's very good at thermal transfer. This helps make it easy to solder; versus a traditional dielectric board, the solder took to the SinkPAD in roughly half the time. This also comes in handy for reflow work and rework, as mounting and un-mounting an LED from the board is quick thanks to the PCB being constructed out of high-conductivity aluminum and copper. Repairing any mistakes takes minimal time and effort, a must for any prototype designer or hobbyist. The solder resist is also very solid, during my reflow cycling test I never once saw a solder bridge form, nor had any shorts. The same solder resist is quite durable against thermal shock as well. I torched it until it was smoking and crackling, near-black, and with no problems I was able to mount an LED to it and light it right up, and the remainder of the pad transferred heat exactly as it should. This durability is exactly what the military and police forces could use for LED-assisted weapons solutions, such as barrel or grip-mounted lights, or IR night-vision equipment. The aluminum boards are light-weight and stiff, very difficult to bend, even with a pair of clamps and pliers and my full weight applied.

Now for the cons. Due to how the thermal pathway is manufactured, there's an indentation on the back of the board. This means you need to use more thermal paste/epoxy, and you need to be more careful not to introduce air bubbles and gaps. On the aluminum boards, there is no solder pad on the back side, which would help greatly since solders tend to be more conductive than thermal pastes and adhesives. The copper board likely wouldn't need it, since copper is easy to solder. Speaking of copper, there are a distinct lack of copper options for many of their PCB choices. While this is understandable (many of these LEDs simply don't need it) the ability to have broader choice in materials would be nice. My final bad note is the difference between prototype boards I have versus some of the production boards. An older X-Lamp XR-E board had four screw holes for improved mounting, and they were placed so that most screws that could fit those holes would not have the heads short out on the contact pads. These holes are nonexistent on the star and disc boards. The star have their electrical contact points too close to the indents at the outside to allow screws to be attached there, but the circular boards look like they might have a better chance.

Overall, this is a solid piece of thermal management tech for LEDs. The ease of use and durability should appeal to a wide variety of LED consumers, from hobbyists, designers, and engineers to police, military, and even more groups of people and industries. Most of the cons are personal nitpicks and aren't exactly game-breaking important, and the pros certainly outweigh any con I've been able to notice. Highly recommended.

User Journal

Journal: SinkPAD & CREE, Together At Last!

Journal by Khyber

I finally got the right boards personally delivered to me. It was quite funny, exchanging tiny baggies of what any person at least ten feet away would assume were drugs right in the middle of a Del Taco parking lot. Then watching an accident happen not fifteen feet from us, between bicyclist and car.

Anyways, I just shot a little run of the MK-R, both of them, driven at 18w each. Note how they obliterate the Cree XR-E to the left, driven at 1.2-1.3w or so (3.7V rated Vf @ 350mA.)

I'm going to go step outside to get a night shot. Updates later in this journal post, in the comments.

User Journal

Journal: Night Test of the MK-R 2

Journal by Khyber

Finally got around to doing a night test of the MK-R.

I went out onto my patio, and put it right beside my 100w incan that typically lights the porch.

And here's the result. As you can see, absolutely no contest. Fluoro equivalent to 100w incan is a ~26w bulb, so we're looking at around 25% the power consumption of an equivalent CFL (LED driven at 6w) and way, way more light.

Still unable to drive full power. SinkPad mistakenly sent me XM-L boards, not MK-R.

User Journal

Journal: Checking Out The New Cree MK-R LED 9

Journal by Khyber

I managed to get some engineering samples of the Cree MKR LED announced on /. a couple weeks or so back.

Needless to say, I was thoroughly impressed. I ran 12V @ 200mA, 500mA, and 1250mA. You know how us geeks/nerds say do not look into laser with remaining eye? You don't HAVE any eyes left with the new MK-R, even at a measly 200mA of direct current at 12V. At 1250 mA, I only needed half a second before the diode packaging began to melt the green PCB, the heatsink was absolutely ineffective. Be prepared to use metal SMD-mount boards for these diodes, because that cheap green board used for most SMD LEDs simply will not cut it. Pushing 15w in a 7mm x 7mm package, you need serious heat dissipation capabilities. Also, DO NOT HOLD THE DIODE WHEN DRIVING AT 200mA! You will blister yourself.

I no longer have my light meter, as I would use that to check flux levels and lumen levels, but needless to say, you drop a quad of very-efficient nearly 4-watt diodes into a single tight array like this, and the world is going to light up before your eyes. I would love to see one of these in a Mag-Lite. At 200mA, 500mA, and 1250mA, you would have all the light you need. Remember that commercial where the kids with the flashlight are using new Energizer batteries and it looks like they're suddenly using a 1,000,000 candlepower light? It's even better than that.

Just one of these would light up my entire 15' x 24' living room, quite possibly more, with wonderful light levels. I could see this sitting in the center of my ceiling. Two bedrooms, two bathrooms, living room, kitchen, dining room. Seven rooms, 105w operating power with all in operation. I could leave my whole home lit up for the same cost of a single 100w incandescent light bulb, which barely lights one room. Oh wait, I forgot the porch and patio. Make that 135w operating power. And in reality, even at 500mA drive current, I think that's more than enough light for an area roughly 25' x 25'. Still beating the pants off of equivalent incandescent or fluorescent. It's the end of them, and I already have the belief that these will truly smash HID systems.

I want single-color versions of these. Cree, if you're reading/listening, you have technology in your hands right now to take over an entire segment of the optoelectronic market. You dropped out of said market (and a couple of others) when you quit making 660nm diodes, and you could really take this class of diode to great heights if you could specialize single-wavelength versions of this.

"Look! There! Evil!.. pure and simple, total evil from the Eighth Dimension!" -- Buckaroo Banzai