Building New Panels - Sharing the Process

[rrog]

Mine is 22" so I have more LED per square foot, so that's why I thought I'll run a little hotter. I have two 6" axial fans going in.

I love that sink.

Rives, when I think of sinks, I now think of you, so how about a question. The LEDs are all on the soldered and the driver wires are twisted together with wire nuts so there's a complete circuit with each string. This is now safe to move about ungrounded? So far this has only been on the grounded mat with a grounding clamp also on the sink.


I have the various control box components in hand, Fuses, MOVs. I feel like the dope looking at a ticking bomb with 15 different colored wires in a bad b-movie.

However: I did receive my multimeter yesterday. I'll need to see how the hell it works and what setting to test these LED strings.

 

[luvdemshrooms]

Not many more. I'm running 1 LED per 5sq inch, you're running 1 LED per 4.6sq inch. With 2 fans running on it I bet your temps will be in the mid to low 40's C. Even if it squeaks up more toward 50C you still get a long and healthy life span from those XML's.

Current LED's are much less prone to ESD damage than earlier ones.

Testing LED's with a meter is simple (once you've done it). There is a diode check position on the meter dial. Looks like this ->|- The arrow points in the direction of current flow. My Fluke meters use a 9v battery that will light up individual LED's quite nicely.If your meter runs on batteries that are higher in total voltage than the LED's require... yours should do the same. Red lead to the positive terminal, black to the negative. You will NOT be able to light the entire string with a meter.

Maybe this will help:

http://www.youtube.com/watch?v=1FYir4uJYhw

 

[rrog]

Thanks- I was previously using the little LED battery tester you see everyone use to test individual LEDs. So now I can do the same with the multimeter, but neither light the string.

I imagine there's a way to test the string? Not light them but just verify correct wiring and solder connections?

 

[luvdemshrooms]

The way I do it is this:

Visually inspect LED's for orientation and solder joints.
Set your meter on the diode check position.



Then, I touch my probes to the LED's adjacent to the one I want to light up.



This test's the LED I'm interested in AND the solder joints.

The only way I know to test the entire string at once is to light it up. Either with your driver, or with a current limiting power supply. Set the current to a low number (say 50mA) and the voltage to zero. Connect your leads and slowly turn up the voltage. When the voltage approaches the sum of the forward current of all the LED's in that string, they will begin to glow.

If you have 20 XML's in series, that will require almost 60 volts to light the string. In practice though the voltage needed will be a touch lower than the math would indicate.

As most people do not have current limiting power supplies in their homes, the multimeter method is the most common. On a positive note, it also is the method LEAST likely to pop a LED.

 

[rrog]

Thanks! I have 10 XM-L per string in series. I see how you're checking individual LEDs. So what does this multimeter do for me other than shoot a little juice into one LED at a time?

 

[luvdemshrooms]

Almost forgot. Too much solder on the pads next to the screw heads can result in a short and your string will not light.

Some use fiber or nylon washers under the screw heads. Some only solder to pads not adjacent to the screw heads. I choose to be careful with the amount of solder I use.

To test for shorts here... set your meter to the Ω position (note that some meters have an audible function for this). Put one probe (doesn't matter which) onto the heat sink and with the other probe go to each and every screw and test. If your meter has an audible function, a short will beep. If it does not have an audible function, any reading (other than the one you see when your probes are not touching each other) will indicate a short.

 

[luvdemshrooms]

So what does this multimeter do for me other than shoot a little juice into one LED at a time?

You can use it to test the output current and voltage of your drivers.

Blown fuses. Electrical outlets. Batteries. Incandescent bulbs.

There are many uses for the meter and everyone should have at least one.

 

[rrog]

Sounds groovy! Rives had me use plastic washers under all my screw heads.

Rives- That whole recommendation regarding the drill bit size, exact screw and washer all worked perfectly. Everything was just right. Working in that 3/8" thick aluminum deck was a treat. I Plan to make 5 more for the new room, now that I understand what the hell I'm doing (so far).

 

[rives]

Damn, you guys got rolling early this morning! Glad to see things coming together for you Rrog. Once you start using that meter, you will find more and more uses for it. Regarding your question on lighting the entire string, I also use a bench-model current-limiting power supply.

 

[rrog]

Well, I've decided I hate thermal grease. What a freakin mess. Of course I used too much, so it's popping out from under the stars. Please tell me this stuff isn't electrically conductive...

EDIT:

I'm reading that this is electrically conductive... So I've ordered up another tube as well as screws and washers. I'll take the stars off and clean them if needed.

Does this stuff drip when things get warm?

 

[luvdemshrooms]

It is conductive, well... some are. It's one of the BIG mistakes people make when installing their new CPU coolers.

A thin film is all you need. We're talking a few thousandths of an inch thick. About the thickness of a piece of notebook paper. Too thick and it impedes the transfer of heat.

Did I mention "thin"?

If you put that much on I'd recommend removing the stars and reapplying the grease.

Did I mention to put on only a "thin" film?

 

[rrog]

I'm not sure if this should be a thick or thin coat...

I'm using Arctic Silver 5. I have a bunch of cleaning to do... I screwed them down tight and the layer is quite thin. It's just that it's squeezed out and ready to short the string. This part is a real liability if you're not careful.

 

[luvdemshrooms]

I'm not sure if this should be a thick or thin coat...

I was afraid I didn't say "thin" enough times.



I feel your pain.

 

[rrog]

Seems to wash off readily with 99% Iso

 

[rives]

THIN!! That shit makes about as much of a mess as Petrox, NoAlox, etc. Every damn thing that you touch comes up gray. As LDM said, some are conductive. Sounds like a perfect experiment for the new meter! And yes, some Q-tips and Iso are probably the ticket.

 

[luvdemshrooms]

EDIT:

I'm reading that this is electrically conductive... So I've ordered up another tube as well as screws and washers. I'll take the stars off and clean them if needed.

Does this stuff drip when things get warm?

When you apply the grease, just put a "small" dab-o-grease (about the size of a BB. .125in or 3mm) under the center of the star. Place the star where you want it and while applying pressure swirl the star. Lift the first one or two to inspect. You should be able to easily see through the grease and it should have covered most of the stars bottom.

Thermal grease is intended to fill voids in the surfaces. The thicker the layer... the worse the thermal transfer. If you could make the star bottom and the heat sink PERFECTLY smooth, you would use no grease. Alas... nothing is perfect.

I doubt Arctic Silver will drip.

 

[rrog]

Will do. I should have asked more questions before doing this step.

Rives- Have you ever tried to remove a screw from a hole where a screw head sheered off? I torqued the head off a screw. Filed it down and will re-drill a new hole when more screws arrive. Hopefully as I remove all these stars to re-grease I will not shear any more off.

 

[rives]

Will do. I should have asked more questions before doing this step.

Rives- Have you ever tried to remove a screw from a hole where a screw head sheered off? I torqued the head off a screw. Filed it down and will re-drill a new hole when more screws arrive. Hopefully as I remove all these stars to re-grease I will not shear any more off.

Yes I have, and didn't have any luck. Those are far too small to get an easy-out into, freon wouldn't shrink it enough, etc. I just drilled new ones. I think that I induced my own problem (as usual!) by using the same screw repeatedly to self-tap the threads into the aluminum. Too much cumulative torque for the little guys.

 

[rrog]

I've set my drill to a low torque.

I cleaned and re-set some stars today. After cleaning, I applied a tiny drop. It's crazy how such a small drop will still squeeze out some on the edge. I've really dialed down the amount I use

 

[luvdemshrooms]

I torqued the head off a screw.

I hate when that happens. When inserting the screws just make them lightly snug. They don't need to be very tight.

Filed it down and will re-drill a new hole when more screws arrive.

That's about all you can do with those tiny screws.

Hopefully as I remove all these stars to re-grease I will not shear any more off.

Good luck to you.

[FONT=Arial, Helvetica, sans-serif]It's crazy how such a small drop will still squeeze out some on the edge. I've really dialed down the amount I use [/FONT]

Just like Brill-Cream... a little dab will do you.