LED Resistor Calculations ...

[eccerr0r]

Re: LED Resistor Calculations ...

Aluminum PCB just means the backing is aluminum, of course this will not function as a regular PCB, so there's a layer of dielectric on the aluminum. Then there's another layer of copper over the dielectric which gets etched. Heat transfer still needs to go through the thin dielectric layer, so it's not as good as being soldered straight to the aluminum but it's better than FR-4.

[EasyGoing1]

Re: LED Resistor Calculations ...

Of course it conducts, it;s bare aluminium. You don't connect anything to it. The centre contact is the anode.
https://www.pcbway.com/pcb_prototype...minum_PCB.html

So then how would you connect the LED to the aluminum backing for heat dissipation? Or is it just having the aluminum close to the LEDs that it will pick up the heat they are kicking out and transfer it away?

[diif]

Re: LED Resistor Calculations ...

Of course it conducts, it;s bare aluminium. You don't connect anything to it. The centre contact is the anode.
https://www.pcbway.com/pcb_prototype...minum_PCB.html

[EasyGoing1]

Re: LED Resistor Calculations ...

Your panel should really be Aluminium not FR4.

OK, what I wanna know is ... if one uses an Aluminum backed PCB, aluminum can still conduct electricity ... unless they dope it with something that prevents that? I'm a little confused on how that works if you were to connect those center pins to an aluminum backing...

[EasyGoing1]

Re: LED Resistor Calculations ...

The footprint model should have marked that node to be a DNC node so it would show up as a DRC or LVS error.

If you're talking about the footprint that is used for the PCB design layout, I completely agree... that center pad should be electrically equivalent to the anode so that someone (like me) doesn't think it's available for a separate network in the schematic.

[EasyGoing1]

Re: LED Resistor Calculations ...

It's a little crude as I did it in paint.https://www.aliexpress.com/item/4000932468693.html

ummmm ... yeah ...

Unless you're in the LED manufacturing industry and are completely aware that those shitty drawings indicate where the connections are ... ain't no one on the planet going to see it ... whoever thought that was good idea, should be demoted.

[eccerr0r]

Re: LED Resistor Calculations ...

That's a very poor indication that they're connected, it could also imply the plastic fill too. If it didn't draw a line through the tabs it'd be more convincing but then it'd be wrong because you can't connect to the plastic.

The footprint model should have marked that node to be a DNC node so it would show up as a DRC or LVS error.

[diif]

Re: LED Resistor Calculations ...

It's a little crude as I did it in paint.

Aliexpress is awash in cheap LED COBs. https://www.aliexpress.com/item/4000932468693.html

[EasyGoing1]

Re: LED Resistor Calculations ...

The side view drawing shows the right pad and lower pad as the same physical piece.

They could have stated it specifically, but dealing with vague datasheets is another joy of design! If in doubt - test!

ok wait ... you're talking about THIS ... RIGHT HERE???

THIS DRAWING ...



Cause there isn't a court in any country on the planet who would acquit them for that shit right there! Hieroglyphics are clearer and more specific ... I mean I'm staring at the drawing TRYING to find some connection that I could in some way see where the two pins are connected ... but man ... I just cant make that connection with whats on that paper right there.

[Agent24]

Re: LED Resistor Calculations ...

Is it now .... well, Here's Page 2 ... kindly tell me which words indicate that the center pad and the anode are electrically equivalent...

The side view drawing shows the right pad and lower pad as the same physical piece.

They could have stated it specifically, but dealing with vague datasheets is another joy of design! If in doubt - test!

[EasyGoing1]

Re: LED Resistor Calculations ...

Yes it comes with a bridge rectifier, heatsink, and diffuser. It gets quite hot too but it's expected.

But 800 lumens ... for a dollar? That's just bliss ... where did you score that?

[eccerr0r]

Re: LED Resistor Calculations ...

Bridge Rectifier and large caps?

Yes it comes with a bridge rectifier, heatsink, and diffuser. It gets quite hot too but it's expected.

[EasyGoing1]

Re: LED Resistor Calculations ...

Yeah I have a couple of those round ones whose pad seems to not be connected. I've not tried sending 1W into these either, so not sure how much heatsinking is needed.

At this point I wonder... I can buy $1 800lm LED modules, just need to figure out how to feed 120V into it.

I just pushed one of those bigger ones at 280ma and according to my Fluke, that center pad was 70C ... YIKES! that's friggin hot.

[EasyGoing1]

Re: LED Resistor Calculations ...

Yeah I have a couple of those round ones whose pad seems to not be connected. I've not tried sending 1W into these either, so not sure how much heatsinking is needed.

At this point I wonder... I can buy $1 800lm LED modules, just need to figure out how to feed 120V into it.

Bridge Rectifier and large caps?

[eccerr0r]

Re: LED Resistor Calculations ...

Yeah I have a couple of those round ones whose pad seems to not be connected. I've not tried sending 1W into these either, so not sure how much heatsinking is needed.

At this point I wonder... I can buy $1 800lm LED modules, just need to figure out how to feed 120V into it.

[EasyGoing1]

Re: LED Resistor Calculations ...

I have a bag of these LEDS as well ... and I just checked and the center pad is NOT connected to either the anode nor the cathode ... strictly heat dissipation... albeit its a much larger LED and a hell of a lot brighter too.

[EasyGoing1]

Re: LED Resistor Calculations ...

Crap datasheet should have warned that the center tab must be kept isolated (unless it's in chinese). I would have to assume that, at least check, since it's not bonded to one of the main terminals.

I guess it's annoying that DRC and LVS don't technically pass on this board, which is easy to slip through the cracks when the models are wrong and you're only using software checks... Oh well, live and learn?

And BTW yes technically liquid nitrogen could help dissipation ... up until the thermal resistance and this will limit the watts you can dump into them. Usually this theoretical number is written in the datasheet too, but tends to be to typical ambient temperature like 25°C, which has no correlation to boiling nitrogen.

I ran this panel for the last 12 hours on 18 volts ... drawing only 380ma and the temp of the panel has been steady at around 36C ... and it still kicks out a ton of light ... not the blistering sun I was hoping for ... but certainly very useful ... if it had a diffuser and was hanging in the center of the room at the ceiling ... it would certainly be all that a room would need for light ... and only pulling 6.8 watts ... that's not bad ... considering back in the day, it was common place to have 60 - 100 watt bulbs in every fixture.

[EasyGoing1]

Re: LED Resistor Calculations ...

It's indicated on page 2.

Is it now .... well, Here's Page 2 ... kindly tell me which words indicate that the center pad and the anode are electrically equivalent...

[diif]

Re: LED Resistor Calculations ...

OK, now where does it say on that datasheet that the center pin is equivalent to the anode or the cathode?

It's indicated on page 2.

[eccerr0r]

Re: LED Resistor Calculations ...

Crap datasheet should have warned that the center tab must be kept isolated (unless it's in chinese). I would have to assume that, at least check, since it's not bonded to one of the main terminals.

I guess it's annoying that DRC and LVS don't technically pass on this board, which is easy to slip through the cracks when the models are wrong and you're only using software checks... Oh well, live and learn?

And BTW yes technically liquid nitrogen could help dissipation ... up until the thermal resistance and this will limit the watts you can dump into them. Usually this theoretical number is written in the datasheet too, but tends to be to typical ambient temperature like 25°C, which has no correlation to boiling nitrogen.