Looking forward to seeing what you come up with.
Here is a little quick idea for a “night light”…
24awg solid wire depicted on an XP-E2 which can reach up to about 150 lumens using a 380ma AMC7135.
An XP-L should perform similarly and uses the same footprint.
The tab on the 7135 can be used for additional heat sinking.
DavidEF, connect the wires to the LED, hold the LED between your fingers and slowly increase the power. When you can’t hold on anymore, that’s about how much current you should power them with. WRT connecting one fat wire between one input and the thermal path, if your wires are insulated, would it make sense?
Thanks, that is very helpful! With that much going to the LED, it’s over 1/3 of its rated capacity. That would be pretty good, considering the lack of good thermal management! But, would the LED and the 7135 being back-to-back create a heat trap between them? Both of them dissipate heat through the back, don’t they? I guess I could try it and see.
Yeah, that makes good sense. So, if it’s just a bit hotter than I can stand to hold on to, then it’s right at the point where it shouldn’t be run any hotter, is that the idea?
If you do the back to back idea(nifty by the way) then one wire goes to the led+/Vdd pair and the other to the ground tab with the led- connection made to the Vout pin.
I was just thinking autonomy per LED by adding either 350 or 380ma 7135 to each one. These little things can tolerate so much heat that there shouldn’t be any issues with both mounted back to back. Did you not say you were going to do this in an array?
I was thinking that the 4.2V will be limited as the batteries normalize to 3.6V fairly quickly. In an array of 10 pieces, that would be 3.8/3.5 amps which is a pretty good load on any conventional Li-Ion cell.
I wish I had some spare XP-G’s to test this on. However, hot as in the limits for the LED (300 degrees F) and solder (450 degrees F) is very hot. Boiling water is hot at 212 degrees F. I estimate my XM-L to reach maybe 125 degrees F judging by putting your finger on hot asphalt on a bright hot sunny day at around 100 degrees F. Of course, ambient temps are a factor. Temperature differential is normally rated at degrees above ambient… so in a 70 degree F room, the 125 degree F would be equivalent to a 55 degree F rise above ambient.
Yep. Freeze the vid at 12 seconds. You will see I marked the AMC7135 with the pin definitions.
You could also solder a heavier wire to the 7135 tab for even more heat-sinking. This could even become structural if desired.
Part of this idea is to kick up the output of the little cheapo blinky bicycle tail lights. This can easily replace those little led’s directly and letting the gate be triggered by the light’s built in modes. Since they are normally driven by 3V anyway, it is a perfect marriage.
I estimate the ‘ouch this s&$# is hot!’ point to be from 60-65 deg C. A little more should be ok. Anyone else have temp based recommendations ?