Btw, make sure you clean the flux residue. At higher current the residue will thin and wet the LEDs sides. When things gets hotter, it will cause phosphor burns.
Boiling IPA and nylon brush is all it takes
Isopropyl Alcohol. Usually available at the drug counter in 70% - 95. Higher concentration doesn’t seems to work any better. IPA needs water to dissolve fats such as solder grease/flux residue. I bought 95 just to save money. I added 50% filtered distilled water taken from my air conditioner condensation (It’s very humid here at 88-90%, I usually got ~20 liter distilled water per day).
Nice, I’m planning the same on my wizard, but with a pretty wild combo of emitters
2x 3000k + 5000k + 6500k. I’m curious how the resulting tint will look. Should be around 4500k I guess.
New VR16SP4 (alpha board) thermal test result from Jensen567:
At the same 5A the new board is 26C cooler. Peaked in ~6A (the resolution used was 500mA). Using my previous lux-current test result, the peak should be ~6,4A (3,1A/LED). As you can see from the phosphor temp result, the BLF max rating is 4,8-5 A in 2s2p config (2,4A/LED).
So if I ordered a VR16SP4 WITH 4 E21 SM403’’ and put them in a Convoy S2+ or S3 with a MTN 17DDm would I want to MCPCB set up as 3V, 6V, 9V, or 12V? I’m really curious to see how these would perform in a Convoy S2+ or S3 host with a high drain 18650.
I supposed the new 3, 6, 12V order received is yours. Only heavy OP reflector and pebbled/frosted/faceted TIR optics will work with the new quadtrix board. The extra gap made for maximum performance creates donut with SMO/plain tight TIR optics.
No wasn’t me, I don’t want to order until I figure out how I should set the LED’s up. Would it be possible to run these on a 8x 7135 driver at 3 volts or 6 volts?
I can’t comment on this since I never use/try any 7135 based drivers. As long as the drivers of your choice designed for 3/6/9/12V LED config it’s good to go. The max safe current limit (BLF max) for each configurations (using 4 x E21A) are: 10A(5,6A) for 3V, 5A(2,8A) for 6V, and 2,5A(1,4A) for both 9V / 12V. Lower values in brackets are OEM L70 50.000 hours absolute max.
However FET fot that configuration is not the best choice. I tried quad E21A with Lexel’s bistro-hd driver and it ended very fast, I saw 16A on clamp meter, then one of E21A died…
It itches me to find a pair of jumpers to connect the pairs of minus soldering pads of a quad board. With the jumpers closed all 4 LEDs are powered with jumpers open - just 2 LEDs. In my new 2x 5000K + 2x 6500K config it’d give me an easy access to 6500K, 6200K, and 5700K output. Just screw out the bezel, pop in/out the jumper(s) and put the bezel back.
I noticed today the VR16S1 I’ve been testing for a few weeks is ever so slightly too large for the cutout in my Wizard Pro, which meant it hadn’t been making any contact with the body of the light. To get it to contact the surface I had to file a tiny bit off all the edges of the MCPCB and carefully deburr everything. It’s now seated properly and making good contact. Performance isn’t a whole lot different, at least on 18350, which I assume is because the 6V emitter is very underdriven. I will have to make similar adjustments to the VR16SP4.
Posting this just to prove it actually worked before I put it out of its misery
This was mostly just very rough job to see how easy the VR16SP4 is to modify for the Wizard. Definitely not something you want to do with a populated board you care about, since the phosphor coating is just as fragile as reported. Now of course I need to order more boards…