I have been in the hobby for a few years now, and have gotten to the point where I am tinkering with my lights, but running into issues with emitter swaps. I am currently stuck with a 50% failure rate. Right now I have on my desk both a DM11 and a T9R that I cannot get any light out of. (and dissembled D4v2 from even further back.) The DM11’s aux lights still work when powered, and I was able to reflash the T9R. For reference the original LEDs and replacements below:
DM11 XP-L HI → FFL351A
T9R W1 → SFT40
I did test the reflowed LEDs with my power supply before installing them, so they have proper board contact. I also tested the T9R (with the power supply) after LED install and got light.
For further reference I have had success when I swapped 519a LEDs into a SP36 and various Reylights.
Have I simply damaged the connections on the driver side somehow? Have I cooked said drivers? Something weirder? If anyone could shed some light on this I would deeply appreciate it.
Always check for a short between positive and host before installing a battery, especially with a reflector light. Just do a continuity check across the two.
Other than catastrophic disaster (overheated MCPCB, the top layer peeled off); the most common reason my LED swaps don’t work is the LED wires shorting against the back of the reflector.
If the ground shorts the reflector, it bypasses the driver and runs direct. If the positive shorts most of the time fries the driver. A continuity check across the two will tell you if the positive is shorted.
I do the following tests after every emitter swap before installing the star into a light:
Gap test - using a small aspheric lens as a magnifying glass I check all edges where the LED meets the star to make sure there are no gaps. I especially want to make sure that the central heat sink pad is bonded all the way across and that the positive and negative pads are bonded properly. I also inspect to see if any solder blobs are bridging across pads.
Short Test - I place the freshly reflowed star onto the anvil of a small vise and then use a DMM set to check for continuity. Specifically, I check to make sure there are no shorts between positive, negative, and ground (the underside of the star).
Function Test - I check to make sure the LED(s) lights up properly. I have an old Radio Shack AA battery holder with a low-power 14500 inside and 2 wires attached. I touch the wires to the positive and negative pads on the star to make sure the LED(s) actually lights up. This only applies to 3v emitters not wired in series.
If any test fails, I redo the reflow. Since I’ve started doing these tests with reflows my success rate has gone up to nearly 100%.
Probably one of the only things I do correctly because I have an old triple nimh pack with 2 “ wires off it that has absolutely no other value in life but works perfectly for this .
Why not just use a multimeter in a diode test mode ? Am I missing something?
Unfortunately it’s uncappable of testing 6V diodes as well as your method.
Or just use a lab power supply with very low current setting. This way I also can check the polarity of the LED, because sometimes there are no datasheets and no markings on the LED for correct polarity on the MCPCB so I have to rely on exerience and a little bit of luck.
I only use my lab power supply after reflowing and it is fine. I have 100 percent success rate for reflowing LEDs on MCPCB, even without checking for gaps or something like that. A failure was long time ago, I had some difficulties with a 219BT-V1 emitter swap (7 LEDs) on a board of the Astrolux MF01 mini due to the board itself (inaccurately manufactured footprints I think, the LEDs were not self-centering and produced constantly shorts after reflow).
Just checked, I do in fact see voltage on the leads, which raises more questions!
Smart! Just checked this and we’re clear on shorts.
I have overheated an mcpcb once before, but I don’t believe this is what has happened here. No reflector installed as of yet since I am/was testing before fully reassembling.
Yep! Ordered a couple from Fireflies directly for this project.
I recommend using a proper constant current power supply, it is just safer for LED and power supply. A lab power supply is very handy for modding anyway, something like Riden RK6006 or similar would be enough for your needs I think.
Well we know the reflector didn’t cause a short in the DM11. Also check the wires and solder connections. If they are touching the DTP layer of the MCPCB then they’ll short and fry the driver.
Just make sure to get the power supply module together with the power supply brick and case. Riden sells the RK6006 and other modules also stand-alone without fitting power supply.
I have done that before on other things and didn’t think to check it this time. Sadly not that though…