I'm working on a project, and I connected 6 SSC P4 stars in series. HOWEVER, when I connect power to the loop, the first 2 leds do not light up, the 3rd one flickers a little and hte last 3 come on medium brightness. When I bypass the first 2 leds, the last 4 come on strong. I took a multimeter and it says the first 3 leds are direct shorts. But, when i hooked each led up to a 3V source, the 1st one flickers one everytime you touch the power, but goes out. The 2nd onward seem to come on fine. I'm very confused about what is going on here. Are 3 of them fried? All I did was connect each led up to a 18650 battery momentarily to check it was good. How could that have hurt the first 3 leds, but not the last 3?! Why would it read the 2nd led as a direct short, but it still seems to work? The only thing I can think of is that I damaged the first 3 somehow with the 18650, or that pack was bad. I got 2 packs of 3 each and the first 3 were from the same pack. But this too seems unlikely because i bought them from the same source, so i'm sure all 6 were in line on the reel and just cut into segments of 3 to fit the box better. I'm very confused.
Are these fried?
- Thread starter havand
- Start date
Help Support Candle Power Flashlight Forum
the heat sink slugs on the P4 are not isolated - they are connected to the positive power lead. if you shorted two or more slugs together by them being attached to a common heat sink, then it will do exactly as you describe.
Here's how to check whats going on.
Set your multimeter to the continuity setting ( or the lowest resistance setting)
place one lead on the + power lead of one of the LEDs.
place the other lead on the heatsink
If you get continuity (either your DMM beeps, shows 0V, or shows 0 ohms), then that particular LED is NOT isolated from the heat sink. Repeat for all LEDs. If any of your LEDs are not isolated, then you have a problem, and will not be able to run them in series. You need to re-attach them, ensuring that they do remain isolated.
Here's how to check whats going on.
Set your multimeter to the continuity setting ( or the lowest resistance setting)
place one lead on the + power lead of one of the LEDs.
place the other lead on the heatsink
If you get continuity (either your DMM beeps, shows 0V, or shows 0 ohms), then that particular LED is NOT isolated from the heat sink. Repeat for all LEDs. If any of your LEDs are not isolated, then you have a problem, and will not be able to run them in series. You need to re-attach them, ensuring that they do remain isolated.
Ok, I took a multimeter and did just as you described. Took a lead, put it on the + of the led, the other on the heatsink (not the star). Nothing beeped. Even more perplexing is I was playing around with the benchpower supply some more and that first led, when I apply power, flashes, then goes dim (lit, but dim). If i wiggle the contacts around more, it sometimes lights up seemingly full.
I tried using the continuity setting before and jumped the + and - legs of each led itself. The 3 that light either dimly lit or if reversed read ~0.745. The 3 problem children, 2 beeped once, then read a resistance? while the other just continuously beeps. :/ I'm thinking the first 3 are toast.
I tried using the continuity setting before and jumped the + and - legs of each led itself. The 3 that light either dimly lit or if reversed read ~0.745. The 3 problem children, 2 beeped once, then read a resistance? while the other just continuously beeps. :/ I'm thinking the first 3 are toast.
You're probably right that the 3 misbehaving P4s are permanently damaged in some way.
If you're using stars, it's likely everything is isolated properly, so it doesn't sound like this is an isolation problem.
It sounds like your DMM's continuity is also a diode test function. When the LED glows, it means that the LED is forward biased (like when it's lit up), which is why it glows. When you reverse it, you are likely forward biasing the ESD protection device in the LED, which is why you get a reading on your DMM of 0.745 (that's 0.745V, a "normal" voltage for a forward biased diode).
The problem LEDs sound like they failed as a short or intermittent short. I've seen flickering when I've applied too much current to the device (internal damange to the die). Failing as a short is also a failure mode of LEDs that have seen too much current/heat.
Where did you get the stars? Some cheap Seoul P4 stars have mounting issues where the heat sink of the emitter isn't even touching the star, so the LED overheats rather rapidly. That's one possibility.
How are you powering these? If your power supply isn't current regulated, they could have been exposed to too much current.
One last thing to consider is physical damage. The dome of the Seoul P4 is rather delicate, and the bond wires are precariously positioned within the dome to get damaged easily if the dome is pressed/squeezed/deformed, plus there are 4 bond wires, making the chances of damage even higher.
We'd need some very close up pictures of the dead devices to acertain what happened to them and why they failed.
If you're using stars, it's likely everything is isolated properly, so it doesn't sound like this is an isolation problem.
It sounds like your DMM's continuity is also a diode test function. When the LED glows, it means that the LED is forward biased (like when it's lit up), which is why it glows. When you reverse it, you are likely forward biasing the ESD protection device in the LED, which is why you get a reading on your DMM of 0.745 (that's 0.745V, a "normal" voltage for a forward biased diode).
The problem LEDs sound like they failed as a short or intermittent short. I've seen flickering when I've applied too much current to the device (internal damange to the die). Failing as a short is also a failure mode of LEDs that have seen too much current/heat.
Where did you get the stars? Some cheap Seoul P4 stars have mounting issues where the heat sink of the emitter isn't even touching the star, so the LED overheats rather rapidly. That's one possibility.
How are you powering these? If your power supply isn't current regulated, they could have been exposed to too much current.
One last thing to consider is physical damage. The dome of the Seoul P4 is rather delicate, and the bond wires are precariously positioned within the dome to get damaged easily if the dome is pressed/squeezed/deformed, plus there are 4 bond wires, making the chances of damage even higher.
We'd need some very close up pictures of the dead devices to acertain what happened to them and why they failed.
I got them from fred, so they're probably mounted just fine. Most likely for whatever reason testing them with the 18650 did the damage 
That stinks. $30 gone. Weird though, i've always tested my leds with a split second fire from a single li-ion. Never had any damage before.
That stinks. $30 gone. Weird though, i've always tested my leds with a split second fire from a single li-ion. Never had any damage before.
I got them from fred, so they're probably mounted just fine. Most likely for whatever reason testing them with the 18650 did the damage
A hot off the charger 18650 will supply to much current to a low VF LED like the Seoul or Cree.
Indeed. You could be hitting them with well over 2A, which is definitely in the range that could cause immediate damage.
If you just want to see if they light, or figure out polarity or whatever, then use a lower voltage power source (like 2AA batteries), or even a lithium coin cell. With those power sources, you won't be at any risk of damaging the parts.
If you just want to see if they light, or figure out polarity or whatever, then use a lower voltage power source (like 2AA batteries), or even a lithium coin cell. With those power sources, you won't be at any risk of damaging the parts.
I've got some older CREE XR-E emitters at home. I can reorder 3 new leds or try to figure out how to mount the emitters to my aluminum heatsink. Tough call. Probably spend the extra $30 and get the SSC stars again. When i've already dropped like $250 on the project, what's $30 more, right?
We've all learned lessons the hard way.
Back in 2001, when Luxeon star/O (with optics) were $18 each, I fried one the day I got it when I shorted out the current limiting resistor. It was being powered by 4AA NiMH batteries, which ended up direct driving it for about 0.5 seconds, then it went dark permanently.
A few months before that, I wanted to remove a cyan Luxeon emitter from the star board. I unsoldered the leads, then tried to pry it up from the star. The body and dome readily popped off, leaving the heat sink slug, die, and two tiny ripped bond wires stuck to the star. That was another $12 right there. I later learned the proper way to remove an emitter from a star board.
Accidentally destroying things is part of the hobby - however, if you learn something from it, it's not a complete loss, just an expensive lesson.
Back in 2001, when Luxeon star/O (with optics) were $18 each, I fried one the day I got it when I shorted out the current limiting resistor. It was being powered by 4AA NiMH batteries, which ended up direct driving it for about 0.5 seconds, then it went dark permanently.
A few months before that, I wanted to remove a cyan Luxeon emitter from the star board. I unsoldered the leads, then tried to pry it up from the star. The body and dome readily popped off, leaving the heat sink slug, die, and two tiny ripped bond wires stuck to the star. That was another $12 right there. I later learned the proper way to remove an emitter from a star board.
Accidentally destroying things is part of the hobby - however, if you learn something from it, it's not a complete loss, just an expensive lesson.
Last edited:
True. I think I may actually be ok. I may be able to remove those bad leds and replace them with 3 SSC emitters i forgot I had at home. The ones I wrecked wered USVOH. The extras I have are USWOH. So, slight color difference. Why not use the other ones if I have them though, right? Thanks for the input, I appreciate it.
The purpose of the slug is to drain heat to the heat sink. Heat transfer epoxy or alternative shall be used for that: see Luxeon star mounting for reference. The slug is connected to the LED terminal (just do not remember which, but tester will show for sure).
In the vicinity of all this leaving slug floating is OK. As for heat dissipation methinks for 1W LED PCB copper plane would not be enough. Maybe OK at lower power levels, but overheating s fatal for LED's.
RoGuE_StreaK
Enlightened
Holy threadmine Batman!
