angry seouls!

aljsk8

Enlightened
Joined
Sep 7, 2006
Messages
435
Location
Wigan, UK
a while ago i modded a stainless steel civictor with a seoul and it worked great on nimh

i got some 14500 yesterday and now on the high setting after about 2 mins the beam goes blue and reduces output - as soon as a switch off and back on its back to warm white

i dare not leave it on longer than 2 mins incase i cause damage

(the light does get hot to hold so it apears that the heat is getting away from the led)

on the highest setting on nimh there is 330ma going to the led
i cant measure this with the 14500s (as i measured origionally when i was doing the mod) but there is no way its running at close to 1000ma as there just isnt that much light (mabee 100 lumens)

it may be heat that is the issue

can anyone let me know what is going on

if need be i can post my origional mod thread to show what i did
 
yes and they measure 4.15 off the charger

and they dont even get warm when this happens
 
330 mA should not make the color shift unless heat is not leaving the emitter. Tell us more about how you connected the emitter base to the pill. :popcorn:
 
In my case, led was shifting blue:
1. when i reached about 1,2A current
2. when the case temperature gets about 50 degree
 
like i said it was 330 when i first did the mod (nimh) but is obviously more with 14500 (thats why i got them) but i thought we were ok up to 1000ma

im guesing this is around 500ma even if it was 800ma id still think i was safe

but it gets really blue and goes to like 2/3 brightness

heres the mod...

http://www.candlepowerforums.com/vb/showthread.php?t=151669

im not to fussed as the light works great on nimh and even good on medium with 14500s and then the high can just be used for short bursts

but im just confused as to why its happening

Alex
 
Stainless steel is an extremely poor thermal conductor! You're frying the hell out of that poor little LED!

I've been trying to tell people how bad of an idea a stainless steel LED flashlight is, but nobody listens to me....
 
The problem is the poor heat transfer due to the copper washer.
You have to use a massive copper spacer.

Best regards

____
Tom
 
Problem: That's too much epoxy under the emitter. It's acting like a blanket of thermal insulation.
  • You should use a copper disc with more surface area (no hole) if possible.
  • You should do all the isolation under the copper disc, not above it. Preferably by applying the thin and even layer of epoxy to the pill base, and then laying the copper into it and tapping lightly, not hard enough to push it through the epoxy against the metal. Let it cure and check it with an ohmmeter.
  • You should then connect the emitter directly against the copper disc with thermal epoxy or compound. Be sure to wet it out so the emitter is directly against the copper. You may want to bend the leads upwards a little so that they don't lift the emitter back up.
This should be good enough for 500-1000 mA of current depending what epoxy is used and how well it is done. For higher currents, it's best to use an isolated star base, which won't work for this application.
 
Paul, there is no problem to isolate the Led directly at the ermitter with a thin layer of AA-epoxy. I modded a few lights this way and it works very well. In some cases the ermitter sees 1,2 Amps. Also important is the binning of the Led, in my cases (1,2 Amps) I used SSC-P4 U---I bins, no ----H bins.
But the most important part at a Led-light is a propper heatsink, specially at higher driving rates.
I always mention, these tiny lights are not created to drive the Led's extremly hard, because there isn't enough space for a propper heatsink.
aljsk8's heatsink isn't a heatsink, because the copper washer has a hole in the middle.

Best regards

____
Tom
 
I have modded the same light with a P4 and a flupic LED is mounted directly to the heat sink no spacer little off the back of the reflector, flupic runs at least 1 Amp on hi gets quite warm but no angry blue, I prefer to run it on level 7 heaps of light and much cooler, not too sure of the current at that level.
Norm
 
Just keep in mind, It's important to maintain the thinnest possible layer of epoxy between the emitter and the pill, however it is done. But if it's under the disc instead of above it, you'll at least get the heat to the copper for greater total surface area of a directly heated region, instead of leaving it all at the emitter.
 
great replies people - this was my 1st mod so im not suprised i messed it up

(chevro - yes stainless is poor conducter - but stainless is blingest material! - better than ti to me coz of heft and shine)

i was worried about the epoxy as i didnt want to not put enough on - and short the led

i thought the hole would be ok as the epoxy filled it and i thought the heat would still be transfered ok

so my next question is - how do i fix this now its all done - i dont mind getting a new emitter but i dont want to have to sort out a new pill

like i said - i could leave it - its still bright on medium and i never use high for longer than 1min anyway

but id rather it be a "pro" job

any tips?

Alex
 
Last edited:
norm i would love to see some beamshots of your light vs a common light

also was the flupic mod easy to do? did u post a tread at the time?

Alex
 
Alex. I wouldn't leave it.
You can try to pop out the Led (after unsoldering)including the washer with a tiny peak of a knife-blade, screw-driver, thick needle or something else.
Then cut the ermitter with a thin and sharp blade from the washer.

To reinstall, AA-epoxy the Led directly to the heatsink (or use the copper spacer from the Shoppe/build your own but massive). I would prefer not to use a spacer, because the thermal path at this light isn't perfect due to the weak mass. Better a working light with imperfect focus than a perfect focus with an unreliable light.
To isolate the ermitter you can use a thin layer of AA-epoxy at the Led. But only a thin layer. After cure, a second thin layer to fix it at the heatsink.
Soldering the connections and.....that's all.

Best regards

_____
Tom
 
still looking for responses to the above posts but a side note

if i epoxy straight to the pill from emitter how do i test continuity? without installing batt and shorting?

also how do i test if i epoxy copper to pill also?

Alex
 
Alex,

I documented my method here on another flashlight that runs around 750mA or more without turning blue.

Hope that helps...
 
engrpaul

you are a genious

i might have a bash at removing it and starting again

thanks for all the help

Alex

(everyone else keep ideas / tips coming if you have anything to add)
 
Hi Alex
i got angry SSC-p4 too! mine was factory mounted/heatsinked inside the MTE SSC-P4 Flashlight. I did a runtime test on it and the emitter did not like it( i tested the current and it was 0.680 amps ) when i got back to my flashlight (after 12-15 Min. on High) it was puting out a blue beam , i turned it off , but it was too late ,the emitter was damaged, and when i turnd it back on high, it starts to blink. ( real blinking on/off ,not flickering!). I'm waiting for a new sscp-4 star to arrive, so i can fix it and make new and better heatsink (thanks for sharing the heatsink method engrpaul ! ) ,i will mod. the resistor and run the light on one 17670 batterie and keep the amps lower,about 400-450 ma .I hope this will make my angry blue sscp4 into a happy white emitter.
 
Top