DD 3x123 vs 6xAA in 5W

vacuum3d

Enlightened
Joined
Feb 26, 2004
Messages
553
Location
Bay Area, CA
Sorry, I don't have technical data. Please bear with me.

I put together a 5W UWAU in a Mag 2C DD with 3x123. I also put together a 5W UX1U in a Mag 2D DD with 6AA. The 3x123 maintains max brightness til now (~10 minutes of intermittent use). The 6xAA was bright for 15 Seconds and then drastically dropped brightness to the equivalent of a BB500 LuxIII in a MiniMag, and it stayed at that level for 2+ hours and still going. Fresh batteries used in both cases.

My question is, does it sound right to you guys? I thought 3x123 should be about the same as 6xAA in terms of the amount of power being pushed to the led.

thx,
ernest
 

evan9162

Flashlight Enthusiast
Joined
Apr 18, 2002
Messages
2,639
Location
Boise, ID
I can think of some possibilities:

1) The Vf of your UX1U is quite high, and only the first few moments of use of fresh AAs are able to drive it at high current.

2) The Vf of your UX1U was quite low, and the 6AAs drove it really hard for 15 seconds, then something happened, either it partially gave up the ghost due to overheating (too much current), or overheated from a poor thermal interface (improper epoxying or whatnot). In either case, this covers permanent damage that may have happened to your luxeon.

3) You had a weak electrical connection that overheated/worked loose after 15 seconds.



Try another set of fresh batteries. If they exhibit the same behavior, then 1 may be likely (though it's hard to justify under that explanation). If it stays dim, then #2 or #3 are likely. Check for loose connections, cold solder joints, etc.
 

shiftd

Flashlight Enthusiast
Joined
Nov 17, 2002
Messages
2,261
Location
CA
another posibility is that your AA's are so new that they do not have their rated capacity. So they are easily discharged quite fast.
 

evan9162

Flashlight Enthusiast
Joined
Apr 18, 2002
Messages
2,639
Location
Boise, ID
shiftd,

True if he's using rechargables. I was assuming he was using alkalines, but with rechargables, that would make #1 even more of a possibility.
 

shiftd

Flashlight Enthusiast
Joined
Nov 17, 2002
Messages
2,261
Location
CA
evan,
me too was assuming he is using rechargeable. But he did not mention any in his post, so ... /ubbthreads/images/graemlins/blush.gif

i guess you are quite right. if he is using rechargeable, then number 1 is more of a possibility. /ubbthreads/images/graemlins/smile.gif
 

vacuum3d

Enlightened
Joined
Feb 26, 2004
Messages
553
Location
Bay Area, CA
I was using fresh Energizer AA alkaline. I'll try it with something else like Duracell tonight. Anyhow, it sounds like the phenomenon I'm seeing shouldn't be the case.

thx,
ernest
 

shiftd

Flashlight Enthusiast
Joined
Nov 17, 2002
Messages
2,261
Location
CA
hmm, did you try to run exchange the batteries? i mean use the UX1U with 3 123 instead and see what happened?

dunno what conclusion this may make, but try it anyway. maybe it is the lux that is faulty /ubbthreads/images/graemlins/icon3.gif

using hotlips would indicate proper heatsink, if the heatsink is epoxied securely on the mag body. Also, alkaline cannot deliver too much current due to its internal resistance, so thermal runaway is unlikely.
 

evan9162

Flashlight Enthusiast
Joined
Apr 18, 2002
Messages
2,639
Location
Boise, ID
Alkalines can easily deliver 2A DD into a 5W. That means that the 5W is dissipating more than 15W. The junction temperature in that case is probably over 160C (max is 135C). Thermal runaway can happen rather quickly. I've experimentally determined that the fuse current for the bond wires on a luxeon is around the 3A range, a current that AA alkalines are capable of delivering.
 

shiftd

Flashlight Enthusiast
Joined
Nov 17, 2002
Messages
2,261
Location
CA
/ubbthreads/images/graemlins/blush.gif

evan, i thought AA alkalines have too much internal resistance that it cannot deliver near 2 amp current. I know D cell can deliver much more than 2 amp, but alkaline AA is not supposed to be able to.
/ubbthreads/images/graemlins/icon3.gif
 

evan9162

Flashlight Enthusiast
Joined
Apr 18, 2002
Messages
2,639
Location
Boise, ID
I measured a single AA alkaline as being able to pump 10.6A in a short circuit through my 20A current meter (0.07 ohms resistance).
For reference, a D alkaline pushed 12.6A, a AA NiMH, 13.5A, and two AA NiMH in series, 17.5A

I used 6AAs and 0.5 ohms of resistance with a U2U 5W for a friend - fresh AAs pushed 1.5A through that one. Given a lower Vf luxeon, 2A is quite easily done.
 

Hobo

Newly Enlightened
Joined
Apr 13, 2004
Messages
29
evan9162, what would it be with 3 CR123 batteries? He said that he's run a similar setup off of three of them without apparent trouble. I would think that internal resistance on a CR123 would be less than 2 AA batteries.
 

evan9162

Flashlight Enthusiast
Joined
Apr 18, 2002
Messages
2,639
Location
Boise, ID
From reading datasheets (but not testing), I've observed that the internal resistance of a Li 123A cells is about the same as 2AA alkalines in series. There was a thread quite some time ago where the internal resistance of 123 cells was measured experimentally and found to be about 0.25ohms/cell.

The advantage the 123As have over alkalines is a flatter discharge curve. So the terminal voltage of the cell will be higher at half discharge than 2AAs would be.
 

vacuum3d

Enlightened
Joined
Feb 26, 2004
Messages
553
Location
Bay Area, CA
evan, it looks like your point #2 in your first post is right on. I tried to use 2300mah NiMh AA, no improvement. I then put the 3x123 that I used in the other working flashlight into the dim one, it made no improvement either. Put the 3x123 back to the good light, bright as sh*t. $30 down the drain ;(
When I examined the led closely, all I could see is about 1/5th of the die has a slight discoloration. I could hardly notice the difference, but that was enough. Damage's done. Shoot.

The led is actually a UX1V, not a UX1U I mentioned in my first post, which is even more strange to me since Vf rated even higher. Why would I burn the UX1V but not the UWAU???


Anyhow, thanks for all your help guys.

ernest
 

shiftd

Flashlight Enthusiast
Joined
Nov 17, 2002
Messages
2,261
Location
CA
/ubbthreads/images/graemlins/blush.gif

care to take pic on the 5W?
maybe when you first run it to see if it turns on or not, you forgot to heatsink it properly /ubbthreads/images/graemlins/icon3.gif
 

evan9162

Flashlight Enthusiast
Joined
Apr 18, 2002
Messages
2,639
Location
Boise, ID
[ QUOTE ]
vacuum3d said:
evan, it looks like your point #2 in your first post is right on. I tried to use 2300mah NiMh AA, no improvement. I then put the 3x123 that I used in the other working flashlight into the dim one, it made no improvement either. Put the 3x123 back to the good light, bright as sh*t. $30 down the drain ;(
When I examined the led closely, all I could see is about 1/5th of the die has a slight discoloration. I could hardly notice the difference, but that was enough. Damage's done. Shoot.

The led is actually a UX1V, not a UX1U I mentioned in my first post, which is even more strange to me since Vf rated even higher. Why would I burn the UX1V but not the UWAU???


Anyhow, thanks for all your help guys.

ernest

[/ QUOTE ]


Sometimes the Vf binning is off (hell, sometimes all of the binning is just off). I received a VV0U that is clearly a VV0T (the Vf was about 6.7V @ 700mA). It's wierd that I've seen Vf binning be off so many times with the 5W, but I bought 50 1Ws where the Vf was dead-on (+- 0.05V) for 90% of them. I think the Vf would have to be WAY off for this to happen.

Frying from heat in 15 seconds makes it sound like there wasn't a good thermal interface between the LED and heatsink. How does the joint between the luxeon and heatsink look?

Did you check all solder joints/electrical paths just to make sure? A final check would be to use a 9V battery and connect it directly to the leads of the luxeon, bypassing solder joints, the switch, internal contacts, etc.
 
Top