Low Vf 5W (bin ?3R) extrem overdrive (with data)

Yesterday I replaced the U4T emitter of my Lambda Megalite to an unknown BIN 5W LS emitter. This was my very first little 'mod'.
Removed the old emitter with a sharp pocket knife and a hammer and epoxied the new one with Artic Alumina on Lambda's heatsink.
I measured the forward voltage and current. To be able to measure I removed the reflector part, and the body was lying on my desk, so heat transfer was not ideal.
I measured the current by the voltage drop on a 0,1 ohm 5W rated resistor.

Initial current was over 2 Ampers! I started to think that this must be a very low Vf 5watter.

Results #1 (6x Fresh Sanyo 2100 NiMH cells)
-------------------------------------------
Time Vf[V] I[A] V_battery_unloaded/Remark
00:00 6,57 2,01 8,44 (1,406V/cell)
01:00 6,55 1,80
02:00 6,53 1,77
03:00 6,51 1,71
04:00 6,50 1,658
05:00 6,49 1,635
06:00 6,48 1,61
07:00 6,48 1,598
08:00 6,48 1,591
09:00 6,47 1,58
10:00 6,47 1,581
11:00 6,47 1,59
12:00 6,47 1,599
13:00 6,48 1,603 Body getting hot
14:00 6,49 1,605
15:00 6,48 1,615
16:00 6,48 1,616
17:00 6,48 1,618
18:00 6,48 1,616
19:00 6,48 1,619 Very hot body
20:00 6,48 1,615 7,73 Volts (1,28V/cell)

Here I stopped experimenting. I had to cool it down and deploy the rechargables or else I would never reach 700 mA to have the approximate Vf.

Results #2 (6x somewhat used Sanyo 2100 NiMH cells)
---------------------------------------------------
Time Vf[V] I[A] V_battery_unloaded/Remark
00:00 6,41 1,24 7,62 (1,27V/cell)
05:00 6,30 1,09
10:00 6,27 1,07
20:00 6,23 0,98 Battery goes down. Soon we get 700mA
27:00 6,17 0,85
28:00 6,16 0,82
29:00 6,15 0,78
30:00 6,13 0,73
30:30 6,12 0,71

Hmm. 700mA, Vf=6,12V, perhaps should be modified with the thermal factor, but seems to be an ?3R binned emitter (5,91-6,39V)

Computed internal resistance of the battery + holder + switch is 0.7-0.85 ohms. I think this is dependant of the fill-state and temperature of the rechargables.

Perhaps 5xNiMH cells could be enough for this emitter.
Maybe I am tending towards regulation vs direct drive.

Perhaps this low Vf one would be good for a smaller flashlight where runtime is needed, as a boost circuit won't have to pull so much juice out of the battery. Was the original luxeon damaged in removal?

Pinter,

Wow! That's some very interesting data. I guess I better take a closer look at those no-bin Luxeons. I had no idea the VF was so low.

They could be used for a direct drive 6V mod with such a low vf.

Thanks again for the test results; seems these bin-unknown might have a 'special' place in the mod world....

IsaacHayes,
unfortunately the original luxeon was damaged, but not when removed from the heatsink. It should work, only one firm hit was needed. When I pulled out the heatsink assembly from the body with pliers, it slipped and damaged the dome. /ubbthreads/images/graemlins/frown.gif