*Completed* Converted Cat Eye Luminux Halogen to a direct driven SSC P4

P220C

Newly Enlightened
Joined
Aug 11, 2008
Messages
81
*****************************Completed*****************************
Final Setup

4x AA Eneloop in series
1.47 Ohm Resistor
1.09 A (peak for 30 seconds or so - stabilizes at 1.02A)
3.46 Vf
one hour run down to just under 1A (fairly flat) - haven't tested beyond that.
Seoul P4 U-bin
Constructed out of 1" Aluminum tubing, 1/4" channel and 1"x1/8" flat
Runs warm to the touch without airflow.
if you only used this while riding you wouldn't need the fins I don't think


DSC_6117.jpg


You can see the 1.0 Ohm resistor tucked in between the fins on the between the housing and the lamp assembly. JB Welded just above it is the 0.47 Ohm resistror (in series).
DSC_6129.jpg


DSC_6152.jpg


Surefire C3 w/ Lighthound Drop-In running on three fresh RCR 123A (12.6V) - Cree R2 emitter - The beam has a very tight hotspot, which makes for good long throw, but when riding at night, it's a little too intense of a spot, causing the spill to be less useful. This is my current favorite carry light setup. (not to mention the guilt-free lumens with the rechargeables). It's not the best constructed drop-in I've seen, but it works, and works well at that.
DSC_6147.jpg


My new light running around 1 amp - It looks like I I'm getting similar total output to the previous one (specs show the U binned Seoul P4 used here to have slightly less power that the Cree above at these amp draws), but this beam has much less throw due to the wider hotspot. I haven't ridden with it yet, but I have played with it at night, and the lack of a bright hotspot allows the spill to light up peripheral vision much better. I'm thinking a good setup would be a two floody light like this on the bars, and a hot throwy light strapped to my helmet.
DSC_6149.jpg





*****************************Update 1*****************************

Here is the first iteration
DSC_6098_005_Resized.jpg


After testing this, the heat sinking is woefully inadequate. While underway there will be some airflow, but I want the heat sinking to be able to handle use as a room lighter in a pinch. I am in the process of adding fins (1/4" channel) and have cut off the front plastic to allow more air circulation.

Run time is great, and output is acceptable, but I think I can squeeze a little more out of this this, if I can add enough heatsink. After a little nudging from Linger, and the results from me early testing I am going for 4X AA @ 1.47 Ohm (1.1A) if I can cool it, otherwise I will go 2.0 Ohm (0.89 A)

This is really a fun process.

Phase 1 testing results

76 F ambient
3 AAA Eneloops in series
0.47 Ohm, 5W wire wound resistor

0:00 elapsed
4.06 V resting
3.81 V under load
3.25 Vf
0.82 A

5:00 elapsed
3.94 V resting
0.76 A

Housing became hot to the touch - cut power - let cool - this configuration is too powerful for current heat sink - on a positive note, the backing plate and housing ran approximately the same temperature, which means that my JB weld construction is transferring heat.

____
Phase 2 test

(after housing cooled) Housing placed on copper table, and sandwiched between AL tubes to allow the battery-life test to continue. (timer was stopped during cooling)

10:00 elapsed
3.24 Vf

12:18 elapsed
0.73 A
3.67 V battery under load
3.23 Vf

17:00 elapsed
0.73A
3.21 Vf
3.67 V battery under load

40:00 elapsed
0.64 A
3.23 Vf

55:00 elapsed
0.59 A (If resistor is bridged, the batteries will output 0.91 A at this point)
3.20 Vf
3.54 V battery under load

1:12:00 elapsed
0.19 A (bridged resistor will produce 0.40A)
2.79 Vf
2.85 V battery under load
3.54 V battery resting

Phase 3 Test
4x AA Eneloops in series
5.63 V resting

@ 3.47 Ohm - 0.61 A; 3.2 Vf; 5.38 V under load
@ 3.00 Ohm - 0.66 A
@ 2.47 Ohm - 0.77 A
@ 2.00 Ohm - 0.89 A
@ 1.47 Ohm - 1.09 A; 3.46 Vf (we have a winnah!)










*****************************Original Post*****************************
This is a work in progress.

I have really started to ask questions and learn about the intricacies of lamp design here on CPF for a few weeks now, so please excuse any ignorance on my part if I say something stupid.

I will keep this thread updated as I progress through the design and testing phases.

Here is what I have so far:
3 x NiMH AA in series (4.2V 1,900 mAh)

Seoul P4 U bin with a 20mm IMS smooth reflector

Direct drive - Got a reading of 1.51 amps @ 4.11 resting pack voltage (want to add a resistor to get this down to a nominal 1.0 amp drive)

My Questions for Y'all:
What range of resistor values should I get as a starting point for testing?

Would this design be better if I ran four NiMH AA in a 2S2P configuration at 2.75 V, or would I end up with less light output?

[/B]
My old Cat Eye Halogen Bike light ended up being the host
(In its original state with 4x AA alkaline it put out 40 lumens with a 45 min runtime, which I thought was amazing seven years ago)
Now I'm running 3X NiMh in series.
TestingPhase007.jpg


TestingPhase005.jpg
 
Last edited:
Re: *My First Conversion* Cat Eye Luminux Halogen to direct drive P4

Nice work with the thermal paste. For applications where you need some hold as well, add artic epoxy to your inventory.
1) mix the two part epoxy with a small coating of artic thermal compound already covering the surface to reduce the strength of the bond and make it easier to disassemble your demo

I'd recommend you get a driver, multiple modes will really extend you run times, and if you get one with a slow flash rate that you can use while driving in traffic it will really increase your visibility.
But as such, I don't know jack about resistors.... :thinking: But I do know there is a little application I found:
2) LedPro knows all there is to know about resistors. You download it and run it (no installation, very simple program) enter a few of the numbers and it fills in the blank. For example, enter battery specs and desired drive current, it will tell you resistor values you need to make it happen:twothumbs Totally awesome program

LED Pro said:
LED Pro was created to help you make informed and educated flashlight modding decisions, and to calculate
aspects of your design, by entering certain known variables.

To use LED Pro, you merely enter all but one of the variables and let the program calculate the unknown one.

At this stage, the usage and results are very simple, and as time progresses the LED Pro Team will release updates
that allow you to complete more complicated functions, and do so with greater ease.

As of version 2.6, LED Pro can calculate 6 different aspects of flashlight modification.
A) Desired LED Current - mA
B) LED Vf (forward) Voltage
C) Battery Input Voltage
D) Battery mAh Capacity
E) Runtime in Hours
F) Converter Efficiency %
G) Resistor Ohms and Wattage

All you need to do is fill in all but one of the fields, and LED Pro will calculate the remaining field.

For most people, LED Pro will be used as a runtime calculator. The user will enter known aspects of their design;
such as the current they will run the LED at (i.e. BadBoy 750), the LED Vf (the average 5W Luxeon Star has a Vf of
around 6.8 volts. 1W Luxeon Stars has a Vf of around 3.4 Volts), battery input voltage (this is generally dependent
upon the light host), battery capacity (future versions of LED Pro will feature preprogrammed capacities for various
battery types), and converter efficiency % (preset at 83% for the dat2zip BadBoy sandwich).

The Resistor View, allows you to determine the proper resistors needed in a configuration of LED's in either series or parallel configurations.

That leaves only the runtime blank. The user then clicks "Calculate", and LED Pro will mathematically calculate the
approximate runtime, given the variables you supplied.

LED Pro is also backwards compatible, meaning that you can conclude any aspect of the design, by entering the
other 5 aspects. Basically: leave one field blank, and LED Pro will calculate the remaining field.

==================== Resistor Help ======================
If you get any negitive numbers from your calculations, this means that your current configuration will not work. You might have a problem such as: not enouugh Voltage
to run the LED's, too much resistance, or trying to get too many mA to the LED's with too little Voltage, or too much Resistance.

Thank you for using LED Pro!

-The LED Pro Team (jtice, rothrandir, logicnerd411, dat2zip)

Any Questions or Comments can be sent to [email protected]

Depending on the vf of your led, 2s might work out just fine. I strongly recommend you try it. You're learning right? Then try if 2s2p is any different.
3)rare earth magnets, to join batteries and on the end of wire leads, make it easy to adjust battery configurations. My suggestion is to go all out getting those 4 AA cells in there, just to increase by %33 the power on board. You figure out how to use it.
The reflector is nice, sure you have a great little ceiling bounce there, but I don't find that beam the most useful when riding. I like a uniform brightness. If you haven't tried it yet, I do suggest giving a bare emiter a chance. The best flood beam is from an aspheric lens really defocused, it gathers the light and spreads it evenly giving you a smooth beam perfect for spotting bumps / depressions.
The ledil Boom-M is a great reflector - completely smooth slow transition from dim at the edges towards bright in the center and back out to dim nearing the edge. Highly recommend it.
Conversely there are some great optics - medium (10 degree) is good for handle bar mounts, spot is good for helmets. I like optics to keep the beam away from on-coming traffic. I have a k2 tffc in a spot optic that just works pefectly on my helmet (shone in front of me it is just wider than my bike, shone ahead it diverges to perfectly reveal the path ahead.
 
Last edited:
Re: *My First Conversion* Cat Eye Luminux Halogen to direct drive P4

Dude, thanks. That program is exactly what I need.

I will look into that driver for my next build (achesalot/nightrider's dual MCE/P4 build). I want this one to stay simple. This is just practice for my next build.
 
Re: *Update 1* Cat Eye Luminux Halogen to direct drive P4

**OP Update 1 **
 
Re: *Update 1* Cat Eye Luminux Halogen to direct drive P4

I'll be interested to see how the heatsinking works out. I've got a couple of Cateye Micro II's that might get converted to a Cree. After doing a similar conversion on an old Vistalight, I find that at least 3 or 4 square inches of aluminum is needed to keep the heat manageable. The trick is doing this while maintaining a watertight housing.

good luck,
Steve K.
 
Re: *Update 1* Cat Eye Luminux Halogen to direct drive P4

Steve, I don't see any way to complete the conversion and have the battery housing be water tight. the switch itself appears to be encapsulated in plastic[edit: it's not, but it is behind a wall that is separated from the batter compartment when the top is on], so i should be good there, and the aluminum housing with the emitter will be sealed up with plexiglass and silicone.

I am way overkill on the heat-sinking now for on-bike use, and I should know tonight if it is enough to keep it cool while sitting still.


I cut the front part of the bottom housing
DSC_6103_000_Resized.jpg


DSC_6104_001_Resized.jpg


DSC_6106_003_Resized.jpg


I think I am going to epoxy some channel to the bottom of the housing and along the sides to connect the reflector housing to. My fear is that the thing has become so front heavy that it will fatigue the plastic unless I do this.

I am becoming tempted to use the existing housing as a battery pack, and just clamp the aluminum housing onto the handle bar and call it good.
 
Last edited:
Re: *Update 1* Cat Eye Luminux Halogen to direct drive P4

okay, I didn't see X-wing fighter resemblance before, but now.... yep!

The heatsink does need to be braced to the battery compartment somehow. It does seem like there's a point where the original light is such a small part of the design that it would be better to start with a clean sheet of paper. Maybe just keep the Cateye mounting hardware, and chuck the rest? Hard to say.

Steve K.
 
Top