Fully functional U2 modding with SSC P4 USXOI

Inspired by ICUDoc's U2 mod, this is my first step of U2 modding, I am testing initially with external mounted emitters.



Following is the U2 behavior when driving SSC P4 USXOI / CREE P4 WC, current measured at the emitters. U2 head was connected directly external power supply set at 3.8 Volt.Level 1......23.8 mA ..........23.5 mA
Level 2......50 mA .............50 mA
Level 3......280 mA............324 mA
Level 4......390 mA............351 mA
Level 5......450 mA............425 mA
Level 6......800 mA............799 mA

At level 6 when driving SSC, current is regulated at 800 mA from 3.7 V until 4.37 V. Above 4.37 V, current will increase exponentially, looks pretty much like direct drive. What a magic number ! it is only a little bit above 4.2 Volt. Nevertheless, this is very good as we can achieve 800 mA from 3.7 Volt cell. If connected to WX0S, U2 could only supply 700 mA from 3.7 V input.

Before the real installation of SSC P4, I would like to at least try to narrow the gap between level 2 and 3. Will try to serial 1-2 Ohm resistor. I hope power lost at the resister will not be too much => 0.64 – 1.28 W. I have yet to find 1-2 Ohm resistor for further testing.

If you guys have any other idea how to narrow the gap between level 2-3, please kindly suggest !

Re: First step of SF U2 modding with SSC P4 USXOI

cm048
Great job- thanks for the data.
You see the same gap between 2 and 3 that I saw visually- a big one.
Your data is nice looking and well presented- thanks.
I don't know what the resistor will achieve as the regulator in the U2 tries to supply constant current, it would seem, so it might just boost the voltage to maintain the same I across your increased R.
I don't mind the jump too much- tell me, is your SSC mod bright as heck?

Re: First step of SF U2 modding with SSC P4 USXOI

It would also be interesting to see what the Vf of the LED is at each of those drive currents, and what happens to the drive current of the lower levels when the Vin is in the 4.10 - 4.20 v range.

Re: First step of SF U2 modding with SSC P4 USXOI

Very good news, I found the way to make U2 fully functional with SSC P4 USXOI.

The trick is to serial one diode with SSC emitter. It will be around 0.7 Volt drop at the diode. Though I mod it with 1N4004, 1N400X will be fine. The result is so beautiful. The current in the table below measured at the emitter unit in mA.

Vin (supply)................3.2.......3.6........3.7......3.8........3.9......4.2........4.4
Level 1 (mA)..…........…..0…….24…..…24….…24…...…24……24…..…28
Level 2 (mA)..…...….......0…….50…..…50….…50…...…50……50…..…50
Level 3 (mA)..…...….......0…….93…..…93….…93…...…93…...125…...211
Level 4 (mA)..…........…..0……195…...195…...195…....195…...195…..240
Level 5 (mA)..…........…..0……376…...380…...380…....380…...380…..380
Level 6 (mA)..…...….......0……762…...781…...796…....797…...799…..799


I mounted 1N4004 Diode with thermal glue at the back side of the emitter sink. The maximum power lost at the diode is only 0.7*0.8 = 0.56 watt !, moded U2 is fully function with 3.7 unprotected cell. The stepping from level 1 to 6 is so smooth like it was with LuxV. Of course, it is a lot brighter !

I hope this will be helpful for you all, it is time to open up your U2.

Loosing some runtime but your levels are back. Still probably better runtime than the LuxV even with the diode.

Legend!

Awesome find !

Re: First step of SF U2 modding with SSC P4 USXOI

cm048 said:

Very good news, I found the way to make U2 fully functional with SSC P4 USXOI.

The trick is to serial one diode with SSC emitter. It will be around 0.7 Volt drop at the diode. Though I mod it with 1N4004, 1N400X will be fine. The result is so beautiful. The current in the table below measured at the emitter unit in mA.

Click to expand...

Good work! You could do the same thing by installing an SSC with an unusually high Vf. What is the Vf of the LED that you used in this installation?

cm048 said:

...At level 6 when driving SSC, current is regulated at 800 mA from 3.7 V until 4.37 V. Above 4.37 V, current will increase exponentially, looks pretty much like direct drive...

Click to expand...

Thank you for that data. I have long suspected that the claim that the circuit is buck/boost was in error. This should save other people from modding their U2 and then popping in 2xCR123's or 2xRCR123's and then blowing a $250 flashlight.

CM said:

Thank you for that data. I have long suspected that the claim that the circuit is buck/boost was in error. This should save other people from modding their U2 and then popping in 2xCR123's or 2xRCR123's and then blowing a $250 flashlight.

Click to expand...

I don't think anyone has ever argued that the U2 circuit functioned properly when Vin was much above Vf of the LED: it certainly isn't a buck circuit as we usually think of it. I suppose it is a matter of semantics, because it does function with some buck properties when Vin is below the LED Vf. Under those conditions, the U2 circuit is able to actively limit current to the LED to levels well under what the current would be direct drive - in a fashion similar to "buck" circuits.

So what do you call it? I guess it doesn't much matter as long as it is understood exactly what the circuit is and is not able to do.

nice work... have you tried 6V similar to using 2x CR123?

with five watt luxeon and single 18650, most could not get regulation in highest level depending upon vf of luxeon.

:kewlpics:

(you know what I mean...)

cy said:

nice work... have you tried 6V similar to using 2x CR123?

with five watt luxeon and single 18650, most could not get regulation in highest level depending upon vf of luxeon.

Click to expand...

Adding a diode with a Vf of about .7 v to a Luxeon with a Vf of 3.6 v gets you to a Vf of 4.3 volts - thats not going to regulate with 6 volts Vin, it will be direct drive at a level likely to damage the emitter and perhaps the U2 circuit.

Very nice to find!
Has anyone looked at the U2 driver circuit itself? I am sure its microprocessor controlled but would still like to know how it does current sense. It would be great to do your mod but drive the emitter at 1.2 to 1.5 amps on high.
If I can get one of my U2's open I'll take a look.(never had any luck with my current ones though)

Edit - it appears to have the U2's regulation circuitry functioning properly on all levels the trick is to have Vf greater then Vin while Vin is greater then 3 volts.

are you measuring these at emitter or at cell. what is the vf at emitter?

cm048 said:

Vf @ 800 mA around 3.5 V. Technically, serial extra diode will be virtually the same as unusually high Vf emitter, but don't forget that we have more control with diode or zener diode.

Click to expand...

i cant bare to forgo option of using 2 xcr123 disposables and sticking to rechargebles all the time.there are times when disposables are more convenient to be used. is there any way of modding the u2 while preserving the disposable batt option?

would be curious to know the outcome of using smd resistors serialed to ssc leds and if it would also work for rechargebles.

kudos to Icudoc and cm048 for informing a 2xcr123 would fry a direct u2 ssc mod.

You could make a U2 with a P4 work properly with 6 volt primaries by putting about 4 power diodes in series with the LED, but I'm not sure it would be worthwhile to do so. Those serialed diodes would be dissipating almost as much power as the LED, and runtime would suffer. You would basically get the same runtime as a U2 with a Lux V.

Is the P4 enough brighter than a Lux V for it to be worthwhile to do this kind of mod? I suspect it isn't. It would be an interesting experiment, though.

Measuring at the emitter, Vin for the cell = Vin supply in the table.

cy said:

are you measuring these at emitter or at cell. what is the vf at emitter?

Click to expand...

It is difficult to be on both side, 6 V Vin will require maybe 2.7 V drop at the diode. You can try 2.7V 3 watt zener diode with around 2.2 watt power lost at the diode. I think it is too much, and does not make much sense for moving from LuxV. More over, you won't be able to get 800 ma level 6 with 18650.

Fat said:

i cant bare to forgo option of using 2 xcr123 disposables and sticking to rechargebles all the time.there are times when disposables are more convenient to be used. is there any way of modding the u2 while preserving the disposable batt option?

would be curious to know the outcome of using smd resistors serialed to ssc leds and if it would also work for rechargebles.

kudos to Icudoc and cm048 for informing a 2xcr123 would fry a direct u2 ssc mod.

Click to expand...