mod --> UltraFire Cree C3 5-Mode Flashlight with 2xAA

Both link suggest this board cannot produce constant brightness, I find this odd after reading this thread: http://candlepowerforums.com/vb/showthread.php?t=190499

And in particular this graph:

From what I learned on this site, constant current is required to have constant brightness. This board is a constant voltage regulator so I expect the current would change depending on the state of the LED. Thus brightness is not constant.

This is really mind-boggling (to my newbie's mind at least). What is the reason for producing constant voltage drivers for LED flashlights? Predictable runtime? I went searching for one constant current driver similar to this and I found this:

http://auction1.taobao.com/auction/item_detail-0db2-783e6dbfbfcc20732083c18375c17d3e.jhtml

Needless to say I recommended to DX but so far nothing happened.

If it doesn't provide constant brightness (although stefan's graph clearly shows it does), I don't understand why it is drawing constant power?

I set mine to 1.6a off a 3.7v battery, if I pop a 4.1v battery in its only pulling 1.4a.

I think some needs to conclusively test this driver.

Arghhh

I have accidently short cuircited the board ... It doéesn't work now (just the modes and a direct useless connexion to the led ...)

I am now waiting for the repalcement driver to arrive ... Still a few weeks

For the moment I am testing the AMC7135. See here for some results (with a "strange" mixed batteries configuration...)



JP

Hi all,

I have done the following modification to my UltraFire C3 with extension tube

- replacement of the LED by a Cree Q5 :
- replacement of the driver by 3xAMC7135 board :

- powering the flashlight with the following configuration :
- 1 NiMh battery 2000 mAh
- 1 protected Li-Ion battery 14500 capacity = 900 mAh

This configuration is safe provided that :
- both batteries are charged independently
- both batteries are fully charged (at the beginning)
- the Li-Ion is protected and has a lower capacity than the NiMh

With the previous precautions, you can be sure that :
- the Li-Ion will empty first (less capacity)
- the Li-Ion cutoff pcb will protect the battery for potential reverse charging when it is empty
- the NiMh will never be empty before the Li-Ion (so it will never enter into reverse charge)

Advantages :
- real constant current through the led (1A in high mode)
- AMC7135 is used in its high efficiency curve
- you will empty the Li-Ion with a led at constant max brightness :candle:

- it can also work in the single 1 Li-Ion configuration (without the NimH) but this config will not garanty the constant brightness as soon as the Bat voltage will be under Vf + 0.1V...

- my wife prefers the look of the UF C3 with the extension tube

- the driver has a memory. Switch off the light, it will go back to its previous setting when switched on again.
- the driver has medium and low modes, this can prevent a fast discharge of the bats and have light up to 4 hours and more.

drawbacks :
- the 2AA configuration does not work very well with two NiMh cells... It is safe but the led is very dim (2.4V instead of 3.5 needed).

- the 2AA configuration will not work with two Li-Ion cells... it would destroy the led and the driver

- be sure that the NiMh has always more capacity than the Li-Ion. Beware that after a few weeks unused a NimH has a greater selfdischarge than a Li-Ion... recharge the NiMh bat to avoid reverse charging of the NiMh by the Li-Ion

Here is the result :

Finally I must say that :
- this config provides a very very bright constant output
- 1A in the led during more than 50 min (ie the real capacity of the 14500 bat ... very close to 900 mAh announced capacity (theorytically 54 min at 1A))
- the cut off works well and insures a safe shutdown of the flashlight
- the heatsinking is good and efficient. The flashlight is hot but you can hold it without burning :devil:

Conclusion : after having tested a lot of configurations, I stay with this one (at least during some time) :wave:

If you are tempted to do the same mod. Please understand the red paragraphs before doing it. All the words are important !

JP

As an added precaution, please read this protected Li-ion battery safety thread before using protected Li-ion batteries.

Thanks for providing this link. It is really important to understand the danger of Li-Ion cells.

BTW the 14500 with protection can be considered as safe in my configuration as :
- extract from the link : "A protected Li-Ion cell that shuts off due to low voltage cut off is less of a concern. The high and low voltage protection circuits are designed for cyclic use, however I am not sure how many cycles they are rated for. I would also expect that there may be a difference in the quality of these circuits that may be reflected in the cell price."
- the cutoff in my configuration is used to avoid over discharge
- overdischarge of a Li-Ion would permanently damage the cell but is not dangerous as the cell is empty.
- So even if the cutoff circuit wouldn't open and thus the flashlight remained ON, it would lead to totally empty the Li-Ion and that's it. No danger just the cell out of order for a new charge...
- however a good advice would be to check that the cutoff pcb is working well on your cell... if not you can garbage the cell (and 4.5$).

- the current drawn into the LED is constant and remains at 1A. That is just the good value for a 1C cell. This 14500 Li-Ion should be safely used up to 1.5A (the protection circuit cuts the cell when above this value). I have checked that the cell is just becoming a little warm but not hot (and this is important for safety reasons).

- bear in mind that this protection is not a "mechanical/chemical/thermal" (venting or a classical PTC or fuse or bi-metallic resettable over temp switch proctection) that should be present into the + end of the cell.
It is a pcb added at the - end of the cell which protects the cell by the electronics.

(following picture is not from my cell... but same trusfire brand)


JP