Ultrafire LZZ-06 AAA light with pocket clip 0.9-4.2v

Thanks again ericandor. Will try cap replacement first. I will report back on results after I work on this tonight.

The more I read the data sheet, the more I like this chip. I need to see if the driver followed the placement recommendations in the data sheet.

No problem! Can’t wait :slight_smile:

Check PAM2801 spec, it is really interesting!

Diode used is 1A (2A recommended), inductor is 3.3uH (4.7uH recommended) and so on…

Well, tonight was frustrating. At the end of yesterday, the light only had high mode and occasionally a firefly mode. I figured I may have partially fried the modes chip. I think the undersized inductor and diode may also be interfering with mode changes. They are probably also limiting current with 2S Ni-mh cells. Highlights:

  • Swapped the stock caps with known good 1uF and 10uF caps. Didn't improve mode changes.
  • I swapped the resistor for a R100. Current with 1S AA was still 310, 470mA with 2S cells. Should have been able to hit 750mA. Modes stayed the same.
  • Swapped in a Nichia 119H while putting the light back together.
  • Checked tail cap current of the assembled light and only got 1.3amps of current on high (Long stock DMM leads probably really reduce current. Funny thing, the modes returned including strobe.
  • Pulled out another light that had not yet been used or changed. It only had one mode for whatever reason. Inspected the driver and noted no visible issues. Swapped the caps and still only had one mode.

I plan on still playing with the driver. Like you alluded to ericandor, the inductor and diode are under rated for this modes chip. I'd like to upgrade them and see what happens.

Correction to diode markings that I posted earlier. It has two marks. At some angles a large "S4" is visible and it covers almost all of the diode top. From straight above, there is a small 6 after the large "S" and there is and 82 going side ways on the end of the diode that does not have the polarity lines.

Absolute Maximum Ratings for Supply Voltage is 6V and there are 1xAA/2xAA flashlights eg. C3 (with extension tube) based on PAM2803/5 working fine, so it is highly unlikely that you have fried PAM with 2xAA (NiMHs or Alkalines).

All PAM based driver (all but one :wink: ) I have seen use undersized inductor and diode, but modes work fine.
I have just spotted that on PAM2805 EV Board (link) inductor is 2.2uH/>2A, Schottky Diode is 2A/20V and it is stated that it can deliver 750mA of emitter current…

That is strange, maybe there is cold solder joint somewhere, have you already resoldered all the components ? :slight_smile:
Or maybe there was a batch of broken PAMs…

Have you already tried 1R00 (1 Ohm) for Rs to see if it works fine with low currents ?

Best wishes

Sorry, I should have stated that I think heat from repeated soldering and reflows from hot air gun may have damaged (fried) the chip.

Joints are all good. Checked them after each reflow or solder. I think the resistance of the DMM leads used for the tail cap measurements reduced current to the point that the inductor was not getting saturated or something to that effect.. I don't know. I will keep trying different ideas on the driver. A bad batch of chips seems like a reasonable possibility.

I have not tried shorting the voltage sensor pads. I will try to remember to try that next time I work on this driver. Shouldn't help the modes, but it would be interesting to see the impact on current.

We learn a little more each time. Hopefully, the list of possible issues narrows down after the next session with the driver. I need to figure out what inductor and diode to order so that I can see if they are the problem. I know you are not too concerned about their ratings in this application, but maybe one of them has a defect. I would like to upgrade them 4.7uH and 2A while I am swapping.

Thanks again ericandor for your help.

What do you mean by “shorting the voltage sensor pads”?
What is the purpose ?

Ah, I miss read the below question you asked. I thought you asked it I had tried a 0 ohm resister. My apologies. I was reading the monitor from about 5 feet away while getting ready for work.

Have you already tried 1R00 (1 Ohm) for Rs to see if it works fine with low currents ?

The answer to that is no I haven't. Good idea. I will try that out.

Separately, I forgot to report. The firefly mode I reported was actually more of a moon mode. Also, later in the night it reliably cycled from High to Moonlight. Maybe because the battery voltage had dropped. It was actually a very nice Moonlight level. Perfect for night adjusted eyes indoors.

Separately, the Nichia 119H cut lumen output in about half from the stock XP-E emitter.

Received a purple one from AliExpress today. Looks good overall. Comes with a nice plastic case, not too fancy, but much better than the usual cardboard box. The LED is off center on mine, I didn't have success in fixing it. Makes the beam quite ringy, even the hotspot.

The mode-changing isn't perfect - it sometimes stays on the same mode, my sample keeps skipping strobe most of the time (I'm rather happy about it :) ). I was unable to detect PWM on low mode, not by camera, not by shining on PC fan. It's either current regulated or the PWM is very fast, must be much faster than Convoy's.

The head o-ring was very "fat" making it almost impossible to change modes with one hand. After few times of opening and closing the flashlight the o-ring would stuck between the head and the body and make it impossible to close the flashlight. I have replaced it with one from the tailcap and it now works much easier and it can be operated with single hand. But there is also much more play it the threads

Mine didn't come with a case. Maybe because I bought a set of 3.

Even centered, there are some slight artifacts in the beam you can see on a flat wall. Replacing the XP-E with a Nichia 119 made the beam real nice and smooth. I think XP-G would be perfect for this host. Need to find a good AA driver. The 5 mode one at FT is out of stock.

I have put a piece of iPhone matte screen protector on the lens and the beam now perfectly smooth (with the protector - on the left, compared to original beam - on the right)

BTW the color isn’t that blue, it’s just cold white.

Edit: Added the picture of the case:

Cool mod and cases Pregulla. I hope I get cases on my second order as these will be gift lights.

So what was the result ? :slight_smile:
Were you able to make it single mode by removing BLK capacitor?

Haven't had a chance to return to this driver. Thanks for checking. I will report back when I do.

So far, since Post 53 above, the one that I put the Nichia 119H emitter in is cycling between High and Moonlight mode very reliably. I like the modes so much, I will not be modding the light any further. I've been EDC'ing it since I put it together and I'm really digging it. I would like to know why it's has a Moonlight (instead of a Medium and Strobe) and why it is reliably changing modes now, but I know better than to look a gift horse in the mouth.

Hopefully, experimenting with my other Jexree Mini drivers will answer some questions.

I have just received my two Jexree Mini. Both of them suffer mode switching issue. Most of the time mode is H, sometimes M, very rarely STROBE.

I have observed that mode switching works well when head is tight, tail cap is removed and negative battery terminal is shorted manually to the body.

That leads me to conclusion that most likely the problem is in bad electrical contact between head and body (fluctuating/unstable Vin while switching ON and/or OFF) causing PAM2805 start-up problems (and fallback to default mode H, because eg. Cblk is being discharged before PAM start-ups).

Any thoughts on that ?

Best wishes!

I initially wondered if that was the issue too when I talked about a potential driver seating problem early on. I thought I got comfortable that it wasn't that in my case, but I can't recall exactly how I determined that. I think it's a good idea to double check as it seems like a possibility given the design. I don't recall if I pulled the driver and soldered it to wires isolate out the host issues. I guess I should do that to make sure. I will try to do that this weekend and report back.

I didn't notice that mode switching was much better when I had the extra resistance of a DMM between the cells and the driver.

Great to hear you got one of these guys. You seem to have a strong electronics knowledge base. I just received my second order a couple days ago. Please let me know if you want me to try anything on one of mine. I sure would be nice to somehow make these drivers single or multi mode reliable.

EDIT: When I wrote the above I was rushing to head out the door and my mind was in too many places. I'm pretty sure its not just an intermittent electrical contact issue. I did most of my work with the driver out of the light and wires soldered directly to the cell contact pads. I suppose I could have been having s similar issue when touching the wires the cell I was using to test with. Seems unlikely as I work on most drivers this way and haven't had this type of issue before. Still haven't had a chance to work on one this weekend. Hopefully I will get a chance today.

I still believe that it is about the head - body contact, and maybe changing Cin will solve the problem. Could you please try to replace Cin with eg. twice or event ten times the capacity to see does it makes the difference ?

Could you also please try to reproduce what I have observed (working modes) but using something else then DMM leads ?

Is Cin the cap in bay "C1"?

I only have 1uF and 10uF caps. Should I stack a 10uF on C1?

Any suggestions on how I can do what your asking in the last request?

I am unable to tell looking at the picture, and have no opened mine yet (lack of time… btw. how to remove the driver, is it glued ?).
Cin is the one between pin 6 (battery +) and pin 2 (battery -, GND).

Stacking 10uF should do (it would double the Cin capacity).

Unscrew tailcap, tight head, use solid metal piece to short battery “-” with body, (fork will do ;))

Best wishes

OK. thanks. I'll try those things and report back. May be a couple days.

Yes, the driver is secured with some type of substance. It doesn't have a strong bond to the aluminum though. Becareful for the lip around the driver. It is very thin and bends easily.

I just received 3 of these in cases. 1 has a working strobe but none of them change mode reliably. Has anyone found a fix for this?

They’re not much good for gifts if they’re dodgy.