It has 3 modes (Low, Med, High) with good separation. I would prefer even wider separation, but have not seen any stock light in this price range with wider separation than this light.
I will try to take beam shots of the 3 modes tonight.
Thank you for asking freeme.
Thank you for the review. 
I noticed something weird.
I don’t understand. Are you talking about over voltage protection?
I have (had) few variants of this SolarStorm X2 type of headlight. Never purchased in a package with batteres and charger as the battery pack in most cases is rubbish. It has protection circuit but none of them prevents the cell from became unbalanced. With poor cells this happens rather quick and you can't charge it full. It does prevent to overcharge each pair of cells that's why you can't charge it fully to 8.4V. ImA4Wheelr, this might be the case you thought the charger can't pass 8V. Probably it can even overcharge the pack without protection circuit. At least I had a few such ones.
The lights itself are good for their's tipical price. Of course none of them had thermal paste anywhere around the pill. Some pills were press fit, some with threads and some just loose and fell apart when opened. Luckily never got the one with no pill at all.
Modes should be much better separated. All I 've seen so far are 33% 66% 100%. It should be 5%, 35%, 100% or like.
Beam is bit to narrow for MTB, could be good for helmet light, but not for the bar one. I would try to mod it with 21mm TIR lenses to get better beam.
BTW, you can get it for lower price and if you are lucky you can get the head only for around $12 on ebay.
EDIT: no need to be lucky -> see THIS
It does not steps down AFAIK. If you do a bit better closeup pictures of the circuit one can tell more (comparing to others seen so far). It is also unregulated so it is getting dimmer as voltage drops. Voltage drop is important to 3 signaling leds. Better (thicker) wires would prevent bigger voltage drop.
Charger is probably tipical low cost 8.4V 1A. You should test it if does not overcharge with some spare old Panasonics or Sanyos. You can leave them to be overcharged for 0.1-0.2V safely. I had bad experience with some EU plug type ones. Some even does not stop charging or charge even if you connect fully charged pack. That's why protection circuit in the battery pack is important.
lagman wrote:
I don’t understand. Are you talking about over voltage protection?
I don't understand either. I tried to test it to determine what it does (OP has been updated for new testing). It seems to have bricked out. All I ca say is that it connects to the cells like a balancing circuit. It also connects the cells to the light. I it appears to be at minimum a balancing circuit. I may also serve an overcharge/over discharge protection role. I don't have the expertise or equipment to determine. I Googled the heck out of it and found nothing.
ledoman wrote a bunch of stuff:
My testing also determined there is not thermal step down (OP has been updated for new testing). Mine does seem to regulate current fairly well. There is an inherent limit to regulation though when you have 2S cells and 2S emitters. I also agree that buying the charger and battery pack is a waste of money.
As far as those cheap you units you linked too. Hopefully they are the real thing and are not this:
The below driver may be fine, but appears to be less sophisticated.
nice boost in power!
Definitely need to beef up the thermal capabilities when you kick XM-L up to 3A~ or more
Those stock batteries are the typical Chinese bulk stock crap (found in those cheap mobile power units and such)…would be better to build your own pack out of known good cells even laptop pulls would double the mAh rating or more…either way they are functional but uber cheap batteries
I wonder if one Cool White and one Warm White in that light will create a wide spectrum light, or go with 2 T6-4C neutral and be done with it
Also those A2SHB FET’s wonder if a swap out with some of the irlm2502’s would boost the power out
Nice work. I think I have never seen this variant of driver. I've got green and black ones. You may read a looong story about those type of lights at MTBR started by well known Leaftye.
Senese resistor mod is getting you to much current for such small housing. It will overheat unless you will use it on Hi only during moving (cycling).
I would suggest you to correct review about battery protection circuit. I'm 100% it is not balancing in any way. I've never seen it even with beter Magicshine packs. It is basicaly only overcurrent, overvoltage protection similar to those built on the batteries. Difference is only it is for two cells. Have you managed to measure voltage of each pair of cells before dissaembling or separate charging? This would tell you more.
WarHawk-AVR,
Thank for those thoughts. Was thinking about mixing tint too. No sure I want to with just 2. Seems like the different right and left spill would be too distracting. I need to look into that FET idea for just generall knowledge. This light was overdriven stock and 3 amps will just need to be used as a short turbo if there is no air movement.
ledoman,
Thank you for you thoughts. Wow, that's a monster thread. I agree. 3 amps per emitter is too much for this little guy when not in motion.
Not sure what you mean about correct my review. I only speculated about what I think it mostly likely is. Why do you think it connects to all three points in the cell bank? The PCB is clearly marked that it is supposed to be connected that way too. All cells were equal voltage levels when I torn down the pack, but that doesn't prove it's a balancing circuit by itself.
By my experiences all those (I've seen so far) circuits does not do balancing. Three connections are used only to monitor voltage of each pair and doesn't allow to overcharge any of the pairs. It also cuts of power in case of leads shorting or when overdischarged. That is all I know they are doing. As said even pricier Magicshine packets does not have balancing capabilities.
You are right, if all cels were of same voltage it doesn't prove anything but cells are probably equal.
You can easily prove it by connecting two cells of different voltages and try to charge. Maybe yours is different. Who knows with all those chinese variants. Would like to know about it.
I'll update the OP with that info so that it clear that it is not likely a balancing circuit. I hope it's ok if I reference your post as the source. If not, please let me know and I will delete the reference.
I will try the charging thing you mentioned and report back. As stated in the OP, I tried to drive the light with cells of different voltages and it would not transmit any power. Unfortunately, it doesn't transmit power with matched cell after that. So I don't know if it bricks out when tripped with an imbalance or if I inadvertently damaged the circuit. I would like to know more about it too because I could use it if is does include a balancing circuit.
Thanks again ledoman for taking the time to share your knowledge. Good info.
freeme,
I told you I would try to take more beam shots. I didn't take any last night as it was raining last night. I will get some after I finish modding the light. The light will be slightly brighter in the lower modes from stock, but should still give a general idea.
You can take me as reference no problem. If the circuit doesn't give any power it has probably triped out. Short charge should put it back into life, same as with protected 18650.
No problem to adding you some of my knowledge. I've been into bicycle lights for few years now.
ImA - just spotted this review! Pills are just like my SolarStorm X3:
I added the thermal paste and as posted previously elsewhere I also upgraded to XM-L2 T5 5B1 emitters for a glorious tint! I also just finished rigging up a method to measure battery pack currents (i.e. "tailcap current") on these lights (will post in a new thread) by buying these power connectors used for CCTV use and wiring them up like so (bottom of pic):
I then connect up my DMM inline and toggle modes (and also see the drain while in "off" position) - here is High mode on mine:
I'm also using the famous Pannova battery pack/holder. Perhaps you should invest in one. It's available from DX cheaper and found on Ebay too.
-Garry
personally more tempted by a Gemini Duo clone from Yinding for my bike
Sweet setup Garry. I think I need to copy some of that.
Although the pills are pressed in nicely, my testing indicates that they or the aluminum bases are not pulling out the heat fast enough (Had significant thermal sag, but current remained steady at 2amps/emitter and exterior temp only hit 195F). I think I will replacing my pill with copper. Not much copper though because the the limited surface area will be the next bottleneck.
Hi nisse. Welcome to the forum. You got a link to that light you mentioned?
thx! love all the useful info to be found on this forum.
the Yinding sold here there and everywhere but I have not found it without batteries would like to use it with the Pannova holder and some “real” batteries
http://www.fasttech.com/product/1443727-yinding-2-cree-xm-l-u2-4-mode-1800-lumen-white
He's talking about this one also sold by Simon on Ali, though he doesn't appear to have the NW version anymore. As with the X2, MTBR peeps are finding these Gemini Duo clones to be of varying quality. Here's a link to the original Gemini light. Also of interest is this one at Lightmalls, head only, with the programmable 10 step driver (driver reviewed by Cat-man-do on MTBR in his KD Tri-clone review). I've wondered about picking up this one myself as I'm afraid my X3 will suck the battery down pretty quick.
Edit - he beat me to it.
-Garry
and as a roadie what I would really like to see is a light with more of a 16:9 format beamshot ie not circular as a flashlight