Convoy L2/L6 modding thread

One thing that might help is a narsil driver. With ramping it has 150 brightness levels so you can choose what you want instead of being limited by high or turbo.

I do security work with mine. It’s so bright, I can sweep an area in like 3 seconds. I use it in short bursts usually no more than 10 seconds at a time. If I need to keep it on longer, I adjust it down a little.

Another thing to consider, it seems like the various bins of the XHP70.2 vary significantly, looks like between 1520 to 2015lm according to the cree data sheet. Where does one buy top bin XHP70.2 already mounted to the MCPCB?

This is the only one that I found already mounted to a MCPCB but I have no idea what bin it is from:

You can get them in the US.

You need to decide if you want white or pale white, etc… I think P2 bin is currently the highest at the moment.

That’s funny I found that same site just before you posted it. I like the color rendering I get from the 5000K XHP70 so I’ll stick to around that. Looks like 4000K would be better than 5700K. Would you agree? Seems like I can pick up stuff faster with the warmer light (at least when comparing to my much whiter Fenix.

What size board do I need? 16, 20, or 32?

I’m a CW guy, 5700-6500k. I don’t like any yellow in any of my lights.

You’ll need the 20mm.

For throw, I would’ve suggested an L2 instead.

I don’t think he’s looking for a thrower. I know I wouldn’t want to hunt for somebody in the woods using an L2 or a C8 light. You want good spill lighting to see all around you plus a decent sized hotspot to better identify things.

Unno. I seem to recall he wanted a tighter beam and more amperage. The former sounds like something throwier.

Hunting someone in the woods, I’d want a thermal imager. Just sayin’…

Actually, from what I remember of my L2, it had decent spill. Wouldn’t be like a zoomie where you’re projecting the batsignal with zero spill.

The larger head of the L6 with a smooth reflector will give best throw, use a buck driver at 5.5A or 6A to the XP-G2 and de-dome it, should get some 1100 lumens and a tight pencil beam. :wink:

It depends. If you’re surrounded by birches for example, a big spot or bright spill will blind you from the reflections. A small spot goes right between the tress.

Ask geofox784 what he wants, not me.

Hmmmm, anybody put an XP-E2 in an L6 yet? :wink:

I use that to my advantage in the house. Lately, I’ve been plagued by either fruitflies or gnats. I got one of those electronical tennis-racquets that zaps ’em, if I see ’em. A nice bright floody beam doesn’t light up the background all that much, but when the critters are close by, they light up like they’re on fire. Zap!

It’s also the same reason foglights are generally poopooed when it’s not foggy. Sure, they’ll light up the road immediately in front of you nice’n’bright, but that brightness almost blinds you to what’s way out there that your regular lowbeams would light up just fine, but is now relatively in the dark.

C8, if that counts. :smiley:

Doesn’t count, too little. :stuck_out_tongue:

We have one of those zappers as well, works great for the occasional wasp that gets in the house. :wink:

The L6 has significantly more output than your TK16 (~3500lumens vs ~1000lumens) and significantly more beam intensity (~75Kcd vs ~14Kcd). But because of the way our eyes perceive light, even a factor of 3 or 4 increase in intensity doesn’t seem like an incredible increase in brightness. This is probably the reason for your disappointment in the brightness. The 26650 cells you use could also affect the brightness. If they have too high an internal resistance it will limit the brightness.

It sounds like you want a bit more throw power, so I would recommend for you to dedome your XHP70 and increase the current. I actually would not recommend the XHP70.2. The throw with the XHP70.2 will be worse, for a given current, than the XHP70. It has a lower forward voltage so it is more efficient, and if you use a FET driver it can be very very powerful and bright, but in that case it really uses an excessive amount of current which makes it not practical for your use.

Removing the dome will increase the beam intensity by a significant amount, up to a factor of ~1.8, and narrow the beam some. With most LEDs we remove the dome with the help of a solvent such as gasoline, but with the multi-die LEDs like the XHP70 it is more difficult to use this method, though it has been done successfully. Instead we can remove the dome simply by slicing it off with a razor blade. See here for some info on increasing the throw of the L6. If you go this route note that it is important to slice the dome as close as possible to the LED die surface. If you don’t slice very close the increase in beam intensity will be less.

Changing to the smooth L6 reflector will also help with the throw, but when used with the XHP70 there will be a “donut hole” in the beam. This doesn’t make it unusable by any means, but some people are sensitive to this sort of thing.

To increase the current you could do a resistor mod as described in the link above, or you could get a FET driver like the TA driver. There are FET drivers from mtnelectronics that will work in the L6, for example this one. If you go this route you can get the D4 firmware which has a really intuitive and powerful ramping user interface. Using a FET driver will send more current to the LED in the max mode (although you can choose any brightness you want) and might increase the brightness by an additional factor of ~1.5 or so.

As far as your heat failure questions they are sort of difficult to answer both because what fails first depends on several factors, and most of us turn off the light or lower the output before something actually fails, so there is not a lot of actual data. But the components in a flashlight are pretty tough and would most likely not fail even if left on high until the 2x26650 cells ran out. As mentioned above, the light will become too hot to hold well before anything is damaged. Metal becomes too hot to hold with the bare hand at around 55C or so, which is not very high for LEDs or components on the driver.

Note, however, that when LEDs are dedomed or sliced this can leave debris or inconsistencies on the surface of the LED die which can burn and ruin the LED. So keep an eye on the LED surface for the first 10 minutes or so of being on high. A burn will show up as a black spot.

Thanks for all of the great information.

First of all, what is a “normal” amperage though the LED on the stock L6? I just measured 5.43A initially and 5.23A after roughly 30 seconds.

I am using “26650 KeepPower 5200mAh Protected High Discharge Flat Top” batteries. Item number 2239 on ILLUMN. Should have a 9.5 max discharge rate. Those are fine right?

Why would the 70.2 be worse? Seems like the LED’s inside the dome have less of a gap, so there should be less of a donut hole. I ordered one yesterday so I am hoping it wasn’t a wrong decision.

My current plan is to stick the 70.2 in with a SMO reflector. Also do a resistor mod with a R120 resistor. Although now I may go with the FET driver you linked. What amperage should I expect out of that? I don’t see anything listed on the website.

I don’t necessarily want a true “thrower”, just a slightly more focused hot spot with a similar.

With the D4 firmware thermal protection will the light just cut out completely at 60C or will it just tone down the intensity to prevent damage? I couldn’t find much information on it.

No problem, happy to help.

I think around 5A is what the stock L6 sends through the LED.

The battery voltage drops as more current is drawn. “High discharge” cells have low internal resistance which reduces the amount of voltage drop, and this is good for flashlight performance. The driver that comes in the L6 is a buck driver which will send a regulated ~5A of current through the LED as long as the battery voltage is above the forward voltage of the LED at 5A. As the battery is used up the voltage drops and the current and brightness will reduce. A battery with low IR will stay in regulation longer, so you get your high mode for longer. With FET drivers the current that is sent through the LED depends on the battery IR; lower IR cells result in higher current. Those keeppower cells are not bad, but there are better cells available. For example the Shockli 5500mAh cells from mtnelectronics are some of the best available in terms of capacity and low IR.

With a FET driver and the XHP70 you would get between 8A and 12A depending on the cells you use. The XHP70.2 has a much lower forward voltage and so a lot more current will flow when using a FET driver. With high performance cells you might get 20A, which is really not sustainable because of heat and reduced runtime.

You will get less of a donut hole with the XHP70.2, but the beam intensity will be less than the XHP70 at the same current. If you want more beam intensity a dedomed/sliced XHP70 will be better.

That being said, if you just change to the XHP70.2 and keep your stock buck driver, the light will be more efficient and get less hot. But the performance will not be much better.

The D4 firmware thermal protection smoothly reduces the output and will not just cut off completely.

Assuming my “26650 KeepPower 5200mAh Protected High Discharge” batteries regulate the amps to roughly 9.5A (going off the advertised max discharge rate) with the FET driver, would the XHP70.2 be reasonable in direct drive mode? I guess even if it isn’t, I can always just dim it some to get to a sustainable temperature.

What about the XHP70.2 causes less beam intensity? Seems like it is the same other than the smaller gap between “cells”. On the CREE data sheets I see 120 degrees for the viewing angle on the 70 compared to 125 degrees on the 70.2 . Is that what would cause the reduced intensity? What about if they are both dedomed?

Seems like for the same voltage I should get much better performance out of the 70.2 based off of the forward voltage charts in the data sheets.

I’ve got some experience in electronics, but a bunch of this is new to me, so I apologize for the more basic questions.

I beg to differ EasyB, the 70.2 will definitely give more throw in the L6, especially if the smooth reflector is used. A slight bump to the stock driver, with the 70.2 and a smooth reflector and it’s pretty much a drastically different light. Going with an FET driver and that combination does indeed yield huge current draw, not necessarily a good thing, so simply raising the ceiling on the stock driver in conjuction with the more powerful emitter should work very well.

Tighter hot spot is usually more intensity, which is of course where throw comes from. Removing the donut hole emitter in favor of one that doesn’t have this issue is a direct impact on intensity and throw. Win win if a longer beam and throw are the request.

The XHP-70, with the separation between the dies, facilitates the need for an orange peel reflector to blend the beam, which also diffuses intensity and reduces throw, creating a floodier light pattern. The XHP-70.2 alleviates that problem.

Example: SupFire M6 (modified to run a TK61 2S2P carrier) with 3 XHP-70’s and an FET driver. Stock triple reflector. 5,796 lumens. Swap out the 70’s in favor of 70.2’s, 11,764.5 lumens with no other changes. From all flood to a large defined hot spot, like a completely different light. Some spring bypass attention and it went up to 13,558 lumens. The 70.2 is a big step up in the right direction.

The 70.2 is the reason my own L6 makes 9,200 lumens. FET driver and over 18Amps on premium flat top non-protected 26650’s.