Solarforce XM-L U2 vs CUXM2 XP-L V4 3D Runtime

Here's a comparison on the Solarforce XM-L U2 module and IOS XP-L V4 3D module.

The Solarforce module has a Smooth reflector and the CUXM2 module has an Orange Peel reflector.
Same L2T host, same Sanyo UR18650FM battery. Ambient temperature 19C
No foil wrap on the module, no spring modifications.
Solarforce runtime is quite weak considering the fact that current output is around 2A (I guess) with a 18650cell, it's a buck driver with input an of 1.7A measured at 4.2V (current output is greater than input current on buck driver).
Difference in output is drastic as you can see, about 2 times brighter, but that for a more advanced person is to be expected based on the fact that XP-L V4 is rated being about 14% more efficient then XM-L U2 and there is also a higher current on the CUXM2. No matter of those differences the runtimes is still weak on the Solarforce. Probably lots of voltage drop in the driver and not so great efficiency (this is speculation on my part).
The participants:

Thanks. Although in defence of the SF, it’s regulation is much flatter than the IOS.

I have the earlier SF XM-L U2 drop in, I didn’t test it, but visually I found it to not hold high output/regulation very well on a single li-ion. But on 2 c li-ion it not only is brighter, but maintains regulation very well.

Are you able to test the SF drop in with 2 x 18350/16340 or 2 x 18650 as a comparison?

It is flatter just because it has a a lower current output using a buck driver. So I do not find that so special as it is much lower in brightness both due to the driver and emitter used. A you can see the runtime is that I find surprisingly weak, and as i said before the fact that the CUXM2 is brighter is not a surprise.

As for the CUXM2 not being flat, that is going to depend on cells, I would suggest cells that are able to keep high voltage under 3A. Still yo ustart at 80 then the after 1 hour the high mode is done at 60, that is at 30% drop across an hour and even then will not be as dim as the Solarforce.

I can try do to a 2 cell test on the SF module, but I expect the regulation to be flat since the driver is a buck one (otherwise would be quite absurd). The only interesting thing would be how much more relatively brighter it is compared to the 1 cell runtime.

That is a huge difference in efficiency! If you see the surface area under the graphs as the total emitted light on a battery charge, a quick-n-dirty-from-screen measurement tells you that the IOS drop-in gives almost 1.9 times as much light for your energy.

I have used my precise power supply to power the SF module while inside the L2T host to plot a graph while inputting 8.4V down to 5.5V thus simulating 2 very powerful Li-ion cells in series.
The results looks good as expected in terms of regulation.
I use 5 secons for each voltage level down until 8.0V then I used 10 seconds for each voltage level down until 5.5V
Example
8.4V - 5sec
8.3V - 5sec
...etc
7.9V - 10sec
7.8V - 10sec
7.7.V -10sec
..etc
Note: This is not a runtime test just a test of how flat the light out is using high voltage. (2*4.2V)

We start at number 35 then we go down to 29 that is 19% drop in output by the driver from the start until the end, of course output is affected by heat at the LED a bit, but not a whole lot, ambient is pretty low 19C.

I am 100% sure garbage cells would not do a good job even on the Solarforce module, do do not buy them just because the amps are not high, that is not going to work.

The 35 value of relative brightness is directly comparable with the 1 cells test runtime. that indicates using higher voltage does not offer a higher output brightness. Values are not absolute, take or give 5% of anything you want, to account for the lightmeter accuracy.

Great test! You just helped me to decide which one to get: the XP-L.

To be precise, the XP-L V6 2C.

Thank you :beer:

I assume that the SF XM-L2 dropins are unable to maintain regulation as long as what you showed here with the XM-L, due to the higher Vf.

The CUXM2 regulation would be worse if I'd use for example NCR18650A, even if would apparently give more runtime, yeah it would but at a lower brightness. There's more to a cell than capacity, and most importantly that is voltage under load (of course for the exact load you need)

The Sanyo UR18650FM is a very good cell, still after all this time.

Comparing cells with and without protection circuits is not fair IMO.

I would suggest to use the more popular NCR18650B.

My point is that the regulation curve would be different using another cell, and I gave and example of another cell. Does not matter if it is protected or not, that is a cell what some people still have and use, it is a fair comparison for what I am trying to say, that another cell would results in another runtime curve.

This is for you RacerR86, to see the Sanyo UR18650FM are good cells even compared to the popular 3400mah cells.