ttran97 "290 lumen" R2 dropin

I don't have permission to post in the Flashlight Reviews section, so hope this posting is appropriate here.

Here are measured run times for the R2 dropin which has been recently sold at the Marketplace





There's obviously no real regulation when running with an 18650 cell, but the apparent drop in light level isn't that much over the very long run time. The steep drop when running with the RCR123 cells looks a lot like profiles I've seen when the LED heat sinking is inadequate, but it might be due to the regulator.

Shipping was super-fast, the beam is very nice looking, and at $20 I feel it's a good value. The reflector is pretty deep, which results in a smaller spill diameter but brighter hot spot than some other lights with the same diameter reflector.

c_c

Did the head get hot during the RCR123 test? I would probably repeat the runtime but put some aluminum foil around the drop-in. Put it in little by little until you get a tight fit with the head. Then repeat the test and see if the drop-off improves and the heat transfer to the body increases. These drop-ins are biasing the LED at 1A and the regulator efficiency is in the paltry 60's percent. But for $20, it's still pretty darn good for what you get.

As far as 18650, the circuit is a buck and between the FET, inductor, and sense resistor, you're probably looking at a minimum of 0.5 ohms series resistance so you're in direct drive mode with a relatively large series resistor to rob the LED of any current it can get.

How does this compare to the legendary R2 at DX at a much lower price $12.40?

CM said:

Did the head get hot during the RCR123 test? I would probably repeat the runtime but put some aluminum foil around the drop-in. Put it in little by little until you get a tight fit with the head. Then repeat the test and see if the drop-off improves and the heat transfer to the body increases. These drop-ins are biasing the LED at 1A and the regulator efficiency is in the paltry 60's percent. But for $20, it's still pretty darn good for what you get.

Click to expand...

There was a fan blowing on the light while the test was being run, resulting in no noticeable temperature rise of the outside of the light. I'll let someone else do the aluminum foil tests.

As far as 18650, the circuit is a buck and between the FET, inductor, and sense resistor, you're probably looking at a minimum of 0.5 ohms series resistance so you're in direct drive mode with a relatively large series resistor to rob the LED of any current it can get.

Click to expand...

Yes, it's quite obviously a buck regulator. The long run time at moderate output shows that it's reasonably efficient when in an effective direct drive move with the 18650.

c_c

TDKKP said:

How does this compare to the legendary R2 at DX at a much lower price $12.40?

Click to expand...

No idea, since I don't have the DX unit. Maybe someone else has tested both and will post a comparison. One striking difference is delivery time, though. The tested unit got here just a couple of days after I ordered it. I haven't ordered from DX for some time, but when I did, it nearly always took about three weeks to get here.

c_c

Thanks for the graphs and running the tests, C_C! I'm sure this information will be very useful to everyone! :thumbsup:

I have one of the DX's, which I definitely feel is a super drop-in for the price (and not far off from the best at any price) so when I get Tung's I'll be happy to do a comparison. Realistically this will take me a few weeks due to my travel schedule...

I got curious about the heat sinking and discovered that the dropin doesn't really properly seat in my host. The top aluminum piece appears to be a bit longer than optimum, preventing the upper part of the reflector to contact the aluminum of the head. The result is very little thermal contact between the head and flashlight body -- the only contact is along a very narrow rim at the bottom of the dropin. If I get some time, I might try shortening up the reflector part to see if that would make it seat better. But even if the heatsinking was made perfect, the improvement would be less than 30%, judging by the output graph. That wouldn't be enough to visually notice. The fit is just fine with regard to making the light function.

The situation might be different in different hosts. Mine is an old Huntlight FT01PJ, which might not be very typical. I guess what I've shown is that even very poor thermal contact between the dropin and host results in output which is visually just as good as perfect heat sinking.

c_c

I would have been quite disappointed if I bought a "290 lumens" R2 dropin, and it only gave 202 or 125 lumens depending on battery. This is far less light than many Q5 modules are putting out.:shrug:

jenskh said:

I would have been quite disappointed if I bought a "290 lumens" R2 dropin, and it only gave 202 or 125 lumens depending on battery. This is far less light than many Q5 modules are putting out.:shrug:

Click to expand...

Emitter lumens and torch lumens are two different things. The 202lm measured is almost exact if you take into consideration a typical 30% loss. It actually ends up +1 lumen, which is fantastic if you ask me. Most of these knock-off brands won't even come close to the stated rating.

jenskh said:

I would have been quite disappointed if I bought a "290 lumens" R2 dropin, and it only gave 202 or 125 lumens depending on battery. This is far less light than many Q5 modules are putting out.:shrug:

Click to expand...

Probably not. It's just less than a lot of modules are claimed to be putting out. Think for a moment about how many lumens you should be able to get from a Cree LED. The highest bin number shown in the data sheet on their web site is the Q5. It's spec is 107 - 114 lumens @ 350 mA. Another graph shows that at the maximum rated current of 1 amp, the output is 2.2 times the 350 mA output. So we could expect an absolute maximum of about 250 lumens from the best Q5 we could get, when running at its absolute maximum rated current. The difference between Q4 and Q5 is about 7%, so I'd expect maybe another 7% or so for Q5 to R2. That makes the R2 capable of 268 lumens top, if you could magically keep the die at 25 degrees C. I'll bet dollars to donuts that at 1 amp in most lights, the junction temperature is more like 100 C, which reduces the output by 20%. So now the very best R2 is putting out 215 lumens. Even with no reflector or lens -- that's the mythical "bulb lumens" you hear people talk about. Compare this with the claims you see, and try to imagine where all those extra lumens come from. I'll tell you where: the marketing departments, not from the LEDs.

Add a reflector and lens, and you lose another 25%. Now you're down to 161 lumens. You can conveniently make the heating loss go away by specifying the "initial" lumen value -- that is, the lumen output the instant you turn the light on. That'll bring you a 20% or so improvement in your spec without any increase in the actual light output. Then you can have your full 268 "bulb" lumens and about 201 lumens out the front of the light. Guess what -- that's just what I measured. Initially, that is, until it heated up.

You can do a bit better by minimizing the reflector and lens losses, pushing the LED current beyond the specified maximum, and doing an extra good job of heat sinking. But not an awful lot better.

A lot of lumen "specifications" are pure fabrications with the sole purpose of selling flashlights. And it works.

The numbers I posted are from strictly amateur equipment - a Quickbeam type light box - so it has a lot of room for error. But at least the numbers weren't just made up, as so many "specifications" are. I calibrated it using some bare SSC and Cree emitters from known bins, under the assumption that the manufacturers' specifications for the emitters are accurate. And it does produce realistic values for lights with presumed accurate specifications, for example the Novatac 85P which measured 90.9. So I think it's in the ballpark anyway. I have a pretty fair collection of bright lights with Q5 emitters, and they all top out right about at 200 lumens. This dropin produces just about as much total light as any I've seen.

c_c