Powering an XR-E to its max. Driver options?

I was looking at the

5-Mode P7 LED Driver Circuit Board for Flashlight (17mm / DC 5.0~17V) very closely.
- Voltage: DC 5.0~17V (1 x 18650 or 2 x 16340 battery)
- Current: 2000mA
- Diameter: 17mm
I have a 2 x 18650 version of the Solarforce Skyline 1 that Id like to drive to its maximum potential without insta-flashing the LED. I dont care if it reduces emitter life to a few hundred hours... LED's are cheap and easy to replace.

Ive read that some guys run their XR-E's around 1.9-2.1A and get good results if the host is well heat sinked. Has anyone experimented with these higher drive currents with an XR-E? Ideally, Id use a 17mm driver that pumps 1.7-1.8A but havent found one that operates at 8.4V in that size.

I have a light that runs an XR-E at 1.7A with fantastic results (Lunahunter MRV) but the driver is designed to run at 4.2V.

Take a look at the video, how perform an xr-e with an avarage, good cooling at 0.5A-1.9A.

In my opinion, that above 1.5A is waste of energy. There is no significant light increase above 1.5A.

I used a simple light box, so they are relative lumen values.

0.5A~ 148

1A~ 237

1.1A~ 248

1.2A~ 258

1.3A~ 267

1.4A~ 273

1.5A~ 278

1.6A~ 282

1.7A~ 284

1.8A~ 284

1.9A ~282

Isn't that a circular argument because at the higher current the battery will also sag more quickly?

There’s not a whole lot of point to exceeding 1A. It will end up throwing about 5% further at 1.8A than at 1A.

For a 40 lumen gain over 1Amp it's not worth the trouble.

Thanks for your analysis. Were you able to verify the BIN and source of where you aquired it? It appears that only the real R2's are capable if providing significant lumen increases above 1A.

Related CPF article while testing a Cree 7090 XR-E R2 WG:

Im sure output will drop off quickly as junction temperatures increase, although the host has a fair amount of chunky aluminum in the head to sink the emitter. The driver I linked at the top is new and untested, but its the closest match I could find for a cheap 17mm driver spec'd for 2 x lico's to output somewhere between 1.5-2.0A. I dont want to go below 1.5A but Im also not to fond of driving it at 2A. Does anyone have any other driver suggestions?

As for battery sag, Im not to worried since 2A will barley be pushing 2 x 18650's for at least 1.5 hrs.

Of my MRV collection, there is a very distinct difference in luminosity and range (as seen by my eyes) for the ones that run at higher drive currents. One that runs at 1.7A is definitely brighter than another running at 1.35A, which is brighter than one running around 1.05A.

That's a different story all together, for 140lumen gain I might try bumping my R2 gun light with an aspherical up to 2 amps. Thanks

Gezuz....flashpilot, do you have 10 x Lumapower MRVs?!

It's a Q5, but who knows.

Very interesting. I don't have R2. Is there something difference between Q5 and R2 (in structure)?

Actually if you do direct LED to copper bonding, you can overdrive by quite a lot. Won't get you a lot of efficiency, but who cares.....20 to 30lm/W is still cool and in the halogen eff realm.

Lamba is aiming for 150k cd with a XM-L. 73mm head only, no hocus pocus. But I think 135k is finalised, and you know for enthuasists lights this is guaranteed result.

Also already proven countless times by the DRY + ice or dry-ice cooling. Even in tropical countries it's pretty sweet.

I have a feeling that there’s no different in ANY recent XR-E. I think any EZ900 die XR-E is just labeled as a Q5 because they will do at least that well. I don’t think Cree bothers separating out the jems from the duds simply because they are not the latest and greatest and the benefit to properly binning them is minimal.

Just a thought . :)

I have the EZ900 die R2 1.5A pill coming from Dereelight. Quite cheap just over 20 bucks shipped. I mean yeah it's not budgetlight realm but heck it has no PWM. LOL!

Supposed to get 90k on the DBS. Saabluster got 107k. Hmmm....

Ive had a lot more than 10 MRV's when I used to mod and resell them. They do great when you mod with a 8 x 7135 boards and remove 3 of the 7135's, netting 1750 mAh. Its to bad these boards max out at 6V. I agree that getting a great performing XR-E is a crap shoot with Chinese sourced LED's. Ive had enough of them to know that later iterations seem to be much more consistent in delivering high output than earlier versions. I have not noted any real advantages of the EZ900 over the EZ1000 dice.

I think the XR-E R2's incorporated the latest in technology (phosphors and materials) but were based on the same EZ900 dice as all the other second generation XR-E's. There was yet another test I read a while back that confirmed how the R2 outgunned previous iterations of the XR-E when driven @2A, but I couldnt find it. Crees data sheet clearly shows the R2 as the most efficient, but Im sure we all know that.

There are many situations where the XR-E's potential far reach and sheer pencil beam have an advantage over any other emitter. Especially when trying to spotlight something at a great distance without losing your night vision or the often unwanted distraction of a bright spill beam.

Ive followed Lambda's creations closely and recently sold one of his 3 x XML 2D (hydra) mags. Ive noted that he gets many more sustained lumens when bonding the SST90's directly to copper. The latest hydra 3 x XM-L also has the emitters bonded directly to copper. DX also sells an XM-L bonded to a copper star, although I wonder if there are really any significant advantages with an XM-L driven to 3A and mounted on copper versus aluminum; the XM-L is said to have outstanding heat convection transmitted through its large ceramic pads. Most (if not all) other power led's are not as efficient in transmitting heat through their mounting pads, thus the advantage of mounting those others to copper.

That is a good price for quality. Got a link to the one you've ordered? I wonder if these buck/boost regulate on 4.2V and if they also run regulated at 8.4V?

I almost ripped apart my Olight M20 warrior for my project because it has such a great driver, but its to large to fit the intended host. DARNIT!! The M20 would be a cool little light with the guts from a 3 mode trustfire p60 XML transplant.

They have a lot of options. 1 mode. 3 modes. single cell voltage, multiple cell voltages. It's a bit confusing. hehe... Yes it is regulated with 2 cells. XR-E has pretty high Vf so it does not stay regulated much esp at 1.5A....even with the best cells.


Absolutely no PWM detectable on the DMM.

Which driver is this? Link, please...

Or is a butchered 8*7135 the only way?

Thanks for sharing that experience, FlashPilot, as I have an UltraOK (MRV clone) running the XR-E Q5 at 1750mA as well, and was a bit concerned.

With regard to spotlighting at a great distance: yes, that's the issue I have with my TrustFire X9, even with the underdriven XM-L at 1.8A. It seems that air here is never dry enough to eliminate the "noise" from the beam trial in my path of vision. Holding the light at arm's lenght does increase on-target contrast considerably, but is impractical of course.

I was thinking of replacing the XM-L emitter with an XR-E R2, but, since the X9 reflector is 47mm wide (internally), an XP-G R5 might be a better lux/spot_size compromise for 100-200m use.

I don't own a light with an R5 to try, so any advice wold be much appreciated.

When spotting huge distances, you are supposed to be using a telescope (an APO refractor one would be nice!) or at least a pair of binos. Else how would HIDs with huge reflectors and those searchlights with hundreds of million candelas work? hehe....

Here's a nice one..