High Output R2 P60 drop in .

old4570

Flashlight Enthusiast
Joined
Feb 15, 2009
Messages
2,900
Location
Melbourne - Australia
Been playing with drivers N stuff .

And this driver is quickly becoming a favorite for me .

Combined with this pill , Im getting some very nice high output R2 pills for my flashlights . So for you DIY guys , here is a combo worth giving a try .
Just in case your wondering , the one I did today is doing 1.8A at the tail , and has become my highest output R2 to date , and is actually coming close to my Mini L2 MC-E .. 16,000Lux for the R2 and some 17,500Lux for the MC-E in my lightbox .
:thumbsup:
 
Last edited:

tx101

Flashlight Enthusiast
Joined
May 17, 2008
Messages
2,357
Location
London UK
How hot does the dropin get ?
Any idea how much current at the emitter ?

I have been looking for a driver that outputs at least 1500mA with two li-ion cells + 17mm
 

old4570

Flashlight Enthusiast
Joined
Feb 15, 2009
Messages
2,900
Location
Melbourne - Australia
8.4v or 6v .. ?

The driver is up to 4.5v from mem , so mainly for 1 x CR123A 3.7v Rechargeable or a 18650 or sim batt .

Heat , not bad at all , gets warm not hot . My MC-E gets seriously warm / hot after 5 minutes , but the R2's are much better in this regard .

Im not aware of any good 8.4v drivers [ 17mm ] that can do more than 1A .
I had one on order for over 2 months and gave up [ 8.4v driver ] .

I need to make a 18650 batt holder with some gator clips so I can measure current to LED . A must do project that Im working on . Tail readings are so so .
 
Last edited:

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
Based on the description, it isn't a driver. It's a direct drive PWM controller.
 

old4570

Flashlight Enthusiast
Joined
Feb 15, 2009
Messages
2,900
Location
Melbourne - Australia
Based on the description, it isn't a driver. It's a direct drive PWM controller.

Yes , so hopefully most of that 1.8A is making it to the LED , :thinking:

CREE R2 (114-120 Lumens @ 350ma) Pure math = 586 But I dont see that .
Controller Loss + Sag + Lens loss etc , I guesstimated 370Lumen out the front .

I need to make a 18650 batt holder so I can measure current to LED , so far its a lot of guess work .

A lot of fun though ! :candle:
 

tx101

Flashlight Enthusiast
Joined
May 17, 2008
Messages
2,357
Location
London UK
I need to make a 18650 batt holder with some gator clips so I can measure current to LED . A must do project that Im working on .

I use this rig for charging D and C cell li-ions as well as taking measurements
at the emitter


DSC01199.jpg
 

CampingLED

Enlightened
Joined
Nov 13, 2007
Messages
612
Location
South Africa
Heat , not bad at all , gets warm not hot . My MC-E gets seriously warm / hot after 5 minutes , but the R2's are much better in this regard .

My experience of these drop-ins are that the heat transfer from driver to flashlight body is quite poor if you do not wrap a lot of aluminium around it. Perhaps you should leave it on for some time and remove the pill to check the actual heat.
 

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
If you really are delivering 1.8A to the emitter with that pill, you will no doubt rapidly degrade the LED's output from excessive heat if you use the module for any sort of sustained on-time. You have to be careful. Temp of the flashlight or even the heat sink is not necessarily the same as the temp of the junction, and that's what dictates the life of the LED.

I also don't follow how you estimated 586 lumens at 1.8A forward current. The Cree datasheet goes up to only 1A for relative luminous flux vs forward current, but extrapolation to 1.8A gives a relative luminous flux of about 3.5. Definitely not ~5X.

The R2 bin ranges from 114-122 lumens at 350 mA forward current. At 1.8A forward current, the estimated, extrapolated output (at the LED) is about 400-427 lumens. Subtract the usual 20-30% for reflector and window losses, and you get around 300-320 lumens out the front.

In reality, I'd be very surprised if the 1xRCR123A or even the 1x18650 can hold 4V or so and deliver 1.8A. More likely at that current load, the Li-ion will sag to its usual 3.7V or 3.8V and end up delivering the usual 1000mA of forward current or a little more depending on the Vf of the LED.
 

old4570

Flashlight Enthusiast
Joined
Feb 15, 2009
Messages
2,900
Location
Melbourne - Australia
My experience of these drop-ins are that the heat transfer from driver to flashlight body is quite poor if you do not wrap a lot of aluminium around it. Perhaps you should leave it on for some time and remove the pill to check the actual heat.

Al-foil wrap is part of any P60 drop in I do / IE all my P60 host/pills are al-foiled .

I dont run them , not foiled .

:grin2: Like I said , Not even close to a MC-E P60 drop in , when it comes to hand warming .

My MC-E in a Solarforce L2 18650 body @ 2.35A runs HOT ! , after 5 minutes even the tail cap is getting HOT .. This is hand held .

Not one of my R2's comes close to generating this kind of heat .
Even with the MC-E in my Mini Body 1 x CR123A , runs much hotter than any R2 I own ... I plan on doing a SSC P7 - P60 drop in as soon as one of the LED's ordered arrives .. I expect it to run Hot .

To me , R2's run warm , when compared to say MC-E .
 

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
It's not getting hot because it's probably not running at 1.8A to the emitter, except perhaps at the start. It is most likely direct driving at the usual 1A or maybe 1.2A. Supposedly, the DX 11836 driver is the same one as used in the DX 6090 drop-in. When that driver is run at 2xLi-ion, it is about 70% efficient, and thus about 1.5W is wasted as heat. You don't have that driver inefficiency with direct drive. Probably all of the heat being generated is coming from the LED.

At what point during the run time did you make your 1.8A tailcap current measurement? Immediately upon flashlight turn-on?

For comparison, what other lux values have you measured for some well-known lights? What was the lux reading for the R2 before mod'ing? Can you describe your lightbox setup (like the one from flashlightreviews.com?)?

Based on you previous post for your Solarforce Mini L2 MC-E, you were driving that LED at about 1.3A, or about 325mA per core. For a K-bin MC-E, you are probably getting about 290 lumens out the front. So the output for your direct drive R2 doesn't really seem to be overdriven to the degree suggested by a 1.8A drive current.

Edit: Ok, now I see you just posted that your L2 MC-E draws 2.35A (I assume at the tailcap). An MC-E at that forward current (I'll ignore any driver inefficiencies and just assume that the 2.35A figure is also the If) could give roughly 450 lumens out the front. If we use a simple ratio of 16/17.5 to estimate the output of the R2, we get about 410 lumens. That doesn't square with the estimated OTF lumens for an R2 at 1.8A forward current, probably because the actual MC-E forward current wasn't 2.35A. If we assume a driver efficiency of 70% (a typical value for these cheap DX/KD drivers), we get a forward current of 1.6A. That puts the estimated K-bin MC-E OTF output at around 350 lumens, which is getting back to the magnitude of the calculations I made above. Also, I assumed that all of the 1.8A that you measured at the tailcap goes to the R2. In reality, there is always some finite resistance and thus additional voltage drop associated with the springs, wires, contacts, etc. At that level of Vf/If, even a small change in voltage results in a large change in current (you are operating in the steepest part of the LED's V-I curve). If the voltage drop from various contact resistances is 0.3V, that can mean a forward current of around 1.0A instead of 1.8A.
 
Last edited:

old4570

Flashlight Enthusiast
Joined
Feb 15, 2009
Messages
2,900
Location
Melbourne - Australia
I cant concentrate - its Midnight ... MC-E has the same Driver/Controller - so if its Direct drive , loss is ??
Im using the same driver/controller on 2 x R2 and one MC-E .. I havent checked forward current , cos I dont have 4 hands .

Check my CR123A batt test , starting current over 1.8A with many diff CR123A .

IN anyway ,
 
Last edited:

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
IMO, it would be better to graph watts vs time/voltage. That would show you when the driver falls out of regulation and at what voltage.

By eyeball, it looks to me like the driver falls out of regulation right at the start (3.9V-4V), which seems consistent with either direct drive or a buck driver that has the usual 0.5V-0.7V voltage overhead to run in regulation (and thus, it is also running direct drive).

Were you testing a real driver in your RCR123A tests, or using that same KD direct drive PWM controller?

Sure, 1.8A input at the very start is entirely consistent with the initial voltage that a Li-ion can deliver -- around 4V. But an RCR123A cell is going to sag very quickly to 3.7V or so. I'm surprised that your voltage measurements are so high. Did you measure voltage under load or remove the load and measure the voltage (e.g., put your DMM in Volts mode, remove the tailcap, and measure the cell voltage by touching one probe to the cell and the other to the flashlight tail)? And then when you measured current, did you remove the probes, switch your DMM to some high amperage scale, and then make your measurement? If so, I think you are measuring bogus data. You aren't measuring the current draw at time t. You are measuring the current draw of a cell that is allowed to recover for some short amount of time back to some artificially high voltage.

Also, your current loads are WAY TOO HIGH for a standard RCR123A. You are far exceeding the 2C recommended discharge rate. You are going to kill your Li-ions very quickly.
 
Last edited:

bigchelis

Flashlight Enthusiast
Joined
Oct 30, 2008
Messages
3,604
Location
Prunedale, CA
Yes , so hopefully most of that 1.8A is making it to the LED , :thinking:

CREE R2 (114-120 Lumens @ 350ma) Pure math = 586 But I dont see that .
Controller Loss + Sag + Lens loss etc , I guesstimated 370Lumen out the front .

I need to make a 18650 batt holder so I can measure current to LED , so far its a lot of guess work .

A lot of fun though ! :candle:


In my search for the highest lumens possible with a R2 I had MrGman test a couple of R2's well sinked in a Maglight 1D custom from Nailbender.

Using the Cree 7090 XR-E LED. WC R2 LED in 3 formats here are the out the front lumens MrGman got.

1. 1D Magligth with XR-E at 1.4A driver = 170ish out the front. Maglight custom made by Nailbender.

2. 1D Maglight with XR-E with 1A driver =180 ish out the front.

3. Tiablos A10 with XR-E w/ 1A driver = 245 out the front.

I expected the 1.4amp R2 to give well into the 300 lumens out the front, but the extra heat was making the R2 perform very poorly. Even at 1A the R2 made still under 200 lumens. Naibender thinks the hight pedestral heat sink might in the Maglight build is not enough to cool the LED effectively, but I need it this way because in this format I can add an aspheric and it focuses nicely. Testing was done with LOP reflector.

The Tiablo A10 at 1A driver did 245 out the front, but it does have a dedicated heatsink.

From MrGmans tests I can say that over 1A of Current on an R2 will not make it brighter, but if you have a custom dedicated heatsink it might. From helping MrGman test many lights I can say that paper math lumens are always way higher than what you actually get. Even The Malkoff stuff he tested was way way under, but still good quality stuff. EX: The Malkoff MC-E direct drive was thought to make over 1000 out the front lumens at 4.5A at the tail, but it made less than 450 w/ I IMR 18650.

Good luck and hope this helps.

Jose
 
Last edited:

kramer5150

Flashaholic
Joined
Sep 6, 2005
Messages
6,328
Location
Palo Alto, CA
Can I ask where you are getting your R2-WC emitters? I am having a hard time finding a source. I use Q5-WCs from shiningbeam (top notch service IMHO). can you shoot me a PM?

I have had questionable results driving the XR-E-R2 at 1400mah, in a P60 host. I can tell a slight but noticeable dip in output as the LED warms up. My particular R2 (DX:11836) seems to be fading too. It doesn't appear to be as bright overall as it once was... prior up up-ing the current. So, my subjective opinion is that the LED may dim/degrade over its lifespan if driven this hard and this warm.

I also killed an R2-WC emitter in a Home Depot spotlight. It was running direct drive with a current limiting resistor from 3x NIMH-AA... but that one has even less thermal cooling than a 6P host. So I kind of knew it was going to die at some point. Its now very dim and blue tint. When previously it was very close in output to a malkoff M60....
dscn0548mp1.jpg

dscn0545mi1.jpg


I have no way to quantify what I observe though... It could all be a placebo without any way to really measure things.

But kudos to your efforts:twothumbs

My fave driver thus far has been this one.... It can easily be modded for 1400 or 1000mah by de-populating one of the 7135 drivers. Its mode selectable 10/50/100%, so I generally use it at 50% (700mah) to help preserve the LED. it uses PWM for the low modes, but its a high enough frequency such that only running water appears choppy.



I have had reliability issues with this one here... I have 3 of these, the first one just up and died, powering a Q2-WC. So I am hesitant to use the other two. its a NICE driver while it worked, the variable output, and PWM-free low modes are really nice features. StefanFS did some efficiency measurements a while back and its a top performer in that regard too.
166hsnmod6.jpg
 
Last edited:

bigchelis

Flashlight Enthusiast
Joined
Oct 30, 2008
Messages
3,604
Location
Prunedale, CA
Nailbender has a couple. I think KD builds custom boards for him too. He has a personal contact that takes his orders over at KD.

Jose
 

metlarules

Flashlight Enthusiast
Joined
Jan 30, 2008
Messages
1,027
Can I ask where you are getting your R2-WC emitters? I am having a hard time finding a source. I use Q5-WCs from shiningbeam (top notch service IMHO). can you shoot me a PM?

I have had questionable results driving the XR-E-R2 at 1400mah, in a P60 host. I can tell a slight but noticeable dip in output as the LED warms up. My particular R2 (DX:11836) seems to be fading too. It doesn't appear to be as bright overall as it once was... prior up up-ing the current. So, my subjective opinion is that the LED may dim/degrade over its lifespan if driven this hard and this warm.

I also killed an R2-WC emitter in a Home Depot spotlight. It was running direct drive with a current limiting resistor from 3x NIMH-AA... but that one has even less thermal cooling than a 6P host. So I kind of knew it was going to die at some point. Its now very dim and blue tint. When previously it was very close in output to a malkoff M60....
dscn0548mp1.jpg

dscn0545mi1.jpg


I have no way to quantify what I observe though... It could all be a placebo without any way to really measure things.

But kudos to your efforts:twothumbs

My fave driver thus far has been this one.... It can easily be modded for 1400 or 1000mah by de-populating one of the 7135 drivers. Its mode selectable 10/50/100%, so I generally use it at 50% (700mah) to help preserve the LED. it uses PWM for the low modes, but its a high enough frequency such that only running water appears choppy.



I have had reliability issues with this one here... I have 3 of these, the first one just up and died, powering a Q2-WC. So I am hesitant to use the other two. its a NICE driver while it worked, the variable output, and PWM-free low modes are really nice features. StefanFS did some efficiency measurements a while back and its a top performer in that regard too.
166hsnmod6.jpg
I have a question reguarding the second board(NJG-18). It's suppose to have a drive voltage of .8-7v.But this link says "if the output setting is exceeded by the input it will go in to direct drive". http://e-lectronics.net/boost-driver-board-input-voltage-8v7v-njg18-p-422.html Does this mean that it will run at 4.2v on a li-ion? Won't that kill the led?
 

old4570

Flashlight Enthusiast
Joined
Feb 15, 2009
Messages
2,900
Location
Melbourne - Australia
The driver / Yes - powers the MC-E in the batt test .
Also in 2 R2 pills ..
https://www.candlepowerforums.com/threads/231040

This R2 / it ran 12000Lux with 2 x CR123A 8.4v fresh cells .
After driver swap , 16000Lux , or + 30%

So 30% increase for ?? increase in power to led .

The average ligh output in my lightbox for R2's is 11000 to 13000Lux give or take .

I stand corrected , and without knowing actual current to LED , its all guesswork , but the 30% boost is not .. What it actually = , oh well , its still the brightest R2 I own .

But No R2 is going to be exactly the same as another , and My R2's are all over the place for current draw , from 1.2A all the way up to 1.8A with fresh 18650 batts . One of my L2 fact R2's draws 1.75A at the tail , the other is about 1.5A , and both sit on 13000Lux +

Oh well , hopefully ill have a Diamond Dragon to play with shortly , and some SSC P7's .

I still say for the money , its a combination worth trying .
 

old4570

Flashlight Enthusiast
Joined
Feb 15, 2009
Messages
2,900
Location
Melbourne - Australia
IMO, it would be better to graph watts vs time/voltage. That would show you when the driver falls out of regulation and at what voltage.

By eyeball, it looks to me like the driver falls out of regulation right at the start (3.9V-4V), which seems consistent with either direct drive or a buck driver that has the usual 0.5V-0.7V voltage overhead to run in regulation (and thus, it is also running direct drive).

Were you testing a real driver in your RCR123A tests, or using that same KD direct drive PWM controller?

Sure, 1.8A input at the very start is entirely consistent with the initial voltage that a Li-ion can deliver -- around 4V. But an RCR123A cell is going to sag very quickly to 3.7V or so. I'm surprised that your voltage measurements are so high. Did you measure voltage under load or remove the load and measure the voltage (e.g., put your DMM in Volts mode, remove the tailcap, and measure the cell voltage by touching one probe to the cell and the other to the flashlight tail)? And then when you measured current, did you remove the probes, switch your DMM to some high amperage scale, and then make your measurement? If so, I think you are measuring bogus data. You aren't measuring the current draw at time t. You are measuring the current draw of a cell that is allowed to recover for some short amount of time back to some artificially high voltage.

Also, your current loads are WAY TOO HIGH for a standard RCR123A. You are far exceeding the 2C recommended discharge rate. You are going to kill your Li-ions very quickly.

Yes cells were removed for V measurement , right after the 5 minute run , like the 1st 30 seconds , The voltage was stable for about 60 seconds then began to rise , Voltage was not the main thing , Amps and run time were .

I usually run the Mini MC-E in Lo or medium mode , hardly ever in Hi , and If I do so its only for 30 seconds or so .

Yes , a very stressful test for the batteries :
 

Justin Case

Flashlight Enthusiast
Joined
Mar 19, 2008
Messages
3,797
I still say for the money , its a combination worth trying .

For 16340-size cells, the only way I would overdrive the R2 at the level you favor is to use IMRs. For regular LiCoO2 chemistry, IMO you are going to kill the Li-ion cells very rapidly by driving them at 3C-4C discharge rate. I also think that high junction temperature is going to kill your R2 LEDs very rapidly.
 
Top