DIY Mini Mag Mod Guide

[jar3ds]

oh gosh... i'm getting so sick and tired of all the 'heatsink' nannies... who cares! :nana:

seriously... it all practical purposes... most people turn their lights off... admire how bright it is... then turn it off :ooo:...

if your light is getting hot it means its working! :devil:

Thanks for the DIY Mini MOD Guide! lovecpf

 

[Illum]

yeah...while the heatsinking may pose as an issue its still much better than the stock minimag LED.

besides...CREEs are cheap :whistle:
Given the real-estate in a minimag...the only ability to effectively heatsink it would be inserting a brass slug or what not in the body of the light and use a single 14500, but that won't be a "simple" mag mod :grin2:

 

[Fender]

I want the MM mod! :D

 

[linty]

"I want the MM mod"

woot!

 

[old4570]

Great Job!! :thumbsup:

I have a Brinkman (MiniMag clone) that I wanted to upgrade for a long time. Where did you order the emitter and how much was it? I would like to order a few and do this upgrade on a few lights.

Thanks Again!

Cree's = Many sources
I usually get mine from Dealextreme or KaiDomain , you can also head on over to the CPF Marketplace , and check the links/sponsors .

IQ tail switch

And thank you , I figure lots of people have Mini Mags in the draw doing nothing , and if you can get them up and running for $5 , why not !
I've learned a lot here at CPF [ still lots to learn ] and this is just a little Pay Back ... lovecpf

 

[old4570]

Nice DIY How-to-Mod.... I ordered a Q3 from SB just for my MM. I'm going to machine a small Al slug for sink. What sort of draw can NiMHs or 'loops source?

Im not sure I understand the question ...

Im guessing your wondering how much current they can provide , and the answer is around 1.7A+ for a single cell with a Cree XR-E ..
Now thats only in my experience , Hardcore AA users may know better .

 

[Pekka]

Simple yet effective, me likes :clap:
Oh, and I want the MM mod!

 

[dealgrabber2002]

Thank you for this great gesture.

I want to be in the raffle as well.

"I want the MM mod"

Thank you once again.

 

[dubweiser]

This is great!! Exactly what I was looking for.

oh yeah.. I want the MM mod!

 

[DoctaDink]

Thanks for posting the Mod. Definitely looks like one to try.
And...yes, I want the MM Mod.

Thanks again.

 

[nanotech17]

you could use the Fraen reflector to get a better beam.

 

[J.C.F.]

how nice
I want the MM mod!!!!!
cool and thanks.

 

[zozo]

I want the MM mod.

Please.

 

[gunga]

Clever mod.

I'm surprised it can work with 2 batteries since the vF is usually around 3.2 V etc?

I guess it can work with 2 alkalines or lithium, but 2 nimh should not work (or is very dim)?

Oh, btw.

"I want the MM mod"

:devil:

 

[3D black mag]

Nice work and good DIY guide.


"I want the MM mod" :twothumbs

 

[raven1911]

I want the MM mod.

Please.

 

[Justin Case]

Clever mod.

I'm surprised it can work with 2 batteries since the vF is usually around 3.2 V etc?

I guess it can work with 2 alkalines or lithium, but 2 nimh should not work (or is very dim)?

Oh, btw.

"I want the MM mod"

:devil:

Vf is not a single value. It comes as a pair -- Vf/If and spans a range of values depending on drive current. The usual spec for Vf is given at 350mA drive. For a Cree P4 (haven't measured a Q5, although I have a loose one sitting around somewhere), I measured about 150mA drive at roughly 3V in. But at 2.8V, the drive current fell to about 50mA. Each LED will have its own Vf/If curve, but if we assume as a first cut that the Q5 has similar behavior, you can then refer to the relative luminous flux vs drive current curve in the Q5 datasheet to estimate output.

Fresh NiMHs should start out about as bright as 2xalkaline, since NiMHs can be charged to around 1.45V per cell. But the output will probably fall quickly.

At these relatively low drive currents, the LED probably won't heat up drastically, especially if you run the light for brief periods only. I can run my loose Q5 (emitter and substrate sitting on a 14mm circular base board only, not even a star) to re-check that. My test P4 is glued to a heat sink, but I can put a thermocouple to the solder point directly. But direct driving a Q5 mounted only a star using 1xLi-ion is going to heat up the LED rapidly. You essentially have a 4W heat source that is very sensitive to temperature and whose longevity doesn't bode well. Try running a 4W night light some time and then touch the bulb.

 

[lonesouth]

Good job on the mod! Enter me in the drawing!

 

[Ty4752]

"I want the MM mod"
Thanks,

 

[Justin Case]

I put a thermocouple to the solder point of a test Cree XR-E P4, as shown here: http://www.cree.com/products/pdf/XLampSolderingandHandling.pdf

My test P4 is a emitter/substrate reflow soldered to a star. The star has been nibbled away to about a 15mm diameter circle (it was scavenged from a crappy P60 drop-in). I attached the emitter/substrate/star to a light piece of metal using Arctic Alumina compound under the star and then secured the LED to the heat sink using standard epoxy around the perimeter of the star. I suppose that technically, this AA Mini Mag mod has less heat sinking than this (just the round metal mini-board under the substrate).

I powered the P4 at Vin=3.6V and within one minute, the thermocouple temp reached 170F (77C). The datasheet spec for thermal resistance between LED junction and solder point is 8C/W. I drove the LED at [email protected] (this current level is why I didn't run the P4 at 3.7V or 3.8V to simulate a Li-ion more closely), or 4.1W. Thus, the junction temp is calculated to be 77C+8*4.1 = 110C.

At full power, I wouldn't run the light for more than a few sec at a time if direct driving a Cree XR-E Q5 with 1xLi-ion. Parasitic resistance in the light can help here, acting like a built-in dropping resistor.

To simulate the effect of parasitic resistance/dropping resistor, I ran my test P4 at 650mA drive current (3.35V input) and reached 123F (51C) after 3 min. That gives an estimated junction temp of 51+8*(3.35*0.65) = 68C. The thermocouple measurement seemed to be moving upward fairly slowly after 3 min, so I can believe that running for 5 min or perhaps even for 10 min continuous would not heat up the light excessively (however, what you perceive on the outside of the flashlight may not reflect what is going on at the LED, especially since the LED looks to have poor thermal contact to the rest of the metal flashlight body). At 650mA drive current, you could get around 130 lumens out the front.

When I ran my XR-E P4 at low power (3V@~150mA), the solder point temp reached 92F (33C) after 2 minutes and didn't look like it was going to move much higher (if any at all). The calculated junction temp in this case is about 33C+8*0.45 = 37C.

So running the light in direct drive using 2xalkalines or 2xNiMH looks like there should be no heat issue. Of course, output is also correspondingly lower. Relative luminous flux at 3V/150mA drive current is about 50% or around 50 lumens for the Q5. Not bad for a quick and dirty AA Mini Mag direct drive build.