Mag3X descendants : the Mag3XD is born!

KevinL

Flashlight Enthusiast
Joined
Jun 10, 2004
Messages
5,866
Location
At World's End
The Mag3X lineage continues..... /ubbthreads/images/graemlins/grin.gif

modamag was kind enough to relieve me of my money.... *cough* err, I mean *cough* provide me with a Mag1D that I couldn't resist /ubbthreads/images/graemlins/grin.gif /ubbthreads/images/graemlins/twakfl.gif it is a pewter 2D that has been cutdown and re-threaded (and re-threaded much better than the original threads which were off by a little) so as to accomodate only one D cell.

Together with an Elektrolumens 3toD adapter, I put 3 Sanyo 2100mAH cells into the light for 3.6V. 3.6V and K-bin Luxeons means more Mag3X!

This time I requested a custom flat disk sink with no PR base, did the chopping of the Mag switch core as portrayed in the OA4D Cookbook, and soldered the 24AWG lines directly to the switch core for the LEAST possible resistance in the path. Current freaks tend to do that /ubbthreads/images/graemlins/grin.gif (hey, at least I didn't use "J" Vf bins! /ubbthreads/images/graemlins/evilgrin07.gif)

A couple of drops of cyanoacrylate served as thread lockers to bind the head to the body so that some hapless user doesn't tear it loose by accident, however as a safeguard against some he-man using brute force on it, the lines have been left long to provide strain relief. None of the Mag3X are focusable anyway. m

Thus, the Mag3X-1D is born. Shorter, smaller, and lighter. However, given the limitations of D-cell heads, I could not get SO20XAs to fit. Actually, you can get them to fit, but the outer O-ring does not seal tightly, compromising water resistance of the light, so I went with SO17XAs for this particular mod. It's still bright and beautiful, but it doesn't reach out and ping something like the 20mm reflectors do. On the other hand, you get a SUPERB flood close up and much better performance (IMHO) than optics.

Additionally, it weighs less. 500 grams (1.1lbs) vs 750 grams (1.65lbs) when fully loaded with NiMH cells. I like to travel light, too /ubbthreads/images/graemlins/grin.gif

So in a nutshell: 3 x Lux3 just like its precursor, 3 x SO17XA, 3 x AA NiMH.

Since this thread is worthless without pics, let me post a few. More to follow later.

Still looks like an ordinary Mag, but shorter:
mag3xd-1.jpg


Mmmm...pretty colors!
mag3xd-2.jpg


That's the first indication of when you know it's not your normal Mag:
mag3xd-3.jpg
 

daBear

Enlightened
Joined
Dec 26, 2004
Messages
250
Location
Chigger Branch, MS
Well you answered one of my questions about the reflector vs optics. I am also doing a D Mag Mod and have used NX05 optics which are giving way too much flood. I guess it is okay to light up a room inside, but take it outside and there is no throw past 50 feet.

Nice job, by the way.
 

KevinL

Flashlight Enthusiast
Joined
Jun 10, 2004
Messages
5,866
Location
At World's End
/ubbthreads/images/graemlins/thanks.gif guys /ubbthreads/images/graemlins/smile.gif

hotbeam: You can probably guess why I was looking for K-bin Vf LEDs. I could have gone crazy and used Js, but I found that on my original Mag3X (black, 3C housing), eventually the Vf will fall and the light will become brighter. Comparing by the ceiling bounce test, the Mag3X-1D is a little dimmer, but that's because the Mag3X-3C original has had more than 20 full pack discharges on it already. Vf drop is well documented by others using more sophisticated instrumentation than I have.

This one should have a tad less resistance too, the original was using a PR base because I wanted it to be field removable. Hardwired lines should provide the least resistance of all compared to mechanical linkage.

daBear: yup, by strange coincidence I managed to finish up my SO17XA based mod. IMHO reflectors offer a lot more than optics (now I can see why each and every one of my MicroMagMadMax+ customers opted for SO17XAs, even though I gave them a choice of that and the NX05).

I have an NX05 in a Surefire KL1 as well, and it is very much of a flood. I wouldn't even count on it throwing much beyond 30ft /ubbthreads/images/graemlins/grin.gif

Craig from the LED Museum did a great job of comparing the SO17XAs vs NX05, his findings are here. He used it on the original Inretech Trilight3, which was my inspiration for the Mag3X project. I have always wanted a Trilight3 since day zero here on CPF, the thought of 240 (raw) lumens has always been extremely appealing, and frankly, still is. /ubbthreads/images/graemlins/grin.gif

The CadResearch reflector is now the IMS SO17XA. Check out his phenomenal gains using the NiCad cells and reflectors - 633,000 mcd to 1.5 million mcd!

Edited to add pix: Enough of my yapping...let the pics do the talking.

The original, the upstart, and a coppertop for size comparison:
mag3xd-4.jpg


Side by side - Mag3X Original and 1D. If you're seeing a hotter spot, blame the reflector.
mag3xd-5.jpg


Tightest hotspot from the X-bin U2's massive 1.47" bezel and extremely deep reflector. The massive amount of flood is probably due to the X-bin /ubbthreads/images/graemlins/grin.gif Love my U2!!
mag3xd-8.jpg


Medium hotspot from SO20XA reflectors:
mag3xd-6.jpg


Diffuse hotspot but very generous side spill from SO17XA:
mag3xd-7.jpg


Last 3 beamshots shot at F2.8, 1/30 sec exposure on a Canon Powershot S40, ISO speed 100 on manual mode.

Yeah but which one is the light??!?!! /ubbthreads/images/graemlins/grin.gif
mag3xd-9.jpg
 

AvroArrow

Enlightened
Joined
Jan 21, 2004
Messages
290
Location
Vancouver, BC
KevinL, I was about to do a 3x D mod (thanks to your very detailed thread on the C body mod /ubbthreads/images/graemlins/grin.gif ) and was trying to figure out whether to use NX05s or SO17XAs. Guess you just answered that.

I do have a couple questions if you don't mind. What was the thickness of your heatsink/heatspreader? And did you have to cut the legs off the SO17XAs? I was looking up the measurements and even using a 1/8" heatsink, LuxIII emitters, and SO17XAs with the legs partly chopped, it would just fit... barely. How did you manage to get Stars to fit?
 

IsaacHayes

Flashlight Enthusiast
Joined
Jan 30, 2003
Messages
5,876
Location
Missouri
Since you decided to lock the head down and not use a PR, you could of used 20mm reflectors if you would of put the luxeons on a heatsink in the body tube and raised the head up some for extra clearence. I think that's what I'm going to end up doing for mine, as I've done it once for the Fraen FT3 Tri-optic. I used removable set screws in mine.

oh, btw What current does it pull?
 

KevinL

Flashlight Enthusiast
Joined
Jun 10, 2004
Messages
5,866
Location
At World's End
/ubbthreads/images/graemlins/grin.gif

I discovered something interesting last night. The Mag3X-1D is versatile enough to be configured for two-stage output, but not quite the 'two stage' (a la McE2S/Surefire L2) we are used to.

With the stock EL 3toD holder, the light draws 1.2A peak/800mA sustained current. This is a combined current reading taken from the base of the light, thus divide by 3 to determine what each star is receiving. This configuration is surprisingly bright - above beamshots were taken on the 3toD - and runtime is LONG, we're looking at around 3 hours! However, that explains why the light is a little dimmer than what I'm used to.

I hacked up a quick-n-dirty (dirty is the word. Aluminium foil and wire. I refuse to provide pix /ubbthreads/images/graemlins/tongue.gif) holder for 4AA cells since Modamag's 4toD holders will not be available till next week or the week after, and loaded it up. "hot off the charger", literally hot to hold cells, delivered 3.3A current draw settling down to 2.5A in the next couple of minutes. But WOWOOWOWOW CRAZY BRIGHT!! I love crazy bright!! *hides Mag85* /ubbthreads/images/graemlins/rolleye11.gif

The trade off is that your runtime may be under an hour, but WOW look at the LUMENS!! It leaves the original Mag3X and the U2 diving for cover! And all you need for this two-stage output is to change the battery mags and you are in business. In fact, if you buy modamag's adapter and a dummy cell, you can toggle between extended-run and extreme-brightness configurations. I didn't realize it until I'd slept on it for a while.


LEDModMan: Thanks /ubbthreads/images/graemlins/smile.gif part of the reason for its existence was that I wanted a shorter light to put in a bag. Most people would never suspect..... /ubbthreads/images/graemlins/grin.gif

AvroArrow: glad it was of some help. The original 3X uses a 1/4" thick heatsink, whereas this one uses a 1/8" thick sink. The SO17XAs do not extend all the way to the top, so I retained the Mag's polycarbonate lens to act as a spacer. The primary lens is an Ultra Clear Lens (UCL), so as not to cause 2X the losses. In fact, the problem is they're shorter, rather than too tall. But that's ok, the Mag's original lens comes in very useful.

IssacHayes: Come to think of it yeah, that could have been done. Hmm..... next time? /ubbthreads/images/graemlins/smile.gif Current draw is as stated above.
 

KevinL

Flashlight Enthusiast
Joined
Jun 10, 2004
Messages
5,866
Location
At World's End
PS: Some may ask how I can get away direct-driving 4 NiMH on a Luxeon 3, which seems to be too much for it. Good question.

I've discovered that 3 Luxeons in parallel behave very differently from one Luxeon (like the one in the OA4D Cookbook) on a single Hotlips. My first Mag3X taught me that - its current draw was actually lower than what I'd targeted. It's not a matter of plugging "multiply by 3" into all my calculations as I thought it was /ubbthreads/images/graemlins/grin.gif it's close to linear but there are a few other things.

They draw more current, thus causing more voltage sag, especially for AA cells. Yes, even NiMH AAs can sag, as evidenced by the Mag85. Pulling 3+ amps of current puts it in '85 territory! Anyway, it drops back to 2.5A after a short while, which remains safe (~850mA per Luxeon). Even peak load is only 1.1A per Lux3, and I have been known to drive some of my single Lux3 lights at 1.5A /ubbthreads/images/graemlins/rolleye11.gif so even hot off the charger cells measuring 1.45V are fine. I deliberately used hot cells to test.

For this specific reason I DO NOT recommend that you use ultra-high-current cells like KAN 1300s or CBP 1650s. Save those for your incan Mags. Other ordinary consumer-grade AAs will be fine. I use my tried-and-proven Sanyo 2100mAHs.
 

IsaacHayes

Flashlight Enthusiast
Joined
Jan 30, 2003
Messages
5,876
Location
Missouri
Kevin, if you didn't use stars, and used 20mm reflectors with the legs trimmed to fit emitters, you'd probably have just perfect clearence. From my calculations you'd need a 1/8th thick sink for 20mm in a D head.

For a better solution to what you have instead of using 2 lens, just use 1 UCL, and take the stock mag reflector. Cut around it, leaving the large lip. Then drop it in, along with the UCL. Bingo, no more lens rattle, 1 lens too! You could leave the reflector cut a little thicker and paint glow powder on it for a nice glow ring. /ubbthreads/images/graemlins/thumbsup.gif
 

KevinL

Flashlight Enthusiast
Joined
Jun 10, 2004
Messages
5,866
Location
At World's End
Hmm, yes. In fact my measurement indicated that it would fit, but I didn't take into account the stars height because I thought the SO20XA's feet already accounted for that. They would, if the feet stood on the same plane as the star base, but they don't. Oh well. /ubbthreads/images/graemlins/icon3.gif Next time /ubbthreads/images/graemlins/grin.gif

If someone wants to turn down SO20XAs, using sandpaper or other methods to shave a little bit off the front, they could manage to pack them in.

I could have modded the reflector, or a suitably-sized O-ring would take care of that too. Good tip, thanks /ubbthreads/images/graemlins/smile.gif I just don't have the equipment to do such precision cutting.
 

Latest posts

Top