Not sure what to call it: 18AAs, custom switch, MR16, starts fires, the usual...

[mdocod]

I'm always on a budget so I like to try to get the most lumens for my dollar, I took a very popular reverse clickie (the Judco IIRC) and figured out a way to mount it up in a hunk of delrin (I ran a thread on the switch assembly awhile back to see if there was any interest in me making a batch of em, never really saw a lot of responses)..

I'm just running a typical off the shelf 50W "spot" (axial filament) MR-16, you can find em in any hardware store, only a couple few bucks each. On 18 eneloops I'm estimating I'm getting ~19V to the bulb, ~100W re-rated wattage, and around 2400 torch lumens... I'll test that Vbulb and Tail current after I charge up the pack again, as I'm curious ... The cells have to be rested for at least an hour otherwise it pops the bulb.

[edit in:]I just measured 21+V at the bulb and 5.75A on fresh cells that had some resting time. Or about 120 watts!!! I'm very surprised at this as I was not expecting to see this kind of Vbulb under this kind of load, eneloops are good, but this must mean that the resistance of the package is quite low. 1.166V per cell after resistance losses under a 5+A load is quite good IMO! With this information in mind, I think it's safe to say It's in the 3000+ torch lumen category

I'm pretty happy with this setup. I've been playing with it on and off for the last few months. The axial filament produces a wall of light that is gradually brighter towards the center. It's a pretty smooth beam with minimal artifacts but the beam quality varies heavily from brand to brand of MR16. The ~100W (50W rated bulb) setup should run for over 15 minutes by my calculations. Dropping in a 20W (driven to ~40W) spot works fine as well and extends runtime to around 45 minutes.

Oh- I ordered a thick High Temp O-ring to help hold the MR-16 down in place a little better. seems to hold up to the heat very well, I've run it for at least 2 minutes straight so far and it does fine, head does get very hot after the first minute

Only problem is I keep busting KD glass lenses from the heat when I do fire starting performances (some of the MR16s I found don't have the built in glass)

 

[steed77]

WOW what a great combo. Thank you!

I have a 6D host, looks like I may try to make something similar. :thumbsup:

 

[ICUDoc]

impressive mod- those eneloops are good batts.

 

[DUQ]

Well done man!

 

[SafetyBob]

Mdocod, I finished my first 3x big emoli MR16 flashlight last weekend. As I pulled out all my M@g modded LED flashlights Seoul based P4's running at 500mA, the MR16 just gave out a wall of light. No spot to be realized until you approached a wall around 6 inches or so. Good bit of heat out the front which suprised me a bit. But it sure did overwhelm and overtake any of the LED lights.

The reason I was posting on this old thread was to alert everyone wanting to do MR-16 mods to look closely at yours......you don't need the Kiu tower in place. Make the leads long enough so you can just plug the socket into the bulb. Easy, simple and cheap. Most importantly, it would be a snap to change bulbs...

Got some of the Judco switches and am coming up with my own Delrin replacement switch too. Haven't figured out where I want my charging jack located exactly, but with the new plastic switch housing it should be easy to do.

Over modding may actually make for easier modding in the end....

Bob E.

 

[RKirk]

how do you connect the bulb?

 

[mdocod]

Hi RKirk,

The bulb is plugged into a standard ceramic bi-pin socket like this one:
http://www.1000bulbs.com/Halogen-Sockets/4564/

The socket is wired to the custom switch assembly that I devised for the modification.

 

[TheInvader]

What host is it?
18AA's could fit in a 4D with some trouble, and quite easily in a 5D.
also: is that a de-ano'd head and tail? how'd you do it?

 

[mdocod]

Hello TheInvader,

The host is a 6D mag with 3AA cells abreast for a total of 18AA cells into the space of 6D cells.

The head and tail are just "swapped" with a "silver" colored mag. I think it was a 4D host I picked up in silver. I have a tendency to swap parts with my mags to give them some flare.

 

[ANDREAS FERRARI]

Mdocod-maybe I missed it but aren't the MR16's 12v. lights.(some are 120v.)

I would not have believed you could drive these past 20v. without them going!!!

I have seen these bulbs everywhere and have always wanted to try them in one of my high powered host.I have a 4D/5 IMR/18.5v regulated drive light that these might work in.

Great work!:twothumbs

 

[mdocod]

Hello Andreas,

Long Life halogens rated in the 2000-6000 hour life range will often handle overdrive anywhere from 1.5X-2X+ the specified voltage. And that overdrive is required to get the kind of performance we demand in terms of whiteness/brightness/efficiency.

That said, bulbs that are designed with low life and more intense drive levels in mind are going to deliver a slightly better performance when driven hard. Say for example: an 18V 200Hr bulb vs a 15V 2000 hour bulb. Both will run pretty similar when driven to say, 20V, but the 18V/200hr bulb will usually deliver better lumen maintenance (less dimming through life) and possible better reliability and slightly better efficiency. The reason for this is that the bulb has been optimized for use at higher temperatures, which effects the way fill gases are going to be balanced and what pressure they are at for best results. So... Taking a 12V bulb and over-driving it to 20+V is very doable, but should be viewed as more of a toy than a tool since the bulb life and lumen maintenance is going to be less than ideal.

Most of the popular "builds" on CPF revolve around medical/scientific/projector bulbs that are specifically engineered for use at high drive levels to achieve the best possible CRI and higher white points. (their specified voltage rating is already pushing high temps, whiteness, etc). So, when we overdrive them just a hair (taking say, a 100 hour bulb and dropping the life to ~10 hours), we aren't operating the bulb as "far" away from what the original intent was, which results in a better performance in most cases.

-Eric