Plans for the MC-E

[TigerhawkT3]

I can't wait for the MC-E to come out. I already have a few things I'd like to try.

2S2P star to DD off Li-Ions or LiFePO4 14500s in a Minimag. To keep the focusability, the pill would have to be slightly loose in the body, which would leave only the body's top "lip" for thermal transfer. It might be more reliable to press fit and/or AA the pill in there, make it fixed focus, and switch with a tailswitch like the (non-IQ) Nite-Ize one. Depending on the Vf, ordinary 14500s might be too much voltage, so LiFePO4s could be good. The difficult parts for this build would be the (copper) pill and finding a small enough star and optic. Still, I think it'd be worthwhile.

4S1P star to make a sort of Super Birthday Mag, running off 6x14500 instead of 6xNiMH. Hey... would anyone be interested in buying something like that?

2S2P... to replace LuxVs, like in the U2 and so on. :naughty: I don't have any LuxV lights, so maybe some other brave modder will try that out.

MC-Es could be good for a new, more compact video light. Four in series, each wired 4S1P, mounted on a CPU heatsink and fan, powered by five 12V SLAs could be handy.

Let's hear some more crazy ideas!

 

[Greg G]

I want to put one in a Surefire L5, and then build a nice Mag using one.

 

[LukeA]

I want to put one in a Surefire L5, and then build a nice Mag using one.

Costco had their 3D mag + 2AA minimag pack discounted to $15, so I picked one up for an MC-E. I don't need it yet, but I couldn't resist at that price.

 

[Illum]

4 modes of light without PWM....
all you need is something that can operate 4 switches:nana:

 

[DaFABRICATA]

4 modes of light without PWM....
all you need is something that can operate 4 switches:nana:

I want to a driver that will turn on and off each die in timed intervals to make it appear there is some weird twirling going on inside the reflector.:tinfoil::duh2:

...but seriously, I would like to put one in a KL4.

It should be interesting to see what is possible....maybe 7 McR27XR reflectors with 7 X MC-E's-:devil: in a BIG host:naughty:

 

[Greg G]

I think 4 MC-E's and 4 McR19XR reflectors would make a nice light engine for a Mag host.

That's my plan for a NASCAR Fade Mag host.

Not exactly sure how I'll power it yet. A blue Shark and a battery pack to match how I wire the LED's. I'd like to use a Fivemega 12AA - 3D battery holder.

 

[monkeyboy]

You could use the Fatman driver to drive the MC-E in 4S configuration using either 2 or 3 li-ion in series. This would be a better/ more compact driver solution than anything I've seen for the SSC P7.

 

[Illum]

I want to a driver that will turn on and off each die in timed intervals to make it appear there is some weird twirling going on inside the reflector.:tinfoil::duh2:

hey, haven't thought of that :naughty:
chasing LED circuits are pretty mainstream too

PWM set to exactly a quarter of the period to another LED for 4 strings could potentially allow one LED to be on at any given time, reducing or eliminating a "flicker" effect even on PWM...:green:

 

[znomit]

I want to a driver that will turn on and off each die in timed intervals to make it appear there is some weird twirling going on inside the reflector.:tinfoil::duh2:

Here you go. 10 channel sequencer for power LEDs.

Im thinking of a running three dies off my bike dynamo and the 4th off battery (for when I'm stopped).

 

[VanIsleDSM]

I have a 4 dies in series buck driver designed to run from 6 Li-Ion cells already built, made for the MC-E, I'll be publishing the efficiency figures at different inputs and outputs soon, but it's around 94% efficient.

I have a 3D model for a flashlight already designed to work with the boomerang reflector aswell.. all sealed using reed switching for on-off and brightness levels.

I'll definitely have some extra drivers for CPFers to have a go with.. more to come soon on all this.

 

[rantanplan]

I think I´ll upgrade my "Space Needle"-clone with my first MC-E ... simple Mag 2C with two 18650, a Maxflex/Shark and for the first try the original Mag reflector ... hopefully the four dies will work with reasonable throw and without big holes.

Putting four MC-Es in a Mag would be thrilling :naughty: ... but difficult to realize I guess.

 

[chimo]

It will be nice to see some MR16 halogen replacement products be developed with these. With the 4 die connected in series, it should be a short development period.

 

[2xTrinity]

It will be nice to see some MR16 halogen replacement products be developed with these. With the 4 die connected in series, it should be a short development period.

For the 12V type, a simple bridge rectifier, and regulator would work. or 120v, a small 10-1 stepdown transformer might be needed first, but then otherwise, same thing. I ceertainly hope they offer the MR-16s in neutral shades. They will almost certainly be offered in warm (similar to incan) and cool (most lumens).

 

[flex76italy]

I have a 4 dies in series buck driver designed to run from 6 Li-Ion cells already built, made for the MC-E, I'll be publishing the efficiency figures at different inputs and outputs soon, but it's around 94% efficient.

I have a 3D model for a flashlight already designed to work with the boomerang reflector aswell.. all sealed using reed switching for on-off and brightness levels.

I'll definitely have some extra drivers for CPFers to have a go with.. more to come soon on all this.

Your project sound very interesting....keep me posted!

However another generation of led's...but we are ready to spend a lot of money for mod's! :sick2:

 

[chimo]

For the 12V type, a simple bridge rectifier, and regulator would work. or 120v, a small 10-1 stepdown transformer might be needed first, but then otherwise, same thing. I ceertainly hope they offer the MR-16s in neutral shades. They will almost certainly be offered in warm (similar to incan) and cool (most lumens).

For my use, I was thinking of the 12V type that would work off existing dimmer controls - essentially plug and play. One challenge may be heat dissipation with the current set of fixtures. There should be a sizeable market for those if they can get the price point right. Energy saving and less heat generation.

 

[R33E8]

*removed*

 

[chimo]

I think heat will be the largest issue.. When I interned at Lamina the heat sinks on the MR16s were well over 50C (as in a too hot to hold for long) at 8 watts... And these heat sinks are definitely not just random design ideas..

I agree, see my post above.

If we could get some decent output at the sub 5W level it could be do-able.

 

[2xTrinity]

For my use, I was thinking of the 12V type that would work off existing dimmer controls - essentially plug and play. One challenge may be heat dissipation with the current set of fixtures. There should be a sizeable market for those if they can get the price point right. Energy saving and less heat generation.

I mean a regulator would need to be built into the lamp, not that the end-user would be required to add one externally. Dimming support will still require a regulator to be built into the lamp -- in fact a much more complicated one, at that.

12V AC is actually a sine wave that oscillates between +17 and -17V at 60Hz. Many cheap LED lights only add a current-limiting resitor, and use the LEDs themselves as rectifiers to only use the positive half of the sine wave, producing a very nasty 60Hz flicker (far worse than the 120Hz flicker of magnetic fluorescents)

In order to driver the LEDs properly, it's necessary at the very least to add a rectifier and filter capacitor that will convert 12VAC into 17VDC. This is too high to drive 4LEDs directly, so it needs to be stepped down with some sort of regulator.

The way the typical triac dimmer works is by chopping off a portion of the AC input wave. However, after the rectifying and filtering steps described above, that will not translate into a change in the regulated DC output. A microcontroller would actually need to "look at" the shape of the incoming waveform before it's filtered, then adjust the output current of the regulator accordingly.

For the design to be considered safe, there will probably also be thermal ramping included, where the lamp would be automatically dimmed if the bulb starts to heat up to an unacceptable/unsafe level. This would be accomplised by adding a temperature-sensor as an additional input to the microcontroller.

The above circuit is fairly complicated, and with the LEDs themselevs costing probably over $10/each even in bulk, don't count on these paying for themselves in energy savings anytime soon -- even in air-conditioned spaces. Of course, saving money certainly isn't the main motivation for me personally being involved with LEDs. It would be worth more to me to have a bulb that is long life, higher color temperature, and dimmable to extremely low levels without changing color temperature.

 

[TigerhawkT3]

I have a 4 dies in series buck driver designed to run from 6 Li-Ion cells already built, made for the MC-E, I'll be publishing the efficiency figures at different inputs and outputs soon, but it's around 94% efficient.

I have a 3D model for a flashlight already designed to work with the boomerang reflector aswell.. all sealed using reed switching for on-off and brightness levels.

I'll definitely have some extra drivers for CPFers to have a go with.. more to come soon on all this.

Any visible PWM flicker? Dimmable? How large?

 

[chimo]

I mean a regulator would need to be built into the lamp, not that the end-user would be required to add one externally. Dimming support will still require a regulator to be built into the lamp -- in fact a much more complicated one, at that.

12V AC is actually a sine wave that oscillates between +17 and -17V at 60Hz. Many cheap LED lights only add a current-limiting resitor, and use the LEDs themselves as rectifiers to only use the positive half of the sine wave, producing a very nasty 60Hz flicker (far worse than the 120Hz flicker of magnetic fluorescents)

In order to driver the LEDs properly, it's necessary at the very least to add a rectifier and filter capacitor that will convert 12VAC into 17VDC. This is too high to drive 4LEDs directly, so it needs to be stepped down with some sort of regulator.

The way the typical triac dimmer works is by chopping off a portion of the AC input wave. However, after the rectifying and filtering steps described above, that will not translate into a change in the regulated DC output. A microcontroller would actually need to "look at" the shape of the incoming waveform before it's filtered, then adjust the output current of the regulator accordingly.

For the design to be considered safe, there will probably also be thermal ramping included, where the lamp would be automatically dimmed if the bulb starts to heat up to an unacceptable/unsafe level. This would be accomplised by adding a temperature-sensor as an additional input to the microcontroller.

The above circuit is fairly complicated, and with the LEDs themselevs costing probably over $10/each even in bulk, don't count on these paying for themselves in energy savings anytime soon -- even in air-conditioned spaces. Of course, saving money certainly isn't the main motivation for me personally being involved with LEDs. It would be worth more to me to have a bulb that is long life, higher color temperature, and dimmable to extremely low levels without changing color temperature.

Thanks for taking the time to post this, I know it takes time an effort to present things in a clear manner. However, you didn't have to take the time on my behalf, I'm an EE and understand the theory.

I still don't think the development will take too long (say within a few years) for products to come to the market and drop to a reasonable (cost-effective) price.