Help with MCE

[Crenshaw]

Sort of have been left behind by the p7/MCE madness.

I have a 2D mag sitting around looking sad, so i fixed the Mag LED drop in i have from Kai, the one that came with the Aspheric set, and its working fine now, although it could be brighter.

Is it possible to just replace the current X-re with the MCE? and connected the current driver in series? (series = current constant throughout?) will it be brighter?

Crenshaw

 

[KevinL]

If you connect the LEDs in series, current remains the same however Vf is the sum total of all the LEDs. Assuming (roughly) 3.4V nominal Vf, you'd need something capable of a 13.6V output.

You could run a more forgiving 2S2P setup, 2x3.4V in series, and split the 1000mA across two dies. Such a setup would be a little bit easier to build.

Alternatively, direct drive is the easiest (and the craziest). I had a 3 x Lux3 mod (from nearly 3 years ago), I decided to go DD on that one because of the complexity of the driver setup and the cost, for at that time drivers sure did not come cheap, unlike the DealExtreme stuff we have now. Simply wire everything in the MC-E as parallel paths, stuff a suitable battery pack in (resister if you need) and let the chips fall where they may! Be sure to cater for up to 3 amps of current though..

My DD light turned out to be pretty OK in the end. Pulls about 2+ amps and has been consistently reliable through the years.

 

[Crenshaw]

wow okay so no direct replacement. I may consider building one as stefan has, but it would require taking time to aquire parts.

anyway...thanks Kevin!

Crenshaw

 

[KevinL]

No prob. You could do 1:1 replacement by wiring the MC-E in 4P, retaining same Vf but splitting the current across 4 dies. However the current would be so low that I am not sure if it would be worth it.

 

[Justin Case]

If I understand your current configuration correctly, you are driving a Cree XR-E LED using some sort of driver that outputs 1000mA drive current? The host is a 2D Mag.

You could wire the MC-E in 4P and underdrive it at 250mA per core with your existing driver. If the MC-E is an M flux bin LED, you could get at least 60 lumens more OTF vs a Cree R2, while also pulling significantly less power. I built an AW Turbo Tower using an MC-E driven by a GD1000 boost-buck driver and I measured only 2.95V Vf, or about 3W consumed by the LED. A Cree R2 will put out less lumens, yet consume probably at least 10% more power.

This sort of setup will have no heat buildup and thus no lumens decrease vs runtime due to LED heatup.

 

[Rexlion]

Not that I know anything about doing mods, but why not bypass the driver? I just bought a 2C Mag modded with P7 direct driven on a 18650, and it rocks.

 

[Crenshaw]

If I understand your current configuration correctly, you are driving a Cree XR-E LED using some sort of driver that outputs 1000mA drive current? The host is a 2D Mag.

You could wire the MC-E in 4P and underdrive it at 250mA per core with your existing driver. If the MC-E is an M flux bin LED, you could get at least 60 lumens more OTF vs a Cree R2, while also pulling significantly less power. I built an AW Turbo Tower using an MC-E driven by a GD1000 boost-buck driver and I measured only 2.95V Vf, or about 3W consumed by the LED. A Cree R2 will put out less lumens, yet consume probably at least 10% more power.

This sort of setup will have no heat buildup and thus no lumens decrease vs runtime due to LED heatup.

so i would actually be inscreasing efficiency if i were to do a direct swap...hmmmmmm interesting. COnnecting it in series would mean connecting the LED leads to the next one, and then subsequent right?

Crenshaw

 

[Justin Case]

To wire in series, visualize how something like 4x123As are connected in most flashlights (also in series). The top cell's + connection connects to the load +. The top cell's bottom - connection touches the top + connection of the 2nd cell. The 2nd cell's bottom - connection touches the top + connection of the 3rd cell. Etc etc. Finally, the bottom - connection of the 4th cell connects to the load - to complete the circuit (through the tail switch). Wire the MC-E similarly. You need to note the "cut corner" of the MC-E case to know where terminal #1 is so that you wire the MC-E correctly.

However, I propose to wire the MC-E in 4P (solder a wire across all 4 of the + terminals, and solder another wire across all 4 of the - terminals). Drive the 4P MC-E using a BBNG1000 if you are using 2xD NiMH. Use an SOB1000 if you are using 2xD Li-ion.

I've measured ~3V Vf for my 4P K-flux bin MC-E setup driven at 250mA per core. Ceiling bounce is brighter than an identical setup using a Seoul P4, which has Vf ~3.35V. Thus, ~3W power consumption for the MC-E vs. ~3.35W for the P4. And more lumens out for the MC-E vs the P4.

An M-flux bin MC-E will put out even more lumens. Since I don't have any 4P M-bins built, I can't say what the Vf will be vs a K-bin MC-E, but hopefully the same.