Mag 62138 Regulated Build. Originally started as Mag623 DD.

[bigchelis]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

64458 12v 90w

thanks. Just ordered 2 of them. I'll try it with 5 imr 26500's, if I get brave maybe 6 imr 26500's

 

[alpg88]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

thanks. Just ordered 2 of them. I'll try it with 5 imr 26500's, if I get brave maybe 6 imr 26500's

6 will flash, but 5 will work just fine, I use sony 18650 high current cells I took out from Makita drill, and 10A tailswitch, on my light, i'm not sure 100% but I think stock maq switch, even thou will work, but resistance will eat up some voltage. i think you'd get better results with high current switch, http://www.candlepowerforums.com/vb...ker-Switch-in-a-Mag-a-continuation-(Pic-Heavy!)

 

[The_Driver]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

From my experiences, the orientation of the filament effects throw quite substantially. Axial filaments are what you want if you're looking for throw. It allows the reflector to concentrate more of the lumens into the center hotspot. A smaller hotspot will have a longer throw (albiet a smaller diameter) when comparing the same filament if oriented horizontally.

An analogy is looking at the XP-G emitter versus XM-L utilizing the same reflector and same input power. The XP-G will out throw the XM-L due to its concentration of light output in the middle of the reflector.

Actually this is not true. For throw only the luminance (= "surface brightness") of the light source in combination with a reflector/optic counts. It just so happens that the smaller XP-G has a higher luminance than the XM-L.
A current XM-L2 in U3 oder U4 Bin probably has a higher luminance compared to the old XP-G. But it's twice as big.....

The same principle applies to bulbs. The luminance is what matters. The orientation of the filament only has an effect on the shape of the beam.

@bigchelis: I think you should think about getting a regulator from JimmyM. They just make sooo much sense in these kinds of lights. You can precisely regulate any voltage lower than that of your batteries.

The 62138 is an axial bulb. Please take another look at the link I put into one of my earlier posts with the luminance values. The filament orientation of each bulb is in the list.
transverse = horizontal
axial = vertical

Also please take a look at this old thread. It is a very nice collection of beamshots. You can see the difference between the beams of the 64623 and 64458 in the throwmaster reflector. You will see that the at least in his battery-bulb-combinations the 64458 actually has a lower luminance and coesspondingly less throw than the 64623 that you are already using. You can will also see that the 62138 beats both of them when it comes to luminance/throw, but the difference between it and the 64623 is actually not that great. Not when it comes to total brightness though (64623 puts out more lumens). These beamshots also show what I said earlier: 3-inch reflectors will give you even more throw than deep 2.5 inch reflectors.

I think you can probably stay with the 64623. You wont gain much with the other bulbs except for maybe a nicer (more even) beam.

 

[alpg88]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

what you said may be true for bulbs driven at rated voltage, but when overdriven, everything changes, no other bulb you mentioned can take as much and burn as bright as 64458, and very few here run bulbs at rated voltage, no fun in that.
http://www.candlepowerforums.com/vb...ulb-Tests-Updated-8-27-2010-(Newer-Info-Added)

 

[bigchelis]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

Since, I never done a high current switch mod on a stock Maglite housing I actually have a FET DD single mode driver for this build. The FET built by Mtnelectronics for me and for the 12V plus inputs and it runs cool while absorbing all the 10A~ of current. This seems to me a lot easier, I have to assemble it today to confirm how it likes the hotwire.



The Hotwire Driver would be ideal, I am not sure though if they are still made and sold.


I will post some pictures and lux values later on. Pictures do help but I think a measured lux values add some more detail that even pictures sometimes cant show.


Thank you,
bigC

 

[alpg88]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

i'm not sure fet dd driver built by Mtnelectronics could handle 150w let alone 200w of power. overdriven 64458 pulls about 200w (20+v 10+amps).

i actually wrote to owner of mtnelectronics asking about it, i'll let you know when i get answer.

i actually got one few days ago, and build triple xpl wired in parallel on noctigon star, works great, but i doubt it pulls more than 3-4 amps now, i plan on getting high current cell and hopefully high current switch, i'm pretty sure at this point stock SF switch is what limits me due to resistance. i would think the driver woudl handle 8-9A at 3v, but i'd wait for answer before tryng to get 150+w thru it.

p.s. i got an answer, he says no. amc7135 will blow up for sure, and fet most likely can not handle that much power. i also forgot to mention to him, the question was about a halogen bulb, unlike leds, its peak on start up can be 2x as much amps. so you looking at 400w+ for a short instance.

 

[bigchelis]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

Placed my order for these 2 bulbs and Incan Driver to conduct actual Lux test

Osram 64458 which I will DD with 5 IMR 26650's
Osram 62138 with 14V Incan Driver which will use 4 IMR 26650's


What makes it more of a realistic test is if I can use the same reflector, lens, cells, switch, and just add a cell or subtract to test. Tailcap currents will be taken too.


bigC

 

[alpg88]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

you could easy, they are same bulbs dimention wise, just unplug one and plug another. also i noticed when i install any of those monster bulbs, they smoke for about a minute or so, no matter how i clean them, they still smoke a bit. , so leave reflector off untill smoking stops, than put head back on. it only happens once when i install it, after that no smoke.

 

[The_Driver]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

you could easy, they are same bulbs dimention wise, just unplug one and plug another. also i noticed when i install any of those monster bulbs, they smoke for about a minute or so, no matter how i clean them, they still smoke a bit. , so leave reflector off untill smoking stops, than put head back on. it only happens once when i install it, after that no smoke.

but he would need to adjust the Bulb voltage quite a bit. The 64458 is a 4000h bulb while the 62138 is a 50h bulb. This means the 64458 needs a much higher voltage (more "overdrive") compared to the 62138 to reach the same efficiency and color temperature.

By the way - I blew a 62138 at ~13.7V with my regulator from Jimmy. At least if his formulas for adjusting the bulb voltage are accurate....
I am now running it at ~13V and it has been working fine.

 

[alpg88]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

sure he'd need adjust voltage, by removing or adding a cell, if you look at destructive tests, you'll see every bulb has a "spot" where it works without flashing with some number of cells.

62138 flashes at 15v according to same tests, i'm not sure why you blew it at 13,7v, may be it was defective, or its lifespan was on its last minute.

 

[bigchelis]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

sure he'd need adjust voltage, by removing or adding a cell, if you look at destructive tests, you'll see every bulb has a "spot" where it works without flashing with some number of cells.

62138 flashes at 15v according to same tests, i'm not sure why you blew it at 13,7v, may be it was defective, or its lifespan was on its last minute.

Jim himself suggested I keep the voltage at 14V for the 62138.

The Voltage Control pots on the driver itself I can adjust the voltage with a flat head screw driver by turning it right to increase and left to decrease. A full turn is equal to about 4V. So, a quarter turn left will put me at 13V or so in case I do flash the first 62138.


Alternatively; I can use the same driver for the 64458 just give it 2 more cells and give the driver 22V by simply turning right 2 turns. The other voltage pot would be the low voltage one which I would turn 2 turns as well.


bigC

 

[alpg88]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

even better if you can adjust the driver. you can come closer to max brighness as opposed with simply adding \removing cells.

 

[The_Driver]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

sure he'd need adjust voltage, by removing or adding a cell, if you look at destructive tests, you'll see every bulb has a "spot" where it works without flashing with some number of cells.

62138 flashes at 15v according to same tests, i'm not sure why you blew it at 13,7v, may be it was defective, or its lifespan was on its last minute.

One needs to take into account the manufacturing tolerances of Osram bulbs (they are not perfect) and also that LuxLuthor softstarted the bulbs for 30s. He also did not actually test the bulbs lifetimes for different voltages (that many combinations would take years ).
Believe me - I have studied his tests many times, backuped his tables etc.

I guess I just wasn't very lucky with that one bulb. It blew 2s after I first turned it on (slow soft start). I turned down the voltage because I didn't want to blow my only other bulb of this type. At some point I will buy some more...

One really practical aspect of these regulators is that you can actually build a hotwire with a practical runtime (by adding additional batteries in series). Mine with the 62138 runs for a solid 30 minutes with eight Panasonic NCR18650PD cells (2900mAh).

 

[bigchelis]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

Here it is: I got the PHD regulator from Jimmy and he set it on SLOW soft start to give my Osram 62138 a chance, but it only lasted at 14V setting 2 seconds before it flashed:sick2:

Thankfully; I ordered 2 of them. The 2nd one I twisted the Voltage Input pot left/counterclockwise a quarter turn which is roughly 1V input less. So, at ~13V input it works perfect now.

With the same 4D Host I been using and same borefloat lens, KAI SMO Reflector, FM socket, ect...Only now with the 62138 I get a steady 117.5K lux.:thumbsup:

 

[bigchelis]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

Ok...now I put the 2.5in Throwmaster bezel and holy crap is that a nice perfect white hotspot. Utter perfection.

Unfortunately; the 2nd Osram 62138 bulb also flashed about 30 seconds into the run.

While I did just Fry $15 bucks I will definitely order another set of 5~10 bulbs. The beam is just too good to give up now.

Maybe give the pot another 1/4 turn for 12~13V input approx.


bigC

 

[The_Driver]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

Keep us posted

 

[bigchelis]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

The Good news is I got the TRUE RMS and the 5W 1K resistor working great with the JimmyPHD Incan Driver.

The bulbamerica purchased bulbs Osram 62138 at 11.86V set (12V rated bulb) I just Instaflashed 2 of them. Despite having slow soft-start.

Lowered the Voltage to super LOW 10.18V and wallah it doesn't flash. Despite this I get 160K lux which says volumes about how great the 62138 can be if I could get it to survive at 12~12.8V.



Surprisingly; the hotspot is about 50% bigger than the Mag623 with same 2.5in Throwmaster. The corono is noticeably smaller on the 62138 though, so maybe it just puts out a stronger and bigger hot spot.



I will try 11.5V later this week. I already blew 4 bulbs and the mail man lost 2 of my on order bulbs so, that puts this to be a very expensive project.

BTW: With the True RMS meter you get the DC setting and just check exposed leads on 5w1K resistor.

 

[bigchelis]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

How the leads check the resistor
Mag Switch is ON
True RMS DC voltage setting

 

[The_Driver]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

Are you sure that you calculated the bulb voltage in the correct way? You need to know/measure the actual battery voltage under load.

EDIT: try this method of calculating the value.
Vbulb = sqrt(Vavg * Vin)
Vavg = measured voltage at the resistor
Vin = battery voltage under load
sqrt = square root

It is very unlikely that 12V bulbs from Osram don't work at their rated voltage.

 

[bigchelis]

Re: Mag623 with 4D Stock switch/4 imr 26650 and 85K lux

Are you sure that you calculated the bulb voltage in the correct way? You need to know/measure the actual battery voltage under load.

EDIT: try this method of calculating the value.
Vbulb = sqrt(Vavg * Vin)
Vavg = measured voltage at the resistor
Vin = battery voltage under load
sqrt = square root

It is very unlikely that 12V bulbs from Osram don't work at their rated voltage.

I must have misread the voltage setting options. You are correct I have to measure the voltage under load. I assumed as long as I had a TRUE RMS meter I would just put those leads to the resistor and I would get a reading. That's clearly not the case.

I have sent the switch and driver and bulbs to Jimmy. He offered to synchronize them correctly for me.


Will update as soon as I have updates.


bigC