Help -want to create Mag 6D with 3" head significantly brighter the Mag85

Hi all!

Right, I want to build a torch that is significantly brighter than a good Mag85 (I've already got one of those) but with good throw

Now for some reason I quite like the mag 6D with my FM3H 3" head on it - but it really needs to be able to back up its monster looks with monster light output. So that is my base for this light. I already have a load of 9500mAh D cell Nimhs and some pelican bulbs (on their way) for long burn time use.

I want to build something with a minimum continuous run time of 20 mins - or better 30 mins

I'm prepared to use 1/2 D cell Nimhs and I'm also intersted in the D Li-Ions and I'm prepared to extend the torch up to 2 D Cells in length at a push (prefer to keep to a 6 D or less - there might be some good options for a 5 D cell Li-Ion set up)

I'm going to put an order in for one of Delghi's 3" heads with 2 reflectors, one with a small hole and another with the large hole big enough for the super osram bulbs.

When I know what bulb/battery combination to go for, then I'll get one of Andrews/Kiu HD/high temp sockets jobbies

So....can any of you knowledgeable guys offer up some suggestions for solutions for a big torch with big light?

What kind of bulb lumens might I be able to achieve?

P.S. I've been trawling through the forum for the last week or so and it is an absolute mine of information - but it is a big forum with a lot of information to read up on so I'm hoping you guys can help me along a bit )

It's a bit dangerous too - I er accidentally put an order in for the new Wolf Eyes Boxer 24W HID and am looking at, at least one new 3" mag head - might have to get one of FiveMega's new offerings too.

Thanks in advance

Mart

6D NiMH cells would drive an Osram 64275 @ 7v for quite a while. Same with the philips 5761 @ 7.1 volts. These are both 45W+ lights.
Going with 1/2Ds would get 9 into a 6D which would drive the 1185. But you want brighter.
I've got a 7D with 11 1/2Ds that runs a 64625. It's a 115W light. With a 3" head it would be nuts! It can only run for 8-10 minutes because of heat (not the electronics, the external temp). The cells can run that cycle over and over before needing recharging.
There is a "dead zone" between 50W and 110W where I haven't found a lot of battery/bulb combinations. Maybe others can help.
I too have a couple of Delghi's 3" heads on order. It'll make a nice 5D li-Ion MagIRC when I get my 5D HAIII body.

Thanks for the quick reply Jimmy

What sort of beam does the 625 bulb throw, it looks very floody in one of the beam shots I saw of the Torch - will the 3" head clean it up enough for good throw

and what bulb would run with the 5D Li-Ion? what sort of lumens

Got it, Osram Halostar 64447 18V ish

Mart

Only a few of us have ever seen a '625 in a 3" head.

Even a 3" head won't completely clean it up, but it will throw!!

I will have a Mag bodied '625 with a smooth 3" reflector up for sale in the not to distant future, I am arranging for a coating for the bare aluminum bodied light and head right now.

You have to remember that the design purpose of these lamps was for projectors, where you want lots of DIFFUSE light to be then focused through a lens in a theater. They are specifically NOT designed to be used the way we are using them, so focusing them tightly is not going to happen within the confines of anything NORMALLY considered to be a flashlight reflector!

The filaments are BIG, and big filaments don't focus well unless you have BIG reflectors.

Bill

What about the Westinghouse 100W bulb at 14.4Vbatt (Vbulb 14.13) giving 3500L can that be done using four D Li-Ions, mag 4D cell body, 3" head and hot driver and Kiu high temp socket?

Does anyone know what sort of run times I'd get out of that?

Could the hotdriver cope with a 6 D cell Li-Ion set up and bring the voltage down to 14V? and would I still get the extra run time over a 4 D cell Li-Ion set up in doing that??

My electronics knowledge is a bit poo, so any help is appreciated

64447! You got it. 18.5V
The 64625 has a horizontal filament. My Mag625 has a, kind of, oval hotspot when using an 2" LOP. The IRC has an axial filament, so it focuses more nicely. The 3" reflector will make that thing quite the thrower. Good luck with the monster. It seems there are only a few here that are not devotees of the small light. I'm a fan of the large ones. There is something to be said for power and runtime.

Thanks Bill.

What about the stock switch in the mag host - I've read other people have replaced it with something else, is this going to be necessary if I try and use a 100W type bulb?

I Look forward to seeing this mag625 of yours.

bwaites said:

Only a few of us have ever seen a '625 in a 3" head.

Even a 3" head won't completely clean it up, but it will throw!!

I will have a Mag bodied '625 with a smooth 3" reflector up for sale in the not to distant future, I am arranging for a coating for the bare aluminum bodied light and head right now.

You have to remember that the design purpose of these lamps was for projectors, where you want lots of DIFFUSE light to be then focused through a lens in a theater. They are specifically NOT designed to be used the way we are using them, so focusing them tightly is not going to happen within the confines of anything NORMALLY considered to be a flashlight reflector!

The filaments are BIG, and big filaments don't focus well unless you have BIG reflectors.

Bill

Click to expand...

The stock switch will burn-up/melt-down under a 10+ amp load.

The hotdriver uses a linear regulator and dissipates power as heat. Additional voltage will only server to burn it up faster. When Vbulb is close to Vbatt, it is the best thing going.
When you want to run a Vbatt significantly than Vbulb, you need a PWM type regulator like the PIR.

Yup, something about the big lump of a 3" head on the stupid long 6D cell body, and if i can make something with the power and run time to do it justice.....

After a bit more reading I understand now that the 64447 is a 12v bulb being over driven to get the brightness - something in the order of 6k lumens at 18+ volts - holy crap! (from Andrew Wynns testing) http://www.candlepowerforums.com/vb/archive/index.php/t-116634.html

Amazing how progress has come on over the past couple of years (that I've been reading this forum) assuming the D Li-Ions are going to work ok in these type of applications

JimmyM said:

64447! You got it. 18.5V
The 64625 has a horizontal filament. My Mag625 has a, kind of, oval hotspot when using an 2" LOP. The IRC has an axial filament, so it focuses more nicely. The 3" reflector will make that thing quite the thrower. Good luck with the monster. It seems there are only a few here that are not devotees of the small light. I'm a fan of the large ones. There is something to be said for power and runtime.

Click to expand...

JimmyM said:

The stock switch will burn-up/melt-down under a 10+ amp load.

The hotdriver uses a linear regulator and dissipates power as heat. Additional voltage will only server to burn it up faster. When Vbulb is close to Vbatt, it is the best thing going.
When you want to run a Vbatt significantly than Vbulb, you need a PWM type regulator like the PIR.

Click to expand...

And these mods are below 10 amps, right? The 625 draws a bit under 9amps, according to a spread sheet I've found linked to from this site.

OK, I suspected there would have to a heat issue. So in general it is best to get a close Vbatt to Vbulb - as you say.

You've lost me with the PWM and PIR bits though, lol

I'd be happy with a 5D cell though

English_Mart said:

Yup, something about the big lump of a 3" head on the stupid long 6D cell body, and if i can make something with the power and run time to do it justice.....

Click to expand...

Hey! I resemble that remark.

English_Mart said:

After a bit more reading I understand now that the 64447 is a 12v bulb being over driven to get the brightness - something in the order of 6k lumens at 18+ volts - holy crap! (from Andrew Wynns testing) http://www.candlepowerforums.com/vb/archive/index.php/t-116634.html

Click to expand...

I've sent AWR a couple of 64458s to look at. It's a 4000 hour 90W 12V bulb overdriven with 18.5 volts. 8000+ lumens.

The PIR is a PWM (Pulse Width Modulated) regulator. Instead of varying the resistance of a MOSFET (the electrinically variable resistor) to drop the voltage of the battery to that which is suitable for the bulb. It switches from completely on to completey off 40,000 times per second. The PIR varies the ON time so that the effective voltage (RMS voltage) is suitable for the bulb. That way there is no power dissipated in the regulator. It can have 100s of times the power handling capacity of a "linear" regulator like the Hotdriver. The PWM is just a LOT more complicated. The Hotdriver is extremely accuate though.

Jimmy,

On what do you base your statement?

"The stock switch will burn-up/melt-down under a 10+ amp load."

Bill

bwaites said:

Jimmy,

On what do you base your statement?

"The stock switch will burn-up/melt-down under a 10+ amp load."

Bill

Click to expand...

An unmodified Mag switch has a lot of internal resistance. a 10+ amp load will generate a lot of heat. A plastic switch body cand conduct that heat away.
Your statement implies that I have no personal experience. You're correct. It is all from reading here and electrical theory P=I^2*R.
Do you have experience with un-modded mag switches such that they can reliably take 10+ amps? If so, let me in! I'd love some info. I've been thinking about using a MOSFET in a few big lights that don't necessarily NEED regulation, just a high current switch.

Well, a little.

You might want to follow this thread:

http://candlepowerforums.com/vb/showthread.php?t=57486

and this one:

http://candlepowerforums.com/vb/showthread.php?t=56934

and this:

http://candlepowerforums.com/vb/showthread.php?t=76509

and this:

http://candlepowerforums.com/vb/showthread.php?t=108945

All the prototyping for the USL was done with Mag switches that had been modified only by cutting off the bulb post and then soldering to the ribbon portions of the switch. The Mule/Fatmule were both built using stock switchs apart from that change.

I ran dozens and dozens of hours, if not hundreds on those switches and not once saw any damage from those runs. Nothing melted, nothing was apparently damaged.

The switch and body combination does have high resistance, caused mostly by the spring to pedestal contact and the tiny negative contact with the body/screw where the switch is fixed to the body.

As for holding up, it never showed any problems or damage, and in fact I still have two of the prototypes still running.

Bill

All you did was solder/weld to the switch tabs? That's great.
In one of your links below, it mentioned that the resistance welded strips are not up to the task of carrying 8+ amps. Did you end-to-end solder them?

bwaites said:

Well, a little.

You might want to follow this thread:

http://candlepowerforums.com/vb/showthread.php?t=57486

and this one:

http://candlepowerforums.com/vb/showthread.php?t=56934

and this:

http://candlepowerforums.com/vb/showthread.php?t=76509

and this:

http://candlepowerforums.com/vb/showthread.php?t=108945

Click to expand...

I simply soldered to the switch tabs. There is one caveat. I did very few runs for more than 12-13 minutes with the '623, '625, or '138, because the light gets too hot to hold. If you ran for longer than that, you might damage the plastic switch body. In fact, I almost didn't build the USL because of the heat issue. It is why I wouldn't use larger cells than the 1650's we used.

After that post was written, we learned a few new things. Resistance welding does work IF you use large weld points and enough of them. That creates a problem on AA cells, because you don't have much space to work with.

IF you use at least 4 resistance weld points and ribbon wide enough to allow that many, they are fine up to at least 9 amps, and maybe more.

OUR intial packs were all end to end soldered for the USL project, then some cells developed leaking problems blamed on that process. We then had special arrangements made with a pack manufacturer to our specs. Unfortunately, we found that some of the cells leaked, regardless of how they were put together.

We now are resistance welding the packs using wide ribbon and 4 points and have had no problems with performance in comparison to soldered packs AND we have the benefit of not having to worry about a broken or cold solder joint.

Larger cells have no problems with using larger ribbon and even more weld points, and can handle higher current draws because of it.

Bill

There was a lot of good reading in those links you posted, Bill. I was aware of "some" of the work and effort gone into developing the various super lights over the years, but it was a real eye opener going through that material.

There really has been a lot of effort, time, research and perseverance gone into finding solutions for these mad creations. Much respect is due I think!


So, Bill, how difficult is the soldering modification to the switch? I've only had very occasional use of a soldering iron, it doesn't sound too complicated - but I haven't ever pulled out a stock switch and examined one.

Jimmy, I've just realised you're the guy arranging the D Li-Ion group buy, LOL - I think I'll be wanting some of those.

Although this heat issue is something I'll have to bear in mind

Would I be correct in assuming that generally the more lumens any of the typical super bulbs we might be using - the more heat is going to be generated.

So, inevitably, a 8000 lumen bulb will have to be run for shorter periods - or are there some bulbs that seem to run colder whilst still kicking out stupid levels of light??

Hmmm, more about the 64447, looking at a thread you commented in, Bill you mention that this bulb draws 7amps, so this is looking increasingly like a good bulb for me to try, has the axial filament as Jimmy mentioned, too. But if I drive it at around 18V (possibly equating to 6k Lumens) am I going to run into a heat issue with 30 min+ run times??

Mart

English_Mart said:

Would I be correct in assuming that generally the more lumens any of the typical super bulbs we might be using - the more heat is going to be generated.

Click to expand...

Generally that's correct. The IRC (64447) is a little different. It has an IR reflective coating that allows it to keep the filament hotter using less power than other bulbs. This also reduces, somewhat, the emitted IR. I still melted leads that were nearby.

English_Mart said:

So, inevitably, a 8000 lumen bulb will have to be run for shorter periods - or are there some bulbs that seem to run colder whilst still kicking out stupid levels of light??

Click to expand...

Yup. I've begun using aluminum foil (to reflect IR inside the light) and ceramic refractory insulation to increase runtimes before it gets too hot. It works quite well.

English_Mart said:

Hmmm, more about the 64447, looking at a thread you commented in, Bill you mention that this bulb draws 7amps, so this is looking increasingly like a good bulb for me to try, has the axial filament as Jimmy mentioned, too. But if I drive it at around 18V (possibly equating to 6k Lumens) am I going to run into a heat issue with 30 min+ run times??

Click to expand...

The 625 is a whiter bulb. The IRC 65W run at 19 volts (15 1/2Ds holt 19v quite nicely) is super bright, but when I look at it side-by-side with the 625, the 625 is whiter. I'm using different reflectors to the beam patterns are not comparable.
You'll never get a 625 or 447 to run for 30+ minutes regardless of power supply. Heat becomes the real problem.

I think we may be getting away from your original question. If you are dedicated to the 6D Mag, you're kind of limited. There are powerful lights that are smaller but have limited runtime. There are bigger lights that are more powerful and have great run time. If you really want to stuff the 6D your options are limited. The 6D will only hold 9 1/2Ds. That's Mag85 territory.

I'm not exactly sure about "significantly brighter", but the 64275 and 5761 are definitely brighter and can run on 6D NiMH cells and heat is not a major condern.