Mag85, how many REAL lumens

juancho

Enlightened
Joined
Feb 26, 2004
Messages
797
Location
Long Island, New York
Thanks to Hookd_On_Photons and Ginseng for this summary of the bulbs:

WA01111: 6V spec (464.95 lumens), 7.2V push (880 lumens)
WA01160: 5V spec (326.72 lumens), 6V push (618 lumens)
WA01183: 4.7V spec (410.92 lumens), 4.8V push (442 lumens) (can't take much overdrive)
WA01185: 9.6V spec (816.81 lumens), 10.8V push (1234 lumens)
WA01274: 7.2V spec (552.92 lumens), 8.4V push (946 lumens)
WA01318: 9.6V spec (534.07 lumens), 10.8V push (807 lumens)

So, since we are using 10.8 volts, it is 1234 lumens???

Juan C.
 

270winchester

Flashlight Enthusiast
Joined
Aug 9, 2004
Messages
3,983
Location
down the road from Pleasure Point.
Well, that's the bulb lumen, so it sounds about right.

but from real world use, even if the efficiency is not that high from loss of lumens to the reflector and lens, it is a HOT piece of light. the gh24, when focused, can barely cast a small portion of the hotspot that the Mag85 casts...and less white too. I wish I had a 3 inch head to use with it, but my other hobby(knives) is consuming the rest of my disposable income...(those little Karambits are expensive!!!)

Nick
 

Kiu

Enlightened
Joined
Jan 11, 2003
Messages
451
Location
Hong Kong
There is large voltage sag of 9AA nimh cells when driving at such a high current.
I will rated my mag 85 within 500-700 SF's lumen with fresh 9AA nimh.(I could be totally wrong) I think that a ceiling brounce off test with a lux meter can give a good picture of total light output.
 

bwaites

Flashlight Enthusiast
Joined
Nov 27, 2003
Messages
5,035
Location
Central Washington State
The Mag85, on fresh, consumer grade cells puts out between 700-850 Lumens out the front of the light.

On high current cells it might be slightly more.

Bill
 

JimH

Flashlight Enthusiast
Joined
Apr 8, 2004
Messages
2,714
Location
San Jose, CA
Bill,

What about driving the 1185 with batts that measure 10.8 amps under load (approx 3.8 amps)?
 

Ginseng

Flashlight Enthusiast
Joined
Feb 27, 2003
Messages
3,734
While the bulb output rerating formulas are pretty well defined, the actual output "in the torch" can be dramatically different and always in the "less" direction. Welch Allyn uses a rerating formula that's been posted many times so I'll decline to post it here. In any case, output goes up dramatically as a function of driving voltage.

However, some correlation studies carried out by PaulW some time ago indicated that there might be as much as 60-65% loss when the bulb is installed in the light. He arrived at this figure by comparing against SureFire lights and assuming these lights put out the honest to goodness lumens "out the front" as they advertise. The strong suspicion is that SF rates their output with the bulb installed in the torch rather than as a bare bulb.

So, we generally use 60-65% of bulb lumens as the actual torch lumens. So, if the WA01185 is making 1,234 bLu, that translates to 740-802 tLu. I've said it before and let me reiterate: <font color="blue">There is no way to know the actual lumen output of a torch without installing the light in an integrating sphere spectrophotometer (ISS) and measuring it as a complete system, all losses included.</font> Unfortunately, it is much easier to measure lux (ie. hotspot intensity)

The ceiling bounce test is a much better analog to an ISS measurement than is looking at the beam against a white wall.

Wilkey
 

Luna

Enlightened
Joined
Dec 27, 2004
Messages
874
So the Mag85 should only be a little bit brighter (for the lack of a better term) than the M6 or 10x HOLAs?

I can't seem to find any comparison beamshots. Anyone?
 

Ginseng

Flashlight Enthusiast
Joined
Feb 27, 2003
Messages
3,734
I exactly wouldn't call 50-60% "a little bit." With a texturized reflector, you get much better throw and good flood as well. Runtime is also superior to an M6 with better color and output delivery. You can get flatter output from the M6 but you'd have to step up to a regulated pack from Jim Sexton.

I think Silverfox might be able to provide beamshots of the TL85 (an analog based on the Tigerlight platform) and a SFM6.

Wilkey
 

Luna

Enlightened
Joined
Dec 27, 2004
Messages
874
Ok so 'a little bit' is too loose. /ubbthreads/images/graemlins/smile.gif I guess it would be like shining a M6 and plainjane magcharger together. The MC doesn't make that much of a difference when you shine them both in the same spot.

I still have a hard time believing that a properly mirrored reflector would be as inefficient as is suggested by a 1234lumen->800lumens out when you consider the lens is 92% efficient. Any clue as to why it is common to accept 40% losses? I have no problem with UV and IR transmission problems but a decent reflector shouldn't be anywhere near this bad.


To me it just speaks volumes for the performance of the M6 if the reflector/lens really account for that much lost light.


edit:
Do you get better throw with a texturized reflector? I tend to get better with the smooth.
 

StainlessSteel

Enlightened
Joined
Oct 18, 2004
Messages
368
Location
PA
I have been reading these Mag85 threads for a while... and still, I haven't found the answer to this...

How many Lux will a Mag85 give you??

SS
 

Ginseng

Flashlight Enthusiast
Joined
Feb 27, 2003
Messages
3,734
Luna,

Keep in mind that while we cannot provide torch lumen measurement values, SF do not provide bulb lumen values. Thus, the comparison between the two quote systems is lacking any direct linkage. Only the eye can tell. As for the Mag, that the correlation indicates 35-40% loss does not mean that there are actually 400 lumens dissipated as non-light. It simply means that that is the variation between estimate and calculated value. A loose connection at best.

SS,
Lux is not that relevant in a high power hotwire mod. Unlike with even the most powerful LEDs, a superbulb such as the WA01185 will create a hotspot that is almost painful to the unshielded eye at 1 meter. However, I think S4MadMan measured 50-60 klux for the significantly lower output WA01160 which puts 350-400 lumens out the front.

Wilkey
 

udaman

Banned
Joined
Feb 13, 2004
Messages
381
[ QUOTE ]
Kiu said:
There is large voltage sag of 9AA nimh cells when driving at such a high current.
I will rated my mag 85 within 500-700 SF's lumen with fresh 9AA nimh.(I could be totally wrong) I think that a ceiling brounce off test with a lux meter can give a good picture of total light output.

[/ QUOTE ]

Umm, I do not know about what we use to measure the relativity of "good" picture of total light output, but there is voltage sag of varying degree with different types, year of manufacture/versions, chemistries, sizes, capacities of batteries. Good way to approximate voltage under load would be to extrapolate data gathered by SilverFox's tests. The more batteries of the types most favored for use, that SilverFox can test; the more knowledge we have about how at least the batteries handle current loads, and how much they suffer from voltage sag. But SilverFox does not test batteries installed your own particular light!

[ QUOTE ]
JimH said:
Bill,

What about driving the 1185 with batts that measure 10.8 amps under load (approx 3.8 amps)?

[/ QUOTE ]


This would only be a guide of course, how the batteries are setup in the flashlight will affect how much voltage and current actually makes it to the bulb itself. Spec's you read are for lab tests, most likely done by the bulb manufacturer under their designed controlled conditions. So your power source (battery) and the electrical pathway to the bulb will determine just how much voltage/current gets to the filament. How much voltage does a bulb in a M*g2C get, a M*gChr, or a M*gD cell body; is also dependent on the electrical pathway, whether or not is is stock or modified(bi-pin socket from a M*gChr put into a M*gD or M*gC, M*gD-cell switch assembly with cutoff pedestal and soldered switch contacts as in bwaites Mule/UnStealth Light)? This can and will make a difference as to whether or not you get X amount of lumens out the front of the flashlight, whether or not the light is white or a bit more yellow, or you end up instaflashing a bulb. Having CBP1650ma AA high-current capable batteries, which is what I think Jim H is trying to imply in his short post; is but part of the puzzle, not the sole determinant...sadly JimH, it's not as simple as we would like it to be /ubbthreads/images/graemlins/frown.gif


[ QUOTE ]
Ginseng said:
While the bulb output rerating formulas are pretty well defined, the actual output "in the torch" can be dramatically different and always in the "less" direction. Welch Allyn uses a rerating formula that's been posted many times so I'll decline to post it here. In any case, output goes up dramatically as a function of driving voltage.

However, some correlation studies carried out by PaulW some time ago indicated that there might be as much as 60-65% loss when the bulb is installed in the light. He arrived at this figure by comparing against SureFire lights and assuming these lights put out the honest to goodness lumens "out the front" as they advertise. The strong suspicion is that SF rates their output with the bulb installed in the torch rather than as a bare bulb.

So, we generally use 60-65% of bulb lumens as the actual torch lumens. So, if the WA01185 is making 1,234 bLu, that translates to 740-802 tLu. I've said it before and let me reiterate: <font color="blue">There is no way to know the actual lumen output of a torch without installing the light in an integrating sphere spectrophotometer (ISS) and measuring it as a complete system, all losses included.</font> Unfortunately, it is much easier to measure lux (ie. hotspot intensity)

The ceiling bounce test is a much better analog to an ISS measurement than is looking at the beam against a white wall.

Wilkey

[/ QUOTE ]

I will agree with most of Ginseng's contentions here (and it's a better explanation he gave me when I asked him how he came up with his figures last year), except for the ceiling bounce or white wall, as neither are particularly accurate. Analog it maybe, but the analogy does not make for good quantitative comparison. Not very scientific, angle of beam can be repeatedly set to the exact same spot and distance? Reflectivity of ceiling is a known amount? Too much of a 'fudge' factor going on here for me to feel assured that the ceiling bounce is giving much in the way of accuracy. Measuring Intensity of a hotspot with a lux meter is not such a bad way of determining relative brightness as far as hotspot is considered for comparision purposes.

Honestly, most of this lumens output thing is not very impressive if you are only speaking in terms of small percentage differences...you cannot see those differences for the most part until they become large enough. How much is large enough? Well I suppose the Coleman 12v/100w SLA spotlight is one comparison. It has a much larger reflector, but should be putting out as many lumens as the WA1185, if not more. Now there is this pushbutton switch on the Coleman that will get you supposedly 1/2 power. But do your eyes see an apparent 1/2 difference in output? Not even close, the difference appears to be minor rather than large. So if going from 1,000 lumens, and hitting a switch which drops output in 1/2 doesn't seem like it's all that much, how do you suppose you'd be able to distinguish output changes of 100 lumens or less?

And BTW, 'out the front' lumens, at what distance is this presumably measured (10mm in front of the lens or 1,000mm?), knowing that the light intensity drops off at a formula of 1/4d squared (I don't know how to post that formula and forget the exact terms)?

A lot of this discussion is all academic at best; and consideration of filament width as it relates to how tight a hotspot can be obtained, is far more important to me. The flash of a camera, either electronically generated by a tube, or by one-shot bulb will temporarily blind you with intense output. A big fat wide filament will get you huge flood, but is that all you want from your light?

I agree with Ginseng's contention that you need a light meter to do these comparisons and that you could use measuring the celling bounce, but where do you put the sensor for the light meter then? All just guides to actual lumen output, and WA's rerating formula's, they are just guides also. Do they publish an figures on how accurately that re-rated formula corresponds to actual lumens output for all scenarios? Has anyone actually tested and measured these assumptions we are making about a generalized formula?

So when do we start the fund raising drive to get an inexpensive integrating sphere measuring device, administrated by Quickbeam perhaps, with relatively low cost testing of any submitted light (with partial proceeds of these tests to fund CPF?). That way, someone could submit any damned battery/bulb combination, in any flashlight, pay a fee, and have it tested for real world tested lumens output? No more haggling over what this or that outputs. Here's a thread that tvodrd started that not too many paid attention to Integrating sphere. Sorry but I don't have time to track down the link for a inexpensive spectrometer sold by an USA company, made in China, runs about $1,500...someone posted a link to the company site on one of the many aquarium sites I skimmed over 2 years ago.

I'll add my admonition, and state the need for this in Ginseng's website page on M*gMods, the statement again....IT DEPENDS...what are all the components that make up your light outputting device; only when you have understood and analyzed the whole, will you be able to estimate true output, whether or not you will get bright white/dull yellow or instaflashed bulbs /ubbthreads/images/graemlins/frown.gif.

Ginseng has posted charts before, NiMH will charge up to about 1.45volts, and that is the brief temporary (if you don't instaflash you bulb) starting voltage. Underload, the battery quickly drops to some other voltage which maybe greater than the 'nominal' 1.2volts the NiMH are rated to supply. Before the batteries use up the full capacity, they drop below 1.2v. So you are getting a range of lumen output, and unless you have regulated voltage as is the case with the js M6-R mod, the bulb will change from bright white of varying color until it dims to a more yellow color. The higher the current demand on the battery (given equal capability, which we know can varying greatly between RC style high-current batteries and consumer batteries) the steeper the voltage drop slope. So you go from white to dimmer yellowish light more noticeably. One exception would be a type of Lithium battery SilverFox tested from PeLu in the alkaline battery tests...straight line output as though it was regulated!

Examine, comprehend SilverFox's test data:
123 Battery Shoot Out
Alkaline Battery Shoot Out


Quoting the world's most influential (but many of times pretty much full of himself thinking he is an expert on the level with winemaking gurus like someone I know, Burt Williams... formerly the famed icon of the 'Ultimate Garage Winery' ---Williams-Selyem Winery) wine writer Robert A. Parker:
An Asian proverb says it all, "Knowing in part may make a fine tale, but wisdom comes from seeing the whole."
http://fora.erobertparker.com/cgi-bin/ultimatebb.cgi?ubb=get_topic;f=15;t=004883;p=0


PS, JimH if you read the wine thread linked to here, while my writing style is similar to Robert A. Parker, the similarity ends there; easy to tell the difference...he's a former attorney prone to using excessive hyperbole, and absolute plethora of mindbogglingly repetitive adjectives; in his reviews in his bi-monthy publication The Wine Advocate /ubbthreads/images/graemlins/wink.gif
Re: the never ending debate on the significance of 'terrior' in a wines' character.
"So, like, which of you two is correct?" /ubbthreads/images/graemlins/crackup.gif
 

CroMAGnet

Flashlight Enthusiast
Joined
Sep 4, 2004
Messages
2,540
Location
Los Gatos, CA
I have noticed that Udaman has brought up that "we start the fund raising drive to get an inexpensive integrating sphere measuring device, administrated by Quickbeam "

If CPF has over 7000 registered members and we all chip in $1, we could have one of those fancy contraptions and some extra change for a few highend passarounds.

So just in case you guys are /ignoring this, I thought I'd ehem... illuminate the good idea /ubbthreads/images/graemlins/smile.gif
 

Luna

Enlightened
Joined
Dec 27, 2004
Messages
874
[ QUOTE ]
udaman said:
Ginseng has posted charts before, NiMH will charge up to about 1.45volts, and that is the brief temporary (if you don't instaflash you bulb) starting voltage.

[/ QUOTE ]

And not just Nimh. The other day I instaflashed an Osram 30w with a new stock nicad Magcharger battery.


As for lumen output: You can take 2 lights with 4x difference in the output and project them on a wall at lets say 10ft. With and assumption of equal characteristics but 250lumen vs 1000lumen, the weaker light might project a hotspot of a 1ft across and the 1Klu may only be 2ft across. It isn't always an impressive difference especially since the reality is that the filament tends to be longer for a higher output bulb or else life is compromised. I tend to like whiter more intense hotspots over yellower wider hotspots. Of course there are limits to or else we would all just buy a pocket white laser and be done with the obsession. /ubbthreads/images/graemlins/smile.gif I need a Maxabeam or Megaray /ubbthreads/images/graemlins/grin.gif
 

NewBie

*Retired*
Joined
Feb 18, 2004
Messages
4,944
Location
Oregon- United States of America
[ QUOTE ]
Ginseng said:
While the bulb output rerating formulas are pretty well defined, the actual output "in the torch" can be dramatically different and always in the "less" direction. Welch Allyn uses a rerating formula that's been posted many times so I'll decline to post it here. In any case, output goes up dramatically as a function of driving voltage.

However, some correlation studies carried out by PaulW some time ago indicated that there might be as much as 60-65% loss when the bulb is installed in the light. He arrived at this figure by comparing against SureFire lights and assuming these lights put out the honest to goodness lumens "out the front" as they advertise. The strong suspicion is that SF rates their output with the bulb installed in the torch rather than as a bare bulb.

So, we generally use 60-65% of bulb lumens as the actual torch lumens. So, if the WA01185 is making 1,234 bLu, that translates to 740-802 tLu. I've said it before and let me reiterate: <font color="blue">There is no way to know the actual lumen output of a torch without installing the light in an integrating sphere spectrophotometer (ISS) and measuring it as a complete system, all losses included.</font> Unfortunately, it is much easier to measure lux (ie. hotspot intensity)

The ceiling bounce test is a much better analog to an ISS measurement than is looking at the beam against a white wall.

Wilkey

[/ QUOTE ]


Has anyone tried one of those spendy low loss reflector material coatings, instead of the high loss flash coated flashlight reflectors, yet?

Or polished an aluminum reflector with some good jewelers polishes, while keeping it an inert atomsphere of nitrogen, so that it doesn't instantly oxidize due to the oxygen in the atomsphere (and rapidly causing reflectance value to drop)? Of course you have to always keep it in a nitrogen atomsphere, even afterwards....
 

Donovan

Enlightened
Joined
Jan 6, 2005
Messages
420
Location
North Metro Atlanta, Georgia
[ QUOTE ]
juancho said:

Thanks to Hookd_On_Photons and Ginseng for this summary of the bulbs:

WA01111: 6V spec (464.95 lumens), 7.2V push (880 lumens)
WA01160: 5V spec (326.72 lumens), 6V push (618 lumens)
WA01183: 4.7V spec (410.92 lumens), 4.8V push (442 lumens) (can't take much overdrive)
WA01185: 9.6V spec (816.81 lumens), 10.8V push (1234 lumens)
WA01274: 7.2V spec (552.92 lumens), 8.4V push (946 lumens)
WA01318: 9.6V spec (534.07 lumens), 10.8V push (807 lumens)

So, since we are using 10.8 volts, it is 1234 lumens???

Juan C.

[/ QUOTE ]

Where is this original thread with the bulb/volts matrix?
 

davidra

Enlightened
Joined
Feb 17, 2004
Messages
605
[ QUOTE ]
bwaites said:
The Mag85, on fresh, consumer grade cells puts out between 700-850 Lumens out the front of the light.

On high current cells it might be slightly more.

Bill

[/ QUOTE ]

For newbies, can you give us some specific batteries that are in both categories? And which ones you would recommend? I realize it won't work with alkalines at all (I hear), but which AA's work well in this mod?
 

Lurveleven

Flashlight Enthusiast
Joined
Dec 21, 2004
Messages
1,237
Location
Bergen, Norway

Latest posts

Top