Improve Mag Beam Quality: Frosted Glass Mod

With all this talk about frosted acrylic, frosted glass and the new composite texturized lenses by FlashlightLens.com, I decided to take matters into my own hands. Here's an easy way for all you incandescent afficianados to dramatically improve the beam quality of your Mag torches in flood and spot mode.

Materials:
1. Small jar of Armour Etch hydrofluoric acid glass etching paste
2. Synthetic bristle paintbrush
3. Rubber gloves
4. Lightbulb

Method:
1. Clean bulb capsule with alcohol to remove oils
2. Brush on a thick layer of Armour Etch
3. Wait 5-7 minutes
4. Wash off acid paste with running water
5. Install bulb in torch
6. Enjoy!

This first picture depicts the improvement in spot mode. On the left is a Mag2C with a KPR112 on 3x123. Bright but very obvious spot artifacts. A bit washed out in this photo but you can see the filament shape, and all sorts of swirls and arcs even tightly focused. On the right is the frosted bulb, a KPR118. The spot is extremely even, blending relatively smoothly from spot to inner spill to outer spill. The closest thing I have that it looks like is a 1W LS through an NX-05 optic. Below is a picture of the Armour Etch ($9.50 from a craft store) and a clear and frosted bulb.


The second picture is of the two lights in medium wide flood. You can see the dramatic improvement in illumination quality. While not quite as efficient a diffuser as the SF holographic diffuser filters, it is still pretty darn effective. The diffused spot looks a different color and dimmer than the clear bulb because the KPR118 bulb is not overdriven like the clear KPR112. I did not have an extra KPR112 on hand to do a direct comparison but it would be closer than what you see here.


At wide flood, the Mag hole is still there, but much softer. Obviously this mod would not work for LEDs but for anyone wanting an easy method to smooth out their incandescent beams, this could be it. While I did not do it (I only have one Borofloat lens) you could also etch a glass lens and achieve pretty much the same effect.

Wilkey

restored 12/21/06

Wilkey,

Ah, this is something I have been looking for. And being in a paste form it won't be quite as dangerous (although I'm still going to be very careful). As you pointed out a while ago in a different place, hydrofluoric acid is very nasty stuff. More experimentation on my plate now. I guess that today I'm going to be spending my afternoon searching the local craft stores. /ubbthreads/images/graemlins/grin.gif

I'm thinking about whether there could be value in applying some of the paste to the tip of bulbs with lenses or with imperfections in the glass -- those bulbs that produce the jagged smile beam center. It might clear up that problem but without spreading the beam as much as if the whole bulb were frosted. It could be the solution to cleaning up beams from bulbs like the Carley 805.

Thanks for sharing this promising technique with us.

Paul

Be interesting to compare how effective this is compared with the writerite-on-lens solution and also the new LDF lenses.

anyone try using fine abrasive paper to frost the glass?

hideo

I don't think it's a good idea to decrease the integrity of the bulb through abrasion, as the heat from the filament may cause the bulb to burst, or shorten its life.

I posted a number of months ago that I regularly bead blasted M*g bulbs to improve beam quality. I've done this since at least '86 with no adverse effects.

Finally, though, I have no reason to use M*gs any longer /ubbthreads/images/graemlins/smile.gif

PaulW,
We think alike I think. /ubbthreads/images/graemlins/smile.gif I treated a CL717 lensed-end bulb to try and tame the dreaded "grinwurst" and it works...unbelievably well. The filament projection is completely diffused. And since I only acid etched the lens tip, the hotspot is a tight and as intense as ever. This proves that the construction of the primary hotspot is due to off-angle light that is captured by the reflector. The virgin light contributes only to spill. This simple treatment provides a route to make lensed-end lamps much more aggreable.

I have been doing more experimentation this afternoon and I have found something out that could limit the utility of this technique. The acid paste, though easy to apply and thick enough not to run, works well on silica or quartz glass. I tried a variety of bulbs and lenses and found that it is completely ineffective on borosilicate glass. This means Pyrex, Borofloat and some bulb capsules! For example, the CL809 bulb shows no etching after over 8 minutes exposure. Likewise, I tried to etch the outer perimeter of a Flashlightlens.com Borofloat glass lens and failed. So, this means that the only way to know for sure will be to try it. One could hazard a guess that high output bulbs which would have to withstand large swings in temperature or high temperature might use borosilicate glass. When I get my WA bulbs back from Carley, I will try it with one of my 01318s.


kakster,
I would be willing to etch a reference bulb for someone to do a head to head comparison. For example, if someone wants to volunteer to run beamshots of an etched KPR112 vs. clear bulbs with WriteRight vs. LDF vs. Acrylite on the lens that would be ideal. Even moreso if that person had instrumentation to measure light output.

Hideo,
I tried using 3M Royal Wet/Dry 2000 grit on the CL809 that I could not etch and it was tough going but I was able to scratch it up. I don't think manual sanding is very good from a uniformity perspective. Now if you wanted to use a Dremel with a sanding wheel, that might be better.

mhejl,
There are aircan sandblasting kits you can find at the craft store to blast with. But if you have professional equipment that would be better. The advantage to using acid paste is that I can easily, selectively etch only the regions of the bulb or lens I want to tailor the light pattern.

Wilkey

Well, I found some of the Armour Etch Glass Etching Cream in the second crafts store I went to. All they had was the 12 oz. size, $18.49, far more than I'll ever need. I'd like to be able to share it with others, but I don't think it's a wise thing to send through the mails.

The first thing I did was to frost four bulbs, each differently. I have a large supply of Mag 2-cell bulbs, so I used them. One was frosted on the tip, extending far enough back so that the filament can not be seen when the bulb is in the Mag head. The second got a ring of frosting around its equator, and the third was frosted in a ring near the base. The fourth was completely frosted.

I put these bulbs in a Mag 2C with standard Mag reflector. I have no camera to show you beam shots, but I'll describe what I saw at tight focus as well as I can.
<ul type="square">[*]Bulb 3 (frosting near base) didn't seem to do anything unordinary. It acted exactly like the unfrosted bulb, as far as I could tell.
[*]Bulb 2 (frosting around equator) seemed worse than a non-frosted bulb. The hot spot was more diffuse but the artifacts -- little spots and lines of light here and there on the wall – were quite sharply defined.
[*]Bulb 1 (frosting on tip) seemed best. The hot spot was still small and bright, but the artifacts were more smeared out. There were a few slight rings around the hot spot.
[*]Bulb 4 (completely frosted) had a very smooth beam – of SureFire quality – no artifacts, no rings.[/list]
But the light with bulb 4 seemed much dimmer than that with bulb 1. So I pulled out my trusty light meter, with the following results

_________________________Direct_____Indirect
With bulb 0 (not frosted).___3970 lux____4.86 lux
With bulb 1 (tip frosted)____4780 lux____4.91 lux
With bulb 4 (all frosted)_____460 lux____4.25 lux

A comparison of direct readings shows that frosting the tip only does not affect throw appreciably. But throw is degraded by a factor of ten when the equator is frosted. (I believe that it also shows the variance in brightness among Mag bulbs. Bulb 1 was obviously brighter to begin with.)

It makes sense that etching the tip is so critical. I have noticed that SureFire has rough surfaces on the tips of some of their bulbs (probably ground rather than etched). It would appear they discovered long ago what we are uncovering here. I agree with Wilkey that light directly from the filament is what causes problems. It can be focused purposely by a built-in lens or focused accidentally ever so slightly by bulb imperfections. Either of these causes at least some of the artifacts.

Wilkey – Yes, the inability to etch some kinds of glass is troublesome. I saw on the bottle of acid (or perhaps just a hydro fluoride salt) a notice that it won't work on Pyrex glasses. I would have tried it on the two WA bulbs I have (1318 and 1185), but I want to experiment with the etching process first. The 1318 in 3C is the best light I have at the time, and I'd hate to mess up the only bulb I have for it. Further, I'm not as confident slinging around the chemicals as you are.

I remember one day in high school chemistry mixing nitric acid and turpentine. After I poured the acid into the tur . . . well I'd better save that for later in the Café). /ubbthreads/images/graemlins/grin.gif

Paul

EDIT: Corrected lux numbers by adding zeros. Forgot what scale I was on.

Nice. At tighest spot it actually looks just about perfect. Tight spot with gradual spill. Good dual purpose beam in one.

Paul,

Thanks for doing my work for me /ubbthreads/images/graemlins/grin.gif Hehe, I was planning to do what you did but I'm glad you were able to extend that with light measurements. Excellent findings by the way. I am impressed by the tip-frosted bulb. It makes sense that the direct light should affect the beam quality but the degree to which you were able to improve things was amazing. Also, it is possible to gain some idea of the direct light contribution to the total beam by shining the light with the head/reflector off. Hold it close to the wall and you can see the direct light without the muddling influence of the reflector.

I wonder how a base/tip frosted bulb would perform.

I think a different effect can be arrived at by diffusing at the main lens. To that end, I'm going to try and find a glass lens I can etch. Mineral glass should work. I think there would be more light lost by diffusing at the lens however. Somebody somewhere suspected that better results can be had by placing the diffusion as close to the source as possible. I think this is true, since it would result in greater collection of the refracted rays by the reflector.

Another experiment you might want to try is to vary the etch time. I found that 8 minutes results in a deep etch. On the Mag2C bulbs, it etches a grainy, rough surface. On the CL717, however, the same time results in a beautiful, satiny etch. Also, I think that two-stage etching results in a rougher surface than single stage. So many variables!

Yep, I've played with nasty stuff. I think the two worst chemicals I've worked with were phosgene and tetrachlorodioxin. Oh yeah, elemental bromine was very, very nasty as well.

Great work Paul. Nicely done /ubbthreads/images/graemlins/thumbsup.gif

Wilkey

time to move this thread ... I'm sure it is now the subject of a federal inquiry

... in a teenage life punctuated with strong oxidizers and pyrotechnics, white phosphorus and metallic potassium, the time I laid out a quarter pound of potassium permanganate and doused it with warm glycerin was a standout ...

ooooooooohhhh ..... it burned, IT STUUUUUNK!! /ubbthreads/images/graemlins/icon15.gif

hideo

Hideo,

It sounds like you misbehaved like the rest of us. That's what the teenage years are for, I guess.

I certainly enjoyed playing with chemicals in the early days. I got smoke. I got fire. I got boom. But, you know, I now find this flashlight thing is far more satisfying. It's the same sort of thing, but more subtle and more controlled and therefore more powerful -- in a different way. I still don't understand everything that that means, but I'm learning. /ubbthreads/images/graemlins/grin.gif

Paul

Would be nice with a little red button on the flashlight that said "self destruct", though...

Dang! I have to go look for some Armour Etch tomorrow!

CURSE YOU Wilkey.... LOL!!!!

Although I still think Writeright is easy and works darn good! /ubbthreads/images/graemlins/blush.gif

P.S. Hydroflouric Acid (not in widespread use anymore - had a tendency to leach into the skin and mess up bones!) was probably the most nasty stuff I ever got close to. /ubbthreads/images/graemlins/ooo.gif

White phosphorous? Nothing like burning metal embedded in flesh. Some guy dropped a piece of metallic sodium down a drain in the lab where I did grad work. Let's just say the combustion of released hydrogen was, um impressive and resulted in some impressive damage to black iron pipe.

A full logic flashlight that could be programmed to read the operator's fingerprint to enable operation would be cool.

Joe, what can I say...curiosity loves company /ubbthreads/images/graemlins/wink.gif HF is great with extra virgin olive oil over a bed of tossed greens.

Paul, I didn't know you were a philosopher?

Wilkey

Real cool but I thought that frosted lamps were made already? I have to admit your beam may be a little better then the frosted lamps I have seen so I guess I just answered my own question.
Nice work and an easy modification! /ubbthreads/images/graemlins/bowdown.gif

Yep,

Soloco has purchased frosted lamps from Carley. He really seemed to like them. Forgive me if I can't dig up the thread. The ability to try all sorts of frosting patterns as PaulW has is a real boon to the tinkerer.

Wilkey

I'm presently working on a nice HPR51 halogen bulb, 6.5V 0.7A that unfortunately has extreme rings in the beam due to a flat headed capsule and a pronounced nipple. The glass is etching, just not very easily. I will give it a bit more tomorrow.

Wilkey,

As I was drifting off to sleep last night -- at about the time you were posting about etching an HPR51 bulb -- I was thinking about doing a little on the tip of one of the LAs I have for my SureFire A2. You may remember that Nascar posted on another thread that he was displeased with the large ring around the A2's beam. I am too. I think that just a touch of etch in the right place would clear that up. Ha!

I hope that this thread will become a repository of people's experiences with different bulbs and different techniques.

The word "techniques" reminds me. The bottle of etching cream says to use <font color="red">PROTECTIVE EYE COVERING.</font> I think this is very important. Even though I wear glasses, I put goggles on over them -- far more important, I think, than the rubber gloves.

Regarding being a philosopher, wasn't it the neat Mortimer Adler who said that we're all philosophers, especially taxi drivers? And here at CPF we have a lot of philosophers, some of whom cause me to do some double takes and scratch my chin. /ubbthreads/images/graemlins/smile.gif

Paul

I think tip etching is really becoming a useful technique. I etched the HPR51 today and it was very effective in diffusing the arcs. Still, because of the very sharp "shoulder" at the bulb top, I was unable to smear it all out. Perhaps a little more acid.

Protective eye covering is extremely important in what we are doing. Unlike a piece of plate glass, the objects we are etching are pressurized! After I coat the piece, I place it base down in a foam block to immobilize it then cover the whole thing with a plastic bucket. Just in case an inherent flaw coupled with the etching action causes the glass to fail. I cannot emphasize the safety aspect enough.

Funny you should mention philosopher cabbies. We have a column in the Providence Monthly penned by a cabbie. In it, he mixes his hack-based street wisdom with the inanity of fare chatter. Quite an entertaining mix. I studied philosophy in school, mostly philosophy of science. Dennett was a favorite read of mine.

Wilkey

I'm wondering about the improvement due to frosting the tip only. The light emitted through the tip is not reflected by the reflector and is too diffuse to aid much in the overall beam, so I'm wondering if frosting the tip may actually result in otherwise spilt beam to be reflected back into the reflector. Anyone have the ability to mirror the tip on the inside?

While I believe frosting provides a significant improvement, I think the best improvement that you can make to the bulb is carefully centering the filament within the reflector by bending the flange on the base. Many filaments are not well centered within the bulb, and the bulb is then not well centered in the base.

You'd think that with manufacturing technology these days, you should be able to get perfectly centered filaments, and not have to frost bulbs, stipple reflectors, or use diffusing lenses ... except to soften the filament image.