Replacement of carbide lamps by LED lamps

[gillestugan]

Yes, the lumen output is of a carbide lamp is more than enough in most situations. And with a 10cm flame you probably get more than 100Lm. But that is really unimportant. I was just trying to make a comparison for hank.
Regarding the pointed forward led: Some light do will hit the reflector. If the 4" premier reflector is 20m deep it gives an angle from bottom to rims of about 130 degrees. I you put a SSC P4 in the middle 12% of the light would still hit the reflector. (you,re right, it's useless.) And if pointing it towards and old reflector you will probably get a diffused beam, It'll be "almost" like an orange peel reflector, just that is made out of oxide and chrome peel
And yes. Nothing I've seen beats the colour rendition from a carbide lamp. But its getting better. I have a 3000k cree in a petzl micro mod that is very warm, just a little brownish in the tint.

Good idea with the water-control lever for brightness. Many lights have fixed steps, which would make it very convenient if used with a potentiometer. (for example a PT4150 driver) And keeping the batteries loose inside the carbide container will give you the rattle of a full light. That would actually be funny, but I too would not be able to bring myself to gutting one. Not a Premier, but maybe a Justrite... hehe

 

[Barbarin]

I've been thinking about a "CarbiLED" too. In fact that would be the most similar to a pure flood style light. (as carbide is)

Let's imagine. Using a 70 mm diameter alloy disc , 7-8 mm thick, with some finning on the back would allow enough heatsink . A protective "frozen" polycarbonate lens like 3 mm for a single XR-E (maybe 3 in series to reduce glare and improve efficiency) , and the batteries placed on a kind of "generator" holding 4x18650's or alternatively 8 x CR123 , with a potentiometer from 1% to 100% ( 10-1400 mA). Runtimes from 6 hours at 300 lm;24 hours if we just need to be over a good carbide specs (more than 100 lm) ; and ranging from one day to two weeks on low levels.

The generator hanging from your waist, hitting every rock, that flat disc on your helmet will look like an old style flamme reflector, the control next to your waist, the cable... That would be for sure really similar to carbide. (Not replacement, but similar). And simple and "cheap"* to make. Sorry, no smell, no heat.

In fact making a pure flood light will make things really easier... and if you just need a "thrower" for very short moments, why don't carry it on a pouch?? Just a different approach.

Not the purpose of this thread, as the idea is to get the best lamp available designed for the needs of caving, with the technology we have today. But just wanted to talk about it.

Javier

*It could be made cheaper if using no driver, DD resistored down with the potentiomenter, and a 3 X D NiMH battery pack with this battery holder on the 3D version and "moddified" to be IP67 at least.(The C version looks better and can be used for 26650's ).

 

[gillestugan]

Regarding the anti-glare:
I just made some tests with a piece of lightly frosted glass from a reflector bulb. The loss was very small and it reduced the glare significantly. Unfortunately it is frosted on the inside, so it wont be possible to glue it on top of the led with optical resin, and it is very fragile in it self. Would be nice to find some kind of thicker and curved lens that is just as lightly frosted.
Do you think it would be possible to diffuse the Carclo 10403 120 deg Bubble optics?
The Khatod PLJT20 20mm Diffused dome has been mentioned, but I am afraid it will give too low output. Especially as it is 170 degrees. The test report doesn't give any information on the output, as the output the led used in the test is not specified.

EDIT: I just saw there is. But I cant understand and match the numbers to what is said in the text. Someone please explain. How much lower is the lux level of the LED with the lens on? 1/3,2?

 

[Tobias Bossert]

I will complete the test report. Gottfried, the tester of test device_2 ('Bucherl'), took some photos. With the link you can compare Scurion P4, Petzl Ultra and my test device_2 (Tobi) side by side.

http://www.hirlatz.at/lampenvergleich.html

"Anmerkung 1:" = Note 1
"comparison of light distribution 1


left: "Scurion P4" flood on level 3,
right: "Tobi 1" work-light low


The light of Scurion decreases strongly with distance, with Tobi's worklight you can see more far. This is achieved by the light-up with smooth transition."


"Anmerkung 2" = Note 2


It is the same text but comparing Scurion P4 maximum flood (level 4) with Tobi's work-light high.


(Comment: Please realize that level 4 of Scurion is about 1A to the LED, but my work-light is 295mA + 91mA only.)


The room in the cave is covered with very dark loam, absorbing a lot of light. The caver stands immediately beside a wall (left to him). The floor decends slowly, increase again a little bit and - in a distance of about 15m - falls again and thus is hidden. The wall at the other side of the large room is far away. With the adapted eye you can see it with each light ghostly, but in the photos with fixed exposure it seems to be totally black - with the exception of my shaft-light.

My personal conclusion from these photos is, that it is NOT possible to use a two-LED construction with fixed beams! For normal use as a work-light, a relativly narrow beam makes no sense - even when combined with a diffuse light: You need a very soft transition from beam to flood and the beam should not reside in the centrum of the flood.

I'm verry interested in Javiers two-LED solution with zoom - but I'm jet not convinced that the transition of the superposition of both parts would be smooth enough to use it round the clock. I'm not shure whether the two-LED solution will be easyer to construct and will have lower weight as the three-LED solution with fixed diffuse/medium-beam/extra-narrow.

But never the less, I'm very eager to see some photos from Javier dealing with this. He is experienced in manufacturing and he also convinced me with his zoom-light: Much better than all I have seen in the past.

I agree with the schematic drawings of Javier completely: That's what we need.

At the moment I'm still looking for spherically domed front glass with diameter 18 to 20mm and dome height of 4 to 5 mm (spherical cap with 90 deg opening) to make some experiments with domed diffusers. This diffused light would be need to avoid glaring anyhow, whether we end up with a two-LED or with a three-LED solution.

Tobias

 

[uk_caver]

My personal conclusion from these photos is, that it is NOT possible to use a two-LED construction with fixed beams! For normal use as a work-light, a relativly narrow beam makes no sense - even when combined with a diffuse light: You need a very soft transition from beam to flood and the beam should not reside in the centrum of the flood.

Maybe it's not *ideal*, but a fixed two-beam solution is not only possible, it's what most people would consider as good enough.
If you mix a flood beam with a rather narrow spot, even a spot that's really too narrow for optimal floor-lighting close-up *and* with a fairly defined edge, that still seems to work as a usable light, since the flood lights up nearby things adequately, if not perfectly.

At the moment I'm still looking for spherically domed front glass with diameter 18 to 20mm and dome height of 4 to 5 mm (spherical cap with 90 deg opening) to make some experiments with domed diffusers. This diffused light would be need to avoid glaring anyhow, whether we end up with a two-LED or with a three-LED solution.Tobias

Practically speaking, I think even having a small and somewhat matte 'reflector' round a flood LED *might* make the glare problem less severe.
Though I'm only speaking from a subjective viewpoint, after I (and most people I cave with) moved from using naked LEDs for flood to LEDs with a small reflective collar, I got less feeling of the other people's flood LEDs being annoyingly glaring.

Possibly it's just that a larger source is somehow easier to avoid looking at than a bare LED, possibly it's just that over time I got gradually better at avoidong looking at other people's lights and the small reflectors had nothing to do with it.
However, when making lights, I do find that there's something about naked LEDs that just doesn't seem to annoy my eyes enough to make me automatically look away from them.
I wouldn't be surprised if there's some kind of brightness-avoiding reflex that isn't properly triggered by very small sources

 

[gillestugan]

Possibly it's just that a larger source is somehow easier to avoid looking at than a bare LED, possibly it's just that over time I got gradually better at avoidong looking at other people's lights and the small reflectors had nothing to do with it.

The smaller the light source is, the smaller is the area it is hitting on the retina when focused by the eye. The smaller the area the lumens hit, the brighter and more annoying it it is to the eye.
This is a problem when you want to make a small and light lamp, as you ideally would have a very large dome on your head

Looking at the pictures (thank you Tobias!) I agree that it is not working very well with the two lamp system. That is with those very narrow optics.

I still think you could get satisfying results with only two lamps, but then I would choose a wider spot and softer edge, around 10 degrees. Of course this gives a loss in shaft lighting performance, but it is not often you have to look down those terrible deep shafts. You usually know beforehand if the caves have shafts like that and can carry a small CR123a handheld for those moments. But then, one Boom medium on a MC-E for hall light would also be very nice to have.

 

[uk_caver]

The smaller the light source is, the smaller is the area it is hitting on the retina when focused by the eye. The smaller the area the lumens hit, the brighter and more annoying it it is to the eye.
This is a problem when you want to make a small and light lamp, as you ideally would have a very large dome on your head.

I take your point, however, I'm talking about surrounding flood LEDs with small semi-matte reflectors, not optics or diffusers.
With a small reflector, someone looking at the light from within ~50 degrees of the lamp axis can still see the naked LED just as before, and just as bright as before, only now with a surrounding area of lit reflector. That does seem to be less dazzling, even though the peak brightness on the retina is the same as previously - possibly it's just easier to automatically avoid looking at.

 

[gillestugan]

Thanks for the clarification, I thought you meant the light reflected by the orange peel reflector. I see your point and agree with you, although a large diffused lens is much less annoying. I have no idea of the cause of the effect with the reflector, it may very well be like you say that the very small sources doesn't trigger the reflexes as it should. Sometimes body (and mind) works in strange ways. The pain caused by dazzle in one eye will for example get lower if you close the other eye. This is really weird.
When talking about glare: Older people suffer more from glare than young people as may take them 3 times as long to recover from it, so if you feel like shining someone in the eyes- go for the youngest.
One thing to consider is that blue light gives a more annoying glare than warmer tints, compared at same brightness level. Another reason to go for a warmer tint... I am a big fan of neutral-warm tints.

 

[Tobias Bossert]

I take your point, however, I'm talking about surrounding flood LEDs with small semi-matte reflectors, not optics or diffusers.
With a small reflector, someone looking at the light from within ~50 degrees of the lamp axis can still see the naked LED just as before, and just as bright as before, only now with a surrounding area of lit reflector. That does seem to be less dazzling, even though the peak brightness on the retina is the same as previously - possibly it's just easier to automatically avoid looking at.

I also made tests with semi-matte reflektors. This works relatively fine with SSC P4 but helps not really with Cree XR-E, since the reflector is hit by a too small fraction of the overall flux when you use an ful width angle of 90 deg. Than the reflector gathers the yellow-greenisch surrounding of the original beam of the XR-E only.

And there is another problem with "only reflector": You get a visible edge!

Therefore I remain so far by the solution of test device_2: XR-E plus Carclo wide angle reflector plus diffuser front lens. This works fine, it gives about the same spatial distribution as a bare XR-E but glaring is reduced dramatically. And the diffuser - I used a plane glass lens etched on the inner side - guarantees that there is no visible edge.

 

[uk_caver]

I also made tests with semi-matte reflektors. This works relatively fine with SSC P4 but helps not really with Cree XR-E, since the reflector is hit by a too small fraction of the overall flux when you use an ful width angle of 90 deg. Than the reflector gathers the yellow-greenisch surrounding of the original beam of the XR-E only.

That's true - I tend to use SSCs most of the time since they work better for me in my layout than the more centrally-biased Cree.
I think that adding a small collar to an SSC can give a nicer flood beam than a naked Cree, while also increasing the apparent size.

 

[gillestugan]

Ledil is coming up with a new lens that looks quite interesting. Twiddle. It has a tiltable diffused lens and will be available in a wide version, but there is no specs of the wide on their website. I emailed them and asked but they said it will take a few weeks until they have the specs, but tefficiency will be over 85%. It probably will be too narrow, but the combination of a tiltable head and diffused lens caught my attention.

 

[Barbarin]

I will complete the test report. Gottfried, the tester of test device_2 ('Bucherl'), took some photos. With the link you can compare Scurion P4, Petzl Ultra and my test device_2 (Tobi) side by side.

http://www.hirlatz.at/lampenvergleich.html

"Anmerkung 1:" = Note 1
"comparison of light distribution 1


left: "Scurion P4" flood on level 3,
right: "Tobi 1" work-light low


The light of Scurion decreases strongly with distance, with Tobi's worklight you can see more far. This is achieved by the light-up with smooth transition."


"Anmerkung 2" = Note 2


It is the same text but comparing Scurion P4 maximum flood (level 4) with Tobi's work-light high.


(Comment: Please realize that level 4 of Scurion is about 1A to the LED, but my work-light is 295mA + 91mA only.)


The room in the cave is covered with very dark loam, absorbing a lot of light. The caver stands immediately beside a wall (left to him). The floor decends slowly, increase again a little bit and - in a distance of about 15m - falls again and thus is hidden. The wall at the other side of the large room is far away. With the adapted eye you can see it with each light ghostly, but in the photos with fixed exposure it seems to be totally black - with the exception of my shaft-light.

My personal conclusion from these photos is, that it is NOT possible to use a two-LED construction with fixed beams! For normal use as a work-light, a relativly narrow beam makes no sense - even when combined with a diffuse light: You need a very soft transition from beam to flood and the beam should not reside in the centrum of the flood.

I'm verry interested in Javiers two-LED solution with zoom - but I'm jet not convinced that the transition of the superposition of both parts would be smooth enough to use it round the clock. I'm not shure whether the two-LED solution will be easyer to construct and will have lower weight as the three-LED solution with fixed diffuse/medium-beam/extra-narrow.

But never the less, I'm very eager to see some photos from Javier dealing with this. He is experienced in manufacturing and he also convinced me with his zoom-light: Much better than all I have seen in the past.

I agree with the schematic drawings of Javier completely: That's what we need.

At the moment I'm still looking for spherically domed front glass with diameter 18 to 20mm and dome height of 4 to 5 mm (spherical cap with 90 deg opening) to make some experiments with domed diffusers. This diffused light would be need to avoid glaring anyhow, whether we end up with a two-LED or with a three-LED solution.

Tobias

Great job!!! Makes my comparison to look like a child work....

This weekend there will be more real cave testing, so I hope to publish pictures by Monday. This time we will test the TWO LED configuration, with different levels of zooming.

Regarding your concerns about blinding your own friends, I think that won't happen with low level pure flood, even with a non diffused CREE... and when it comes to "work light" it is near impossible to avoid if you look at their eyes. I just believe that warmer colors, with less "bluish-greenish" tints should help to reduce the glaring effect, as those colors are the real scotopic vision killers. So according to this, what do you preffer?

1. Use cool white (high lm/watt) and diffusers (efficiency lose) to reduce glare.

2. Use warm white (lower lm/watt) no diffuser. You get better higher CRI, but you need more energy to get the same light.

3. Cool white, no difusser. On a cave efficiency is the priority, above CRI. Less weight, more light. Glare is unavoidable anyway, so will try to not look directly to my friends when work light is on.

What do you think, guys?

Javier

 

[Tobias Bossert]

I just found a sherical diffuser (Khatod PLJT 35 and PLJT 35/02) which I will try to modify:

http://www.optomarket.com/jsp/index.jsp?p_gadgetURL=pagemapping.jsp%3Fid%3Dcatalog_illus&p_contentTopic=&pocs_bnav=yes&p_containerTopic=0000121203&p_Navigator=0000043212|0000043244|0000121203

It is just a hemisphere made from poly carbonate, outer diameter is 30.4mm and inner diameter is 24.4mm, so the material is 3mm thick.
There are two types. One is extremely milky (PLJT 35) and the oher one (PLJT 35/02) they call "transparent". I will test both of them.
The modification will be lathing (or sanding) it down to a hight of about 6..7mm remaining a spherical cap with an opening angle of about 90 deg, an inner diameter of 17..19mm and an outer diameter of 23..26mm.

I guess the milky type diffuses much to strong.
In case the "transparent" one diffuses to weak, I will sand blast it at the inner surface.

Unfortunately the register page of Optomarked.com doesn't work propperly at the moment, so I emailed them.

I will post whether this ends up in an improvement against the plane edtched glass lense.

I'm still convinced, that changing another tint will not solve the glaring problem sufficiently. The 'diffus-light' of test device_2 does the job very well, but to the expense of about 20...30% loss of overall luminous flux depending upon the "etchedness" of the lens.

Tobias

 

[Barbarin]

I just found a sherical diffuser (Khatod PLJT 35 and PLJT 35/02) which I will try to modify:

http://www.optomarket.com/jsp/index.jsp?p_gadgetURL=pagemapping.jsp%3Fid%3Dcatalog_illus&p_contentTopic=&pocs_bnav=yes&p_containerTopic=0000121203&p_Navigator=0000043212|0000043244|0000121203

It is just a hemisphere made from poly carbonate, outer diameter is 30.4mm and inner diameter is 24.4mm, so the material is 3mm thick.
There are two types. One is extremely milky (PLJT 35) and the oher one (PLJT 35/02) they call "transparent". I will test both of them.
The modification will be lathing (or sanding) it down to a hight of about 6..7mm remaining a spherical cap with an opening angle of about 90 deg, an inner diameter of 17..19mm and an outer diameter of 23..26mm.

I guess the milky type diffuses much to strong.
In case the "transparent" one diffuses to weak, I will sand blast it at the inner surface.

Unfortunately the register page of Optomarked.com doesn't work propperly at the moment, so I emailed them.

I will post whether this ends up in an improvement against the plane edtched glass lense.

I'm still convinced, that changing another tint will not solve the glaring problem sufficiently. The 'diffus-light' of test device_2 does the job very well, but to the expense of about 20...30% loss of overall luminous flux depending upon the "etchedness" of the lens.

Tobias

Hi Tobias,

A small exposure to cyanocrilate glue vapours is enough to "frost" slightly almost any transparent plastic. Faster than mill, and easier to test different exposure times.

Javier

 

[Tobias Bossert]

Hi Tobias,

A small exposure to cyanocrilate glue vapours is enough to "frost" slightly almost any transparent plastic. Faster than mill, and easier to test different exposure times.

Javier

Hi Javier,
thanks, that's a good idea!
I always tried to wrap with a solvent moisten cloth, but the result was very irregular, even when the cloth was nearely dry. I will play around with vapour, this is very easy to apply for freezing the inner side of a dome: just put a small drop of glue on a plane surface and place the dome over it.
Tobias

 

[uk_caver]

Frosting from cyanoacrylate does seem to be particularly likely if the dome has a tiny amount of grease from normal fingerprints on the inside.

It also seems possible to wipe the frosting off from at least some surfaces, just using a paper tissue.

 

[gillestugan]

Yes, I also have experience of "frost" that is quite easy to scrape it off, but as long as it is sitting on the inside it will be protected.
Just a thought, but is it better to have it on the outside? Most frosted lenses are frosted on the outside and not the inside. Is there a reason?

 

[likeguymontag]

Yes, I also have experience of "frost" that is quite easy to scrape it off, but as long as it is sitting on the inside it will be protected.
Just a thought, but is it better to have it on the outside? Most frosted lenses are frosted on the outside and not the inside. Is there a reason?

I would recommend having the frosting on the inside. Caves are dirty, and a frosted lens would be very hard to clean. Perhaps there are additional considerations. Also, look into diffusing filters for photography or theater lighting. I would expect that you could get performance data for such products.

 

[Barbarin]

Off-Topic.

This is and extraodinary thread. I'm really happy with it because there is an active discussion about many facts related to portable lighting, reconsidering it from a starting point created one century ago.(carbide lamps)

There has been, there are and there will be billion threads about "tactical" and "notsotacticalbutgadgety" lights, basically handheld lights. But this one is being discussed from a perspective focused on real usage and real users which are not flashaholics, considering visual field, CRI, output, angle... Surprisingly no one has asked or insisted on the importance of a strobe setting (or five of them) and cavers are exposed to real dangerous situations. (IMHO it can be usefull, but not on every flashlight).

Well, as I said an extraordinary thread on every sense of the word. Thanks for starting and following it.

Javier

 

[uk_caver]

Yes, I also have experience of "frost" that is quite easy to scrape it off, but as long as it is sitting on the inside it will be protected.

The upside is that for experimentation, excess frosting may be reversible.

Just a thought, but is it better to have it on the outside? Most frosted lenses are frosted on the outside and not the inside. Is there a reason?

I don't know enough optics to be be able to guess, but could there possibly be a small difference in light transmission for a given amount of frosting?
That is, if the outside surface is frosted, might some fraction of the light reflected back from the frosting into the lens be internally reflected off the inside of the lens and have a second chance at making it out through the front?

The intuitive feel is that optical setups should tend to be 'reversible', but that's maybe more appropriate to point-to-point systems (tracing light paths through lenses) than the kind of more random ('dirtier') point-to-spread setup involved in diffusers.