The issue about bulb/torch lumens

[Swedpat]

I mentioned it in another thread, don't remember which:

A binocular of high quality has a total light transmission of 90% or more. It has at least 2 lenses in the objective and 3 or more lenses in the eyepiece. Plus two prisms. That is several lenses and surfaces to pass through.

A flashlight has one "lens" and one mirror(reflector). A single lens/glass plate has a light transmission of around 95% even without an antireflective coating and a good mirror close to 99% or so. How can a flashlight have a light loss of up to 30% or more? That cannot be correct. Even a cheap low quality flashlight should have a light transmission comparable to the best binocular. Actually should "torch lumens" in any case be 90% or more of "bulb lumens"!
The common number of around 65% has to be a fake made by the manufacturers, maybe is the much higher "bulb lumens"-number something psychological? You don't need to be an optical engineer to understand that something isn't correct in this case! Just simple optical knowledge.

Regards, Patric

 

[Marduke]

Re: The bluff with the Bulb/torch lumens

65% is for incandescent. A decent LED setup is 75-85% (reflector-TIR).

These numbers did not come from manufacturers, they came from CPF testing.

 

[Kiessling]

Re: The bluff with the Bulb/torch lumens

A single lens/glass plate has a light transmission of around 95% even without an antireflective coating and a good mirror close to 99% or so.

This is incorrect. AR coeated sapphire has losses of roughly 12%, and it needs a special UCL to achieve 99% transmission.

Reflectors are even worse.

The 65% was determines by CPF testing for incans. LEDs seem to be higher, but it is in th etesting right now. You can have a look at MrGman's sticky where he does lumen testing.
Although I still think his numbers look too high, the idea is that LEDs with good optical systems will be better than 65%.

bernie

 

[Cydonia]

Re: The bluff with the Bulb/torch lumens

Wait a minute - this is interesting. What if an LED has no reflector or optics at all and is mounted right on a heatsink? What number do we get then?
A soft 180° diffuse glow in all directions - but what efficiency?

 

[Swedpat]

Re: The bluff with the Bulb/torch lumens

What I base this issue on is the fact that a mirror has less light loss than a lens when it comes to optical instruments. I know this from astronomical telescopes. However, lens or mirror, we have two elements included in a flashlight while a binocular has several more to pass through. And still much higher light transmission in the binocular. Something is still wrong...Where does 35% light disappear in the flashlight while only 10-20% get lost in the binocular?

 

[js]

Re: The bluff with the Bulb/torch lumens

I mentioned it in another thread, don't remember which:

A binocular of high quality has a total light transmission of 90% or more. It has at least 2 lenses in the objective and 3 or more lenses in the eyepiece. Plus two prisms. That is several lenses and surfaces to pass through.

A flashlight has one "lens" and one mirror(reflector). A single lens/glass plate has a light transmission of around 95% even without an antireflective coating and a good mirror close to 99% or so. How can a flashlight have a light loss of up to 30% or more? That cannot be correct. Even a cheap low quality flashlight should have a light transmission comparable to the best binocular. Actually should "torch lumens" in any case be 90% or more of "bulb lumens"!
The common number of around 65% has to be a fake made by the manufacturers, maybe is the much higher "bulb lumens"-number something psychological? You don't need to be an optical engineer to understand that something isn't correct in this case! Just simple optical knowledge.

Regards, Patric

Patric,

You're not the first to have this feeling / assessment about bLu vs. tLu, but it is true. Only about 65 percent make it out the front for incan filaments: bLu vs. tLu: IS confirms 65% conversion factor.

I think the flaw in your line of thought is that a lot of the light from a filament goes directly backward into whatever is holding the bulb, and most of that light is lost, plus even though the reflector has a reflective surface that has a high transmission rate, there are multiple (as in double) reflections going on in many cases, if you do the ray tracing.

But whatever the theoretical take, the integrating sphere results tell us that multiplying the incandescent bulb lumens by 0.65 is a pretty good estimate of torch lumens.

This is not correct for LED, where the number is larger (more make it out the front). Undoubtedly, this is due to most of the light being directed that way anyway, vs. the incan scenario and the filaments radiation pattern. But, anyway, believe it or not, it's the truth. It has been confirmed multiple times now by multiple people here on CPF (including me). Check out the link.

 

[Marduke]

Re: The bluff with the Bulb/torch lumens

Binoculars usually have high grade optical glass. Flashlights don't, simply because it would not hold up to the abuse and demands of the light. Regular old glass with some AR coating is the best you can expect usually. Reflectors are also not optical grade mirrors. They have the reflective properties closer to a plastic shaving mirror in a shower.


Sure, you could make a high quality system with fewer losses, but the price goes up exponentially for a relatively small efficiency boost, which is especially unimportant considering the logarithmic human perception of brightness.

 

[Gunner12]

Re: The bluff with the Bulb/torch lumens

I think it has to do with the reflector coating used and maybe also the way light is collected.

The binocular has light coming in at near parallel while the flashlight has light hitting the window at many different angles. From what I figure(no sure if it is right or not), if the light hits at less or more then perpendicular, more light is lost.

 

[Swedpat]

Re: The bluff with the Bulb/torch lumens

Wait a minute - this is interesting. What if an LED has no reflector or optics at all and is mounted right on a heatsink? What number do we get then?
A soft 180° diffuse glow in all directions - but what efficiency?

Yes, that's what I am talking about. Without the reflector it will provide ca 50% more light (100/65) than placed in the reflector according to the usual numbers. Where does the 35% (100-65) dissapear? Through a high quality binocular only 10% or less will get lost.

Regards, Patric

 

[js]

Re: The bluff with the Bulb/torch lumens

Yes, that's what I am talking about. Without the reflector it will provide ca 50% more light (100/65) than placed in the reflector according to the usual numbers. Where does the 35% (100-65) dissapear? Through a high quality binocular only 10% or less will get lost.

Regards, Patric

READ

MY

POST.

 

[Swedpat]

Re: The bluff with the Bulb/torch lumens

Thanks for all replies,

Marduke:

Binoculars usually have high grade optical glass. Flashlights don't, simply because it would not hold up to the abuse and demands of the light. Regular old glass with some AR coating is the best you can expect usually. Reflectors are also not optical grade mirrors. They have the reflective properties closer to a plastic shaving mirror in a shower.

Sure, you could make a high quality system with fewer losses, but the price goes up exponentially for a relatively small efficiency boost, which is especially unimportant considering the logarithmic human perception of brightness.

In some high class flashlights the glass is coated, especially noticable with the green reflection of my EagleTac P10A. The question is if that's mainly a selling argument more than a true improvement of the lights noticable brightness? I mean, with the hunting of lumens a higher class of glass and reflectors would be worth the higher price according to many flashoholics...
Example: Fenix TK10 R2 with 240 lumens when the non R2 has 225 lumens. That difference likely isn't noticable for the eye, and likely would possible to reach by a higher class optics instead of a brighter bulb. Or maybe the improvement is by higher grade optics...

Regards, Patric

 

[flash_bang]

Re: The bluff with the Bulb/torch lumens

even if you had 100% light transmission, it wouldn't matter due to the way the human eye perceives light, that's why unless you compare lights side by side you can't tell if you're looking at a 60 or a 100 lumen light.

 

[Swedpat]

Re: The bluff with the Bulb/torch lumens

READ

MY

POST.

Yes, I have read now. But it's still hard to understand that high light loss...

 

[Kiessling]

Re: The bluff with the Bulb/torch lumens

Example: Fenix TK10 R2 with 240 lumens when the non R2 has 225 lumens. That difference likely isn't noticable for the eye, and likely would possible to reach by a higher class optics instead of a brighter bulb.

Either way, it just doesn't matter for the light and its usefulness. The people do still fall for that trap, but that's life.

Maybe high grade optical stuff has other disadvantages besides cost?

bernie

 

[Kiessling]

Re: The bluff with the Bulb/torch lumens

Yes, I have read now. But it's still hard to understand that high light loss...

1. LEDs loose less than incan. We don't know how much yet, but it seems much less. Read MrGman's thread.

2. incans spread the light in all directions whereas the binoculars have more or less parallel light coming in and needing to be redirected

3. incans thus loose light to the area under the bulb without reflector, to the enveloppe of the bulb itself, to the stipples of an orange peel reflector, to numerous reflections of the light due to the special situation in the reflector that is not present in the binoculars and to areas not covered in reflective substances like the bezel rim etc.

4. It is what it is. It has been shown. Reality is just that ... real.

bernie

 

[Swedpat]

Re: The bluff with the Bulb/torch lumens

even if you had 100% light transmission, it wouldn't matter due to the way the human eye perceives light, that's why unless you compare lights side by side you can't tell if you're looking at a 60 or a 100 lumen light.

Yes, I have noticed that it can be difficult to estimate brightness. As you say, without a side by side comparison I would not always be sure if a flashlight has 60 or 100 lumens. Side by side, I can see the difference though it feels quite moderate.

Regards, Patric

 

[flash_bang]

Re: The bluff with the Bulb/torch lumens

I understand that the human perception of brightness is logarithmic, but I'm not really sure how to apply that. I know that 10 is twice as much as 1, and that 1000 is twice as much as 100, but do I take that literally? as in a beast is only a little more than twice as bright as a KL5A?

Cheers,
Flash

 

[Swedpat]

Re: The bluff with the Bulb/torch lumens

I understand that the human perception of brightness is logarithmic, but I'm not really sure how to apply that. I know that 10 is twice as much as 1, and that 1000 is twice as much as 100, but do I take that literally? as in a beast is only a little more than twice as bright as a KL5A?

Cheers,
Flash

I also have heard that. But I am not sure it's exactly that way. Yes, I think we can say that it will demand about 4 times brightness to provide a WOW! difference. But I want to say that my Fenix 4E20 feels more than twice as bright as a single E20. Maybe not 4 times but close to 3 times brighter...

Regards, Patric

 

[jtr1962]

Re: The bluff with the Bulb/torch lumens

I understand that the human perception of brightness is logarithmic, but I'm not really sure how to apply that. I know that 10 is twice as much as 1, and that 1000 is twice as much as 100, but do I take that literally? as in a beast is only a little more than twice as bright as a KL5A?

What you're referring to is a base 10 logarithm. IIRC human perception of brightness is closer to a base 2 logarithm (four times the brightness is perceived as twice as bright, eight times as three times as bright, etc.)

 

[Swedpat]

Re: The bluff with the Bulb/torch lumens

What you're referring to is a base 10 logarithm. IIRC human perception of brightness is closer to a base 2 logarithm (four times the brightness is perceived as twice as bright, eight times as three times as bright, etc.)

Hi jtr1962,

I have discovered that the eye can be deceptive in some circumstances. My 4E20 is around 90% brighter than the TK10.
When I compared my 4E20 indoors side by side to the TK10 I could see that it was really brighter. But when I brought it out in the forest i didn't experienced that big difference...