Multiple output - the wrong way?

Kiessling

Flashaholic
Joined
Nov 26, 2002
Messages
16,140
Location
Old World
We all want multiple output LED lights. Obviously.
Here's why: to be able to adjust the light output to our needs, meaning little output for close-up and a big nasty punch to spot the dog far away. To avoid being blinded by an insane hot-spot while working on our desk and to be able to light up the whole room with the same light nevertheless.
In the end, it comes down to the simple wish to maintain a more or less constant light input to out eyes, enabling us to see things at a sufficient illumination either near or far away.

Now ... my point:
Basically the goal is to achieve a constant light-input to our eyes that needs less output for close-up tasks and more punch for things far away. This amount of light our eyes need to see clearly is the same for near and distant objects, you just need more light energy to achieve this goal according to the laws of physics when the target is further away.

So ... with multiple output levels we can achieve this ... by changing the output according to the needed input. This is an insufficient solution IMHO. This would be a light that requires attention and tweaking while at work, it actually distracts from the task the user might be working on and, in case of improper use, may even have adverse results, such as accidently loosing your adapted vision when shining the high beam at something close.

Here's the solution:
Build a light that defines its output according to what our eyes see, according to the input level. Sounds impossible? Why?
With all those fancy computer-controlled lights around (Eternalight, LS4, ...), the solution is simple. Just install a photo-diode or something that can measure light-intensity at the front end of the light, shield it from the beam, and have the electronics adjust the output according to the measured input level automatically. This would be same input that your eyes see, the light reflected from the target of the beam. If you swing your light around, it will automatically dim when hitting a closer target and brighten when shining it on something far away. For you (your eyes) it would result in an illumination level that is always at the same intensity, thus preserving your vision and making it unnecessary to fumble with your light to adjust the output. It might save battery power, too.

You could even take this concept one step further and allow for selectable INPUT levels that represent really the light intensity that you need, and not an output level that might be inadequate a second later when moving your light around.

Basically, this concept takes a more physiological approach to what we need, the definition should be INPUT to our eyes and not OUTPUT, this is just a poor construct to achieve the true goal, which is not what a device can do, but what it can do for us to accomplish a task. That would include the sensory input as the important variable to discuss, and not output, which is the dependent variable and should be handled by the device itself without the necessity of constant monitoring by the user.

What do you think?
Some crazy modder listening? PayPal loaded ... /ubbthreads/images/graemlins/grin.gif /ubbthreads/images/graemlins/grin.gif

bernhard
 

Kiessling

Flashaholic
Joined
Nov 26, 2002
Messages
16,140
Location
Old World
Doug ... souns cool! maybe there's hope ....

DrJ ... this would be much more than needed. this whole concept is actually much more simple than a camera ...

bernhard
 

AilSnail

Flashlight Enthusiast
Joined
Jul 11, 2003
Messages
1,299
Location
Bergen, No
I recommend some barn doors on the sensor then, since what you want to read the hotspot and not the secondary beam.

Nice, kiessling!!

I think it would be worth a try.
 

Doug Owen

Flashlight Enthusiast
Joined
Jan 30, 2003
Messages
1,992
Dynamic range and beam angles will be issues, as will be defining what makes the right level of light return.

Consider, for instance, ground return from a few feet as you are trying to see the tree line at 200 yards. It's not really reasonable to expect a simple circuit to resolve the return from the treeline (with the near in clutter disregarded), even if it did understand that you wanted that. OTOH, I know a bunch in advanced vision recognition....

And then you still have the dynamic range (we'll need *decades* of output range) and beam angle (from a few degrees for the treeline, to say 60 for finding your keys.

This one will take a little while, I'm thinkin'

Doug Owen
 

FlashlightOCD

Enlightened
Joined
Apr 7, 2003
Messages
931
Location
Central FL, USA
[ QUOTE ]
DrJ said:
...hysteresis...

[/ QUOTE ]

I'm not familiar with that term but suspect it may be a more scientific term for feedback. I think the suggestion is a very interesting one, but I would like to know how you would prevent the whole system from strobing?
 

dukeleto

Newly Enlightened
Joined
Oct 21, 2002
Messages
187
Location
France
As a matter of fact, I have an old (4 years) front bike light made by Vetta that tried to do something slightly similar, although less ambitious:
it has a photoresistor on the top of the light that reacts to the amount of ambient lighting to change the (originally) incandescent output. It was "dumb" regulation though: just a variable resistance, which is terrible from an electrical conversion point of view for the incandescent bulb. It works better with its current luxeon emitter, but still isn't great, because I never really managed to get a satisfactory output variation according to ambient conditions.

Olivier

PS: the spellchecker doens't accept "luxeon" /ubbthreads/images/graemlins/icon23.gif
 

Kiessling

Flashaholic
Joined
Nov 26, 2002
Messages
16,140
Location
Old World
so ... little technical problems aside /ubbthreads/images/graemlins/grin.gif ... maybe it wuold be doable.
well, since I am not skilled in this field of expertise, I'll just wait and see /ubbthreads/images/graemlins/smile.gif

DrJ, the sensor could well be in the body of the light pointing in the direction of the main beam since this is the area that you want to light up. if you do not look at your target, that is another story /ubbthreads/images/graemlins/smile.gif

and why would it strobe?

bernhard
 

WildRice

Flashlight Enthusiast
Joined
Nov 30, 2003
Messages
1,135
Location
SW Michigan
Why would it strobe... If the reaction time if the sensing circuitry is too fast. light gets sensed, regulator backs off, if it overshoots the light would go high again. If an RCT of about .5 to 1sec is used, the light would ramp up/down when an object is in the proximity of the sensor.
I will try to scribble something down and post it here when I can.
Jeff
 

Doug S

Flashlight Enthusiast
Joined
Jun 20, 2002
Messages
2,712
Location
Chickamauga Georgia
[ QUOTE ]
FlashlightOCD said:
[ QUOTE ]
DrJ said:
...hysteresis...

[/ QUOTE ]

I'm not familiar with that term but suspect it may be a more scientific term for feedback. I think the suggestion is a very interesting one, but I would like to know how you would prevent the whole system from strobing?



[/ QUOTE ]
Hysteresis? Thats one of those female problems /ubbthreads/images/graemlins/icon15.gif
 

WildRice

Flashlight Enthusiast
Joined
Nov 30, 2003
Messages
1,135
Location
SW Michigan
hysteresis....ok... your furnace is set to turn on at 70F. when the temp in your house is 69F the heat kicks on. when it reaches 71F it turns off. Hysteresis is +/- 1F.
Most inverters (ICs) are like that too. example... will turn on at 2.75v but will stay on until the voltage is below 1.8v (voltages are made up). this keeps it from oscilating uncontrolled.

okey-dokey

Jeff
 

JSWrightOC

Enlightened
Joined
Mar 1, 2003
Messages
559
Location
Fort Mill, SC
I think this is at least worth some experimentation and development. If you have a fixed-focus light then you can put optics in front of your sensor and calibrate it to match the beam divergence of the light output. If the light were adjustable throw/flood, then you would have to adjust the sensor optics with the light output. IMHO I think it's a great idea to try and impliment. If you use a high-speed sensor (photodiode or photovoltiac cell, not a CdS sensor) and an LED source, then loop slew rate can be increased while still avoiding oscillations. A "sensitivity" adjustment is a must (to adjust relative brightness), as well as a manual override button for full intensity in case you need it. The only problem I forsee is shining the light at retro-reflective surfaces (signs, license plates, etc.) because a lot of light would be returned to the sensor that may not be returned to your eyes. I think someone should prototype this and really see how well it works.
 

Steelwolf

Flashlight Enthusiast
Joined
Feb 6, 2001
Messages
1,208
Location
Perth, Western Australia
When I was at a welding workshop, I recall the instructor had this cool welding face shield. The lens would remain quite clear (very slight tint) until he struck an arc, then it would immediately darken to protect the eyes. It worked instantaneously so that there wasn't even an initial flash. Apparently, it used an array of photovoltaic cells and a manual trim to adjust the level of darkening so that it could be used for gas welding as well as electrical arc.

I'm not sure how well it would translate to flashlight use as the face shield control is open-loop, ie, the amount of light from the arc will vary on its own, the shield will just darken or lighten according to how much light it senses. In the case of a flashlight, it is close-loop and therefore more prone to instability. The amount of light will be determined by the output from the sensor, which is determine by the amount of light.

I can see how to design the control loop, but my knowledge of electronics is insufficient to design the actual circuit. This is certainly a concept worth looking at, but not an overly important one. I suspect that at the size of flashlights most people are toting around, and the power output of those flashlights, a manual control would make far more sense as they will be running full out most of the time, being turned down only for the occasional closeup work.
 
Top