DIY Light Meter Sensor Quesiton

[Grif]

Any information on sensors for light meters, and color response, sensitivity and information like that? Links, Comments, most anything to get me started int he right direction.

Interested in a low light capable DIY meter project for the winter.

 

[jamesgignac]

Uh, wow - I just want to follow this thread to see what comes up. Sorry Grif I have no info of my own - in fact I couldn't imagine putting together my own meter but I suppose there must be some benefit to doing so. I've always been happy with my Luna Pro but I'm tickled pink to know there's a diy-er out there who's taking on something I never thought of.

Best of luck!

 

[anikin]

I've been thinking about such a project myself. Check out ISL 29102. digi-key has is for under $3. Add a button-cell battery, a variable resistor and an LED and you're done.

 

[BetterSense]

I've made 3 exposure meters plus several densitometers/enlarging meters. Easiest thing I've found for covering the whole brightness range from moonlight to sunlight is a device like the TAOS TSL230R. All my exposure meters are based on this chip. Since it's silicon photodiode based, it requires IR and red filtration for good color response...although with no IR filtration, vegetation is several stops "brighter" than it should be, so maybe you could try to make an IR light meter with it.

You might also check out visual-corrected photodiodes, with integrated fractional-exponent amplifiers, like the LX1973a, LX1970, or LX1971. Note that most B&W film is sensitive to UV and deep red well outside of the luminosity function, however, so you have to decide for yourself if a visually-corrected sensor is really ideal for an exposure meter. Also, most of these will conk out/saturate in full sunlight, so covering the whole brightness range with them might require some engineering.

My best meter I made is a microcontrolled one with a 32-character LCD, AAA powered, 0.05 degree spot/incident meter with Aperture priority, Shutter priority, Ev mode, ability to either take and store readings or 'scan' by holding the button down, brightness range from <<0 to 15, film speeds from 25-3200, and only 1x1x3 inches. So it's pretty much the best lightmeter in the world by my estimation.

 

[Steve Smith]

So it's pretty much the best lightmeter in the world by my estimation.

Then perhaps you should provide a link to the plans and circuit diagram!


Steve.

 

[Grif]

Interesting device, down to 0.3 Lux is good,,, seems like the chart I was looking at had twilight down to 0.1

Anyone know what the (I know,,, how hi is up) light level is at the ground glass of a view camera if I was looking to put a "spot" probe together?

I guess I just need to hack something together on a breadboard and just try it out.

Looked at the ISL 29102 data sheet. Looks like the voltage out is usable as a light level directly, didn't really think about the tempature curve (values) but it did seem to be an issue at ambient temps.

FYI, I'm way more involved in the gizmo effect of the various hobbies I have than the actual hobby. Never been sure if that's good thing, or a bad thing,,, but as things go, and I do like things,,, it's way easier on my liver and marriage than the cute little hottie at the local tav ;-)

 

[domaz]

This blog has a good write up of using the TAOS TSL203R with an Arduino microcontroller to build a light meter.

 

[Grif]

Great info BetterSense,,, I'll go take a look

 

[michaelbsc]

Oh great. Yet another project to fill up my time. Jeez.

 

[Grif]

Naw,,, we can hack up the hardware,,, the big value is the code for the PIC. I'm mostly still a solder and wire sort of programmer during my off hours, even though my paycheck depends on databases and opsys's.

Then perhaps you should provide a link to the plans and circuit diagram!


Steve.

 

[anikin]

For a spot probe you may probably want something more sensitive, I'm not sure. I guess, you could add a small lens in front of it which could improve low-light characteristics. The temperature is in C, so you'll be staying in -10+40 range most of the time which gives about 0.002 LUX error. Does not seem to be too bad. I was thinking about building light meter into a watch, so I could just calibrate it at 98F and temperature would become totally irrelevant ;-).

If it's not too much trouble, please keep us updated about your investigation, because I might be following your footsteps soon.

 

[BetterSense]

Then perhaps you should provide a link to the plans and circuit diagram!

There is no real circuit diagram; all the information is in the TSL230 datasheet. Here's a picture of two of my older meters though.

 

[benjiboy]

I'm in awe at the depth of knowledge and expertise we have from our members, we really have some very talented people in our midst.

 

[anikin]

BetterSense,

It's a cool device, I can see how it could be easily hooked up to a microcontroller - just count number of pulses in a time interval, and you're done. Cool. Also the light spectrum is very wide, so if you add an IR filter, it would make a dream meter for IR shooting.
In your experience, how sensitive is it in low light? Is it usable for star-trails photography? How about measuring a candle-lit room?

Eugene

 

[BetterSense]

You can use it basically arbitrarily dark; it just takes that much longer to take a reading. You can design your software to measure pulse length too. Eventually dark-current will get you, but you can calibrate that out for the most part. My incident meter transitions from frequency-averaging to pulse-width-measuring regimes somewhere around EV0. I didn't bother adding this functionality to my spot meters because you would have to hold still for several seconds on target; at light levels that low, I don't find metering very useful for photographic purposes, with reciprocity failure and all.

 

[Grif]

MicroProcessors

Even found a couple of links in the various comments above for a Basic stamp app, and an AVR example.

 

[Hexavalent]

PICs, Basic Stamps!
C'mon we're supposed to be Analog here - how about a couple 12au7's and a neon voltage reference

 

[holmburgers]

Please, somebody make an IR light meter.

 

[Grif]

Hollowstate lives!!! Good op amp designs were a tought go before IC's, 741 op amp's still considered Analog, right?

So,,, Hexavalent? Chrome?

PICs, Basic Stamps!
C'mon we're supposed to be Analog here - how about a couple 12au7's and a neon voltage reference

 

[Grif]

Best I can tell, the meter sensor that BetterSense used is sensitive to IR,,, just pop an IR filter in front of it and calibrate away.

Please, somebody make an IR light meter.

 

[BetterSense]

Best I can tell, the meter sensor that BetterSense used is sensitive to IR

Indeed; I have never tried to use it as an IR meter, but when I was testing it without an IR filter, green foliage in sunlight read about 2 stops brighter than pavement, which is hardly what you expect based on visible light but is consistent with IR.

 

[Hexavalent]

I suppose 741's are analogue enough, for the purist, a Philbrick op-amp would be better.

In reality, a couple of discretes are all that are needed for a decent light-meter.

and Yes, Chromium, one of my favourite metals.

 

[Steve Smith]

C'mon we're supposed to be Analog here - how about a couple 12au7's and a neon voltage reference

That's my sort of electronics! See Valve Audio link below!


Steve.

 

[David Grenet]

There is no real circuit diagram; all the information is in the TSL230 datasheet. Here's a picture of two of my older meters though.

I'm interested in how you narrowed the field of view for the spot meter (lens?, pinhole?). Does it have a viewfinder so you know what you're pointing it at?

 

[BetterSense]

I just use a lens, exactly like a camera. The longer the focal length, the smaller the angle intercepted by the sensor. Brightness is controlled by F/ratio exactly like a camera, so a longer lens tends to have a worse aperture, though. Right now I'm using a coated, F/.9, 18mm lens, which is very compact, but only gets me a 4 degree spot in low light. Since my sensor can adjust its active area according to brightness, in medium (room) light, my spot size is .4 degrees and in sunlight it's theoretically .04 degrees although with flare I'm not sure that's a meaningful number.

If I wasn't a compactness maniac I have a f/2, 50mm lens that I would use. I suppose this is similar to the lenses used on Pentax spotmeters and so on.

Aiming is accomplished by sighting through a peephole that runs through the length of the meter, which consists of 4-5mm diameter aluminum tubing. It's a bit squinty but works well. I've considered adding some rifle sights, but they would stick out.