New DIY LED light

[wietefras]

There are many reasons that could happen. The inverse square law not being one of them.

Reflection would be there at each height so unlikely reflection would be a major factor.
And this overlap you speak of. Got any citations to back that up? Overlap will follow the inverse square law.

It's really pointless for us to try and explain this in physics principles since you are still unable to apply it and flat out ignore the truth when presented to you on a silver platter.

Even your own measurements and calculations have demonstrated that inverse square law does not apply to distance between fixture and canopy. Malocan did the same with a spectrometer and many others have done it with lux meters. It's really time you stop the cognitive dissonance and let the truth in.

If you really want to see this for yourself, why not build an actual growlight? You pretend to be a "growlight researcher" so how difficult can that be? Hang the damn thing in your tent and find the optimal height for uniformity (whole surface covered up with darkest spots still getting at least 80% of average). Double the distance and see if intensity was cut to 25% of the first measurement.

If you want to do it properly you should measure average light intensity over the whole surface. Single point measurements are unreliable. Measuring the average will also correct for when you got the uniformity wrong (distance too close) and for small measuring errors (wrong angle, wrong position). That is if you have learned how to correctly calculate averages since the first time you appeared here as nofucksgiven. Otherwise just stick with one measurement for each height.

BTW Do you also remember how you were applying "inverse square corrections" to your PPFD measurements back then? Good times, good times.

Lux meters typically measure the number of photons between 500nm and 600nm

Wrong. The lumen CIE curve extends all the way from 380nm to 780nm. The amplitude is less at the extreme ends, but it's still measured and therefore it works perfectly fine to measure light of any (visible) color.

The sensors used in Lux meters are similar (if not the same) as those used in cheap PAR meters. They are also sensitive over the whole range.

For instance this one:

For a lux meter they add a color filter to tone down the blue and red ends. For a cheap PAR meter they add a color filter to squeeze the chart into more of a box shape.

Either way, if the lux meter provides a reading, it's reading the light and it can be used perfectly for relative light intensity comparisons and also for guesstimating PPFD values well within the 600% range that plants can deal with.

Oh and while we are at it. Malocan's spectrometer produced both PPFD and Lux values. If you divide those lux values by the corresponding PPFD they give you exactly the same Lux to PAR conversion factors like we calculate with Alesh's chart or like Apogee posts on their site. Amazing coincidence since that's impossible right?

You would know that if you had the ability to understand optical measurement.

Funny how the measurements showed exactly what I (and others) told you they would show and exactly the opposite of what you claimed. So there.

You know a lot about this stuff.

Finally you say something which is correct. Wow.

 

[GrowLightResearch]

You cannot fool all the people all the time.

Why are you bringing up this curve. I covered this topic with you here:
You are now getting malicious with your attacks. Taking my post and putting your spin it with obvious ill intent.

Ambient Light Sensors

I went easy on you there were other points about your misguided advice that I skipped So I will take yet even more time. Maybe this time something will stick in your troubled mind. Are you blind to how wrong you have been on nearly everything you have said in the past couple of weeks???

The sensors used in Lux meters are similar (if not the same) as those used in cheap PAR meters. They are also sensitive over the whole range.

More bullshit. The Lux meters do NOT use the full Photopic bandwidth.

The ALS curve you unscrupulously shows the bandwidth before filtering. After it is filtered, the range is about 500-600nm.

Wrong. The lumen CIE curve extends all the way from 380nm to 780nm. The amplitude is less at the extreme ends, but it's still measured and therefore it works perfectly fine to measure light of any (visible) color.

First off a lux sensor uses one single photodiode. A photodiode is very similar to a solar panel's photo-voltic cell. When the light excites the photodiode anywhere photodiodes memory effect "stores" the energy created when a photon strikes the photodiode. All that is measured is the number photons. There is no way for it to distinguish between wavelengths and therefore cannot reveal any spectral information.

A PAR meter's sensor use multiple photodiodes with various wavelengths. So NO, lux and PAR meter's do NOT use the same sensor.

The CIE Photopic Luminous Efficacy goes from 390 to 770
At 390nm the sensitivity is 12
At 550nm the sensitivity is 99,495
At 770nm the sensitivity is 3

An ALS is very sensitive to UV and IR so the blues and reds are filtered out. Bu that may not be the reason.

Wrong. The lumen CIE curve extends all the way from 380nm to 780nm. The amplitude is less at the extreme ends, but it's still measured and therefore it works perfectly fine to measure light of any (visible) color.

The center of the curve is 555nm which CIE set its value at 100,000.
One sec in a sensors datasheet is FWHM which is the point in the curve that sensors read and report, Full Width Half Max is the 50% point in the curve. Guess whch wavelengths are at 50% of the CIE Photopic Luminous Efficacy curve. The two points at 50% are 510nm at 50,300 and 610nm at 50,300. So No! A lux meter would never look beyond 510nm and 610nm

The Texas Instruments OTP3001 and OTP3002.both use the same photodiodes as you said. But purposely leaving out the rest of the story is the same a lying.
.
The 3101 is targeted to lux markets e.g Lighting Control Systems.
The 3002 Medical and Scientific Instrumentation

TI says Use the OPT3002 in optical spectral systems that
require detection of a variety of wavelengths, such as
optically-based diagnostic systems.


The OPT3002 Response Curve, looks a lot like your curve that you deceitfully attempted to pass off as a lux sensor's

.

Notice no filter before the sensor





The OPT3001-Q1 device is a single-chip lux meter,
measuring the intensity of light as visible by the
human eye. The precision spectral response and
strong IR rejection of the device enables the
OPT3001-Q1 device to accurately meter the intensity
of light as seen by the human eye, regardless of light
source.

This curve looks like EVERY response curve for Lux only markets.
Notice the filter before the sensor



More wanton and willing deceit

For a lux meter they add a color filter to tone down the blue and red ends

There you go again trying to Baffle with your Bullshit. "Tone Down"

The ALS curve you show is before filtering. When filtered the range is about 500-600nm.

The CIE Photopic Luminous Efficacy goes from 390 to 770
At 390nm the sensitivity is 12
At 550nm the sensitivity is 99,495
At 770nm the sensitivity is 3

An ALS is very sensitive to IR so the reds are filtered out.

I have to assume you do not understand the need for a cosine sensor

A high end sensor has cosine correction. They cannot use the cheap lux sensors.



The cosine sensor has an aperture much smaller than the sensor. The sensor has many photodiodes. The lux sensor has one photodiode. When the light comes through the aperture at an angle the beam of light will not excite the center photodiode. The angle can be obtained from the cosine of the height (aperture to center photodiode) and the angle of the beam. The angle is calculated by using the acrtangent of the height and distance from the sensor's center to the photodiode that was excited by the light beam.

Additionally a really high end lux meter will use an expensive CCD sensor rather than a photodiode.

My spectrometer has a cosine sensor. That is why ISL works. That is why Greying Geeks lux meter did not follow ISL.

I ntices yu did not understand at least tw of the lines of cde where yu accusded me of fudging the numbers. These two line were taking the cosine into the app.

$angle = rad2deg(atan( $offset/$height));

$distance= $height / cos(deg2rad($angle));

More is posted elsewhere explaining these two line in great detail.

Because a lux meter does not have a cosine sensor it cannot follow ISL or detect uniformity very well. It is NOT a good tool to use for adjusting the height of a fixture.

To see why you think ISL does not work, look at my post where you gave Greying Geek your typically bad advice. I tried to explain the same thing to you but you do not listen to what I have to say.

As you can see, his intensity drop from 4 inches to 8 inches was only about 6% - 51k lux to 48.2k lux. ISL predicts a 75% drop. Even at 24" down he still has not hit 50% drop.

I posted a rely to Greying Geeks post regard ISL

This is the post:

Bottom line ISL works.

 

[GrowLightResearch]

If my target is 1000 uMols of flux density at the canopy, the plant is not going to give a rats ass if its really 900 or 1100.

I totally agree with you. In the 70s I grew with 4' grow lux tubes. Four tube over 5 plants. Here's the thing, the grow lux tubes were always less than an inch from the canopy. I see some of the grow setups here and the fixtures are way too far from the canopy for no legitimate reason. And the grow lux tubes did a better job. The CoBs need to keep their distance because the light is concentrated in small area and create hot spots if they are placed too close. I would NEVER use CoBs for that reason. But there is no reason a Samsung Bridgelux strip cannot be placed an inch away from the canopy. You will have to decrease the current and run them cooler but you will likely get more photons the leaves than a massive fixture four feet over the canopy.

So it not about the 900-1100 µmols, it's about whether you want to continue to unnecessarily pay significantly more for the electricity. So do you really want to continue to taking horrible advice that costs you? That is what I see you doing here.

No one cares it's accuracy here,

But remember all the little stuff adds up.

All I am trying to do is make the world a better place. There is not much else to accomplish that than by improving the production of cannabis and lowering the costs to do so. Less cost and improved yields.

 

[nfhiggs]

The cosine sensor has an aperture much smaller than the sensor. The sensor has many photodiodes. The lux sensor has one photodiode. When the light comes through the aperture at an angle the beam of light will not excite the center photodiode. The angle can be obtained from the cosine of the height (aperture to center photodiode) and the angle of the beam. The angle is calculated by using the acrtangent of the height and distance from the sensor's center to the photodiode that was excited by the light beam.

Additionally a really high end lux meter will use an expensive CCD sensor rather than a photodiode.

My spectrometer has a cosine sensor. That is why ISL works.

Because a lux meter does not have a cosine sensor it cannot follow ISL or detect uniformity very well. It is NOT a good tool to use for adjusting the height of a fixture.

To see why you think ISL does not work, look at my post where you gave Greying Geek your typically bad advice. I tried to explain the same thing to you but you do not listen to what I have to say.



I posted a rely to Greying Geeks post regard ISL

This is the post:

Seriously? THAT's your excuse for the extreme deviation by a factor of more than ten? Come on dude. Man up and put it to the test yourself. Samsung Acuity strips can be purchase for 2-3 dollars each on Ebay Buy ten, put them together in your 2.5 foot tent and see for yourself. Prove me wrong. Hell, I'll even donate the two extra ones I have to the cause.

And for the record I offered exactly ZERO "advice" to GG. I merely commented on the fallacy of the supposed "penetration" advantage of a bright point source light (HPS).

 

[GrowLightResearch]

And for the record I offered exactly ZERO "advice" to GG.

Never said you did. You are pointing to a response of mine t some one else. Did you not understand the importance of the cosine sensor??? I spelled it out in detail in the GG thread. Optical measurements are very complicated. It takes time to understand. Read the explanation in the GG thread. A cosine sensor eliminates all the math that would needed otherwise. ISL would need to be measured on each individual LED without a cosine sensor.

see for yourself

I HAVE seen for myself. I have a cosine sensor. I have no issues with getting accurate ISL measurements. Again spelled out in the GG thread.

The accuracy of a cheap lux meter could account for GG's issues. A $1,200 lux meter has more than 6% error. A decent $150 lux meter can easily have more than 10% error before taking cosine accuracy int consideration. . A cheap $25 meter I would not be surprised if it had 50% error.

Fluke 941 Meter cosine angle of deviation from the characteristics according to: 30 °, ± 2% 60 °, ± 6% 80 °, ± 25%

 

[GrowLightResearch]

Have you ever grown any Cannabis?

Yes. But it's just a gateway plant. I now only grow poppies and cocoa.

 

[nfhiggs]

I totally agree with you.

If you agree with me on that point, then why go on and on about the lack of accuracy of using lux values?

Here's the thing, the grow lux tubes were always less than an inch from the canopy

You say that like its some kind of revelation. Lots of people still use them that way - mostly for vegging, but some people still flower under them as well.

The CoBs need to keep their distance because the light is concentrated in small area and create hot spots if they are placed too close.

You won't find me advising anyone to use them either. Strips work much better.

But there is no reason a Samsung Bridgelux strip cannot be placed an inch away from the canopy. You will have to decrease the current and run them cooler but you will likely get more photons the leaves than a massive fixture four feet over the canopy.

Good luck with that. Document it and let us know how it goes. Those of us that have actually been using them already suspect that's less than realistic (4-6 inches is a more realistic distance), but we'd be happy to be proven wrong. But its going to take more than words to convince me (and quite a few others here I suspect), I need to see a successful grow, and so far you've come up a bit short on actual documented grows. No offense, but so far you've been "all show and no grow" - until you start putting up some legit grow pic and details, no one here is ever going to take you serious. I'm a confirmed skeptic, I believe none of what I hear and only half of what I see, and I suspect there's quite a few more here )

So do you really want to continue to taking horrible advice that costs you? That is what I see you doing here.

That's a mighty big assumption. Frankly, you have no clue what advice I have taken and what advice I have discarded.

 

[nfhiggs]

[

Never said you did. You are pointing to a response of mine t some one else

My apologies. I see you were responding to Wietefras there.

HAVE seen for myself. I have a cosine sensor. I have no issues with getting accurate ISL measurements. Again spelled out in the GG thread.

The accuracy of a cheap lux meter could account for GG's issues. A $1,200 lux meter has more than 6% error. A decent $150 lux meter can easily have more than 10% error before taking cosine accuracy int consideration. . A cheap $25 meter I would not be surprised if it had 50% error.

Fluke 941 Meter cosine angle of deviation from the characteristics according to: 30 °, ± 2% 60 °, ± 6% 80 °, ± 25%

Again, do the measurements yourself under a similar light fixture, not a single bar with a measly 16 diodes. Put together a fixture consisting of 400 diodes in a 22" square within a 2x2 reflective tent and take measurements at the same distances GG did.. Or are you afraid of what you will come up with?

 

[nfhiggs]

My spectrometer has a cosine sensor. That is why ISL works. That is why Greying Geeks lux meter did not follow ISL

So.... we are to believe that because its not a cosine sensor that its somehow going to give readings on the order of ten times what your cosine meter would read at that distance (24 inches). Riiiiight.
Two words:
Prove. It.

You have the cosine sensor. 400 diodes. 22" x 22" frame. 10 strips (I'll even donate my two spares). Reflective walls 24" x24". Put up or shut up.

 

[GrowLightResearch]

"all show and no grow"

Where I live it's a felony to grow one plant. There are also mandatory minimum prison sentences. I do not grow here, that's my story and I'm sticking to it.

Again, do the measurements yourself under a similar light fixture,

I do this stuff for a living. I have made measurements hundreds of times with many types fixtures. I just bought some Samsung and Bridgelux strips. I will make a video. I have shit loads of data files down loaded from my spectrometer software. I thought I did a pretty good job of explaining why GG got those numbers. They sound about what I would expect. I will soon be getting a development kit that has a non-cosine sensor. I am working on the design of a very low cost PAR meter / Spectrometer. It will be a kick ass PAR meter. Not sure how well it will do as a spectrometer. The sensor is being marketed as a... well just see for yourself

 

[wietefras]

The Lux meters do NOT use the full Photopic bandwidth.

...

The CIE Photopic Luminous Efficacy goes from 390 to 770
At 390nm the sensitivity is 12
At 550nm the sensitivity is 99,495
At 770nm the sensitivity is 3

So yet again you post facts that 100% disagree with what you say and support 100% of what I said.

Indeed Lux meters measure the whole range from 400 to 700 (and more). So they are perfectly fine for measuring any light that's useful for plants. Thanks for proving my points yet again. Now if only you actually understood what you were copy pasting then this could have been resolved so much faster.

I have to assume you do not understand the need for a cosine sensor

Well I understand what a cosine correction does and yet again it does exactly the opposite of what you yhink it does. A "cosine sensor" is nothing. Just a made up term. Showing again that you don't know what you are talking about. You don't even understand the terminology.

Maybe this will help: https://oceanoptics.com/product-category/cosine-correctors/

"Cosine correctors are typically specified for applications requiring the redistribution of incident light, such as measuring spectral irradiance of a plane surface in air or other media."
"Typical applications include relative and absolute spectral intensity measurements of LEDs, light sources and other radiant sources. Solar irradiance measurements are also performed with cosine correctors."

They cannot use the cheap lux sensors.

WRONG! There is no relation between the two.

The cosine sensor has an aperture much smaller than the sensor.

WRONG! That's not a cosine correction, but exactly the opposite. The hole is used for spot measuring and ignoring ambient light. To measure ambient light correctly they use cosine correction to reduce the effect of incident angles on sensitivity.

My spectrometer has a cosine sensor. That is why ISL works.

Your sensor did NOT show ISL. You would need to end up with 25% of the light after doubling the distance. In fact ytou ended up more on half the intensity because you are using a strip (line) instead of a point light source.

Now try this again with a full fixture and the light will not drop even that much when you increase the distance.

I ntices yu did not understand at least tw of the lines of cde where yu accusded me of fudging the numbers. These two line were taking the cosine into the app.

$angle = rad2deg(atan( $offset/$height));

$distance= $height / cos(deg2rad($angle));

That has nothing to do with cosine correction of light sensors. I anything it's exactly the opposite again. A cosine correction is supposed to reduce dependence on incident angles. Your calculation does the opposite.

But yes your calculated numbers are bogus. You measure only half a strip (8 SMDs) and on only one edge of the 8 SMD's then somehow it is supposed to magically produce the same numbers as a 16 SMD strip in a reflective tent. In fact it matches for shit. You measurement curve is a completely different curve. I showed this to you also. You can clearly see the reflection kicking in in the measured curve while your calculated curve is miles off there.

It only comes back together at one point again because the slope is so much shallower than in your calculations.

The fakery is not so much in the calculation, but in the fact that you stopped exactly at the point where both curves touch again. Double the distance from that last measurement and your measurement and calculatoion will be miles apart.

Because a lux meter does not have a cosine sensor it cannot follow ISL or detect uniformity very well.

WRONG and WRONG! A lux meter DOES have cosine correction. It's that white disc you see on any lux meter. Most lux meters also have a spot measurement lid. So they can do both.

You do not want to use a light sensor covered with a hole to measure uniformity. You want you have a cosine correction (the real one and not the opposite one you seem to think is cosine correction), because you want to measure all light coming from all angles and not just the bit of light straight above the sensor. Plants don't look at a tiny beam of light either.

I do this stuff for a living.

LOL. Do you work as a janitor at a University? Screwing in light bulbs for a living?

 

[GrowLightResearch]

he order of ten times what your cosine meter would read at that distance (24 inches). Riiiiight.

What did you not understand about my write up n the cosine issue. I went through step by step why one LED is only going t measure like 4% when you are expecting to it to drop 16% from 24" to 20". WTF are you talking about 10X??? His measurements are off like 7%. At a flux of 30200 at 24" at 20" by ISL it would be 43400. His reading was 39900 39900/43400 = 1.08771 So 8% off. No where near 10x or100%. As the height gets less the cosine issue does get much worse. But if he went from 44" to 24" the ISL would nearly match. It's the steep cosine angle that screws it up.

 

[GrowLightResearch]

Indeed Lux meters measure the whole range from 400 to 700 (and more).

No! Wrong!

Lux meters measure 490nm to 610nm.

A cosine correction is supposed to reduce dependence on incident angles.

No! the meter makes the correcting caused by the angle. What you are saying is gibberish and nonsensical. More of your attempts at baffling bullshit.

You measure only half a strip (8 SMDs) and on only one edge of the 8 SMD's then somehow it is supposed to magically produce the same numbers as a 16 SMD strip i

Yes exactly. When the sensor is in the center the readings from the center t the end will be identical going from the center t the opposite end. Duh??

because the slope is so much shallower than in your calculations.

Slope??? What slope???

because you want to measure all light coming from all angles

That is not cosine correction. What cosine correction does is fix the ISL for the rays coming in at an angle.

You do not want to use a light sensor covered with a hole to measure uniformity.

Yes you do, You obviously just do not understand. Go back and look at the diagram of the cosine sensor. It would not work without the aperture. The smaller the "hole" the better the accuracy. Read up on how a the slits in a spectrometer work.

LOL. Do you work as a janitor at a University? Screwing in light bulbs for a living?

 

[nfhiggs]

What did you not understand about my write up n the cosine issue. I went through step by step why one LED is only going t measure like 4% when you are expecting to it to drop 16% from 24" to 20". WTF are you talking about 10X??? His measurements are off like 7%. At a flux of 30200 at 24" at 20" by ISL it would be 43400. His reading was 39900 39900/43400 = 1.08771 So 8% off. No where near 10x or100%. As the height gets less the cosine issue does get much worse. But if he went from 44" to 24" the ISL would nearly match. It's the steep cosine angle that screws it up.

LOL. 4" = 51000. 8" = 48000. When it should have been 12500 - that's nearly 400% difference. At 16 inches ISL predicts 12000 (from the 8" measurement) and he got 43000. That's 350% difference. At 12 inches he was at 45000, and at 24 he was at 30000 when ISL predicts 11.200 - still more than 300% off. If you calculate from the first measurement, its off by 10 fold. Can you not see that fact?

I repeat - do an identical measurement yourself with YOUR SENSOR. 400 diodes. 10 strips. 2 ft x 2 ft tent. same measurement distances. What are you afraid of?

 

[wietefras]

Lux meters measure 490nm to 610nm.

Not just WRONG but in fact complete bullshit. Even you must know this.

You obviously just do not understand. Go back and look at the diagram of the cosine sensor.

WRONG!
Anyway by all means, lets look at the picture of the "cosine sensor" you posted, because in fact that is a sun sensor. It's designed to be mounted on satellites so it can determine it's vector to the sun. This has absolutely nothing to do with cosine correction or even with light meters in general. How on earth do you even come up with this ridiculous shit?
http://www.caravan-ny.com/digital_sun_sensors.html

Clearly a sun sensor is not in the least useful for measuring "spectral radiance of a plane surface" as the people who actually sell cosine correctors put it.

Whatever you think a cosine corrector should look like, the main component is a white bit of diffuse glass/acrylic. Cosine correction is designed to give the light sensor a 180 degree field of view. Exactly the opposite of what you think it does. It's supposed to measure ambient light and not just a single beam like spot meters do (with a very narrow FOV). Most lux meters can do both, but for measuring light intensity for plants and light uniformity, you will want the cosine correction.

It's uncanny how everything you say is always pretty much 100% wrong. Seriously, is this some elaborate trolling expedition or are you really this clueless?

Anyway, like nfhiggs says, just measure it properly for once.

 

[GrowLightResearch]

Cosine correction is designed to give the light sensor a 180 degree field of view.

I did an image search for a cosine sensor. That image is what a cosine sensor looks like, it was just being used for different end use than measuring light.

A cosine sensor provide a method to measure the angle of the light beam. The light enters through the aperture and excites one of the pixels of CCD image sensor. It has nothing to do with the FoV.

The meter's CPU get the light level of each CCD pixel. From the the angle is know becasue each CCD pixel is only excited at a a finite angle.

With the angle and energy level the unit can then calculate the ISL for that specific angle.

From the angle of the CCD pixel the unit can calculate the relative distance from the sensor to the light source using the cosine of the angle. When the relative distance is known the flux level of that signal the level of energy detected by the pixel is then mathematically adjusted relative to the level of the center pixel.

The cosine formula is: distance = 1/cosine(angle) where distance is relative to the center pixel

Using my experiment with the 3.93 to 7.75 in four 0.786" (20mm) increments.

On the left side of the diagram is the ISL for the center LED which is compared to the far right 8th LED from center.

The point here is as the distance is increases for the center LEDs the distance for the right most LED increases at a rate much less then the distance for the center LED. The cosine sensor give the angle. In my calculations to get the angle I used:

$angle = rad2deg(atan( $offset/$height)); Where $offset is the horizontal component and $height the vertical component.

The spectrometer gets the angle from the cosine sensor.

As the center LEDs distance to the sensor changes 20mm (0.786") the distance from the right most LED to the sensor changes from 8.94 to 9.43 .

So while the center LED traveled 0.786" away from the sensor the right most LED traveled only 0.49".

So the ISL must be calculated separately using the 0.49 distance rather than 0.786".

What you have been saying is the other LEDs flux should increase at the same rate of the center LEDs. But that does not work becasue the other LEDs are moving toward the sensor at lesser distances.

But you still say the ISL should be 110 when it is going to be less because all the other LEDs have a lesser travel distance toward the sensor for each decrement in height.

And that is what a cosine sensor does.

 

[GrowLightResearch]

Put up or shut up.

In that case seeing you say it like a snotty little piss ant, I'll shut up. I have no desire to converse with a person rude as rude as you.

 

[wietefras]

I did an image search for a cosine sensor.

WRONG! There is no such thing as a "cosine sensor". It's called a cosine corrector or cosine diffuser.

That image is what a cosine sensor looks like, it was just being used for different end use than measuring light.

So you admit you know jack shit and posted a picture of a device that has nothing to do with a the subject at all. Check. Guess that's some progress.

It has nothing to do with the FoV.

WRONG! A cosine corrector's purpose is exactly to increase the FOV of a light meter to 180 degrees.

It's basically a white diffuse disc that takes in the light over a 180 degree FOV and transmits it to the sensor/detector below. The only hole involved here is the one the disc sits in to cover the sides of the cosine corrector to stop the light outside the 180 degree FOV reaching the sensor. So the white disc sits in a hole to shield it from the sides. There is no tiny hole or slit underneath. You are confused with spot meters which are designed to measure a single spot and ignore ambient light. That's pretty much the opposite of a cosine corrector.

A light sensor without this cosine corrector has narrow FOV and a high sensitivity to incident angles.

Picture of an actual cosine corrector (Li Cor):


Or this DIY cosine corrector:

Posted this already, but here goes again:

Cosine correctors are optical diffusers ... to collect signal from 180° field of view. Cosine correctors are specified for setups requiring the redistribution of incident light, such as measuring spectral irradiance of a plane surface in air or other media.​

BTW You probably think that "measuring spectra radiance of a plane surface" means that you measure a single ray of light, but you are WRONG! there too since it does not.

Or this one:

Cosine Correctors – Cosine correctors are planar diffusers placed in front of a fiber or a detector. This provides a 180 degree field of view, and the energy that reaches the detector is the component of the light that is normal to the plane of the surface of the diffuser. These are used primarily in radiometry, to define a measurement plane.​

And this is what LI-COR has to say about Cosine correction:

Cosine Correction: A light sensor should be able to accurately measure light at angles to ~90 of normal incidence (0), and a cosine-corrector allows this. Two cosine-correction types exist - one type is a hemispherical plastic diffuser dome (used by Apogee and Spectrum Technologies), while the other is a plastic cylinder (that should rise slightly above its housing in order to properly collect light, which the Li-Cor sensor does).

All sensors are advertised to be cosine-corrected, meaning their response will be the same to a beam of light, regardless of that beam's angle of incidence to the sensor​

The meter's CPU get the light level of each CCD pixel.

WRONG! You are talking about that sun sensor again. That is NOT at all how a light meter works. Light meters don't use CCD. They usually use one light detector, or in the case of Li

What you have been saying is the other LEDs flux should increase at the same rate of the center LEDs.

WRONG! I'm saying that if the distance doubles the light will have spread 4 times as wide. Which means that the lights from surrounding light sources also spread and all that fills up to the same amount again.

If you could have 100% reflective walls then you could extend the distance infinitely and still have the same light intensity. Which is how glass fiber works to transmit light over vast distances.

But you still say the ISL should be 110 when it is going to be less because all the other LEDs have a lesser travel distance toward the sensor for each decrement in height.

WRONG! You yourself measured 164 down from 440 after doubling the distance. I said an inverse square relation would mean it should have been divided by 4 to get 110. Now you say it should have been less than 110 for ISL to apply? You are so wrong you even start disagreeing with yourself. Oddly enough you then still manage to be wrong. Amazing.

Too bad you are too chicken to take up nfhiggs gauntlet. Guess you couldn't even build a fixture. That crappy looking led strip is all you have huh? Otherwise how much effort would it take to take a few measurements and finally see how wrong you have been all this time?

 

[ttystikk]

@nfhiggs how are your plants doing?

 

[GrowLightResearch]

Now you say it should have been less than 110 for ISL to apply?

That was a mistake. But you knew that already, more or less.

Guess you couldn't even build a fixture. That crappy looking led strip is all you have huh?

Those two crappy PCBs, are from an inter-canopy design done for a landscape architect becasue 220 Central Park South is shaded and they need supplemental lighting. They are happy, I am happy, very happy, why can't you be happy.

Too bad you are too chicken to take up nfhiggs gauntlet.

I previously said what I would do, his test setup is inappropriate and he talks like a piss ant.

In the mean time you could look at how well the ISL worked for GG from 24" to 20". When you get down to the 8" to 4" the measurements have much more error. Not so much from 24" to 20".

Even a good lux meter will have 25% error at 4" especially where the LEDs are around 80°. Look at the specs for an okay lux meter like a Fluke specifically the error in the cosine correction.

Is that due to reflection? or overlap? or is ISL working and the lux meter used the wrong tool?

I would love to chat some more but I have some light bulbs to screw with at a University.