New DIY LED light

wietefras

Well-Known Member
On the other hand to measure PAR you MUST have wavelength intensities.
You can keep repeating this fallacy, but it's simply not true.

So I guess you are the "scientist" who got kicked out of his house because it was in such a crappy state. And proud of it. You seem to have a lot in common with him at least. Unfortunately you're back loud mouthing and you still really have a lack of understanding of how this all works in reality. Too fixed on half understood middle school physics laws to see reality.

The lumen to PAR conversion factor creates a factor to go from lumen curve to PAR "curve" for a known SPD. It's simple maths really based on multiplying both curves. There is a thread here on the forum showing how to do this math.

Your formula and measurements for inverse square are simply wrong too:
1) Inverse square in a reflective tent only applies when you are very close to a point source of light. That's exactly what you orchestrated in your "test" by using only one strip and starting right on the light. In practice this never applies since we hang the light for correct uniformity. Which means light will be overlapping everywhere and there is not just a bright spot in the middle of the tent.

2) Your results are clearly not an inverse square result. When you go from 3.93 to 7.85 distance then that's double the distance. You should end up with a quarter of the light. So it should be 110 there and not 164 or 166. It's not even close.

So, you have in fact proven us right. Thanks.

Try to understand this from a real world perspective. Hang a full fixture in the tent. Measure the light intensity where uniformity is "optimal" (80% rule). Now double the distance. You will see a drop in light of about 10% to 20% depending on how poorly your walls reflect and how close you are to the wall. You do not end up with a quarter of the light intensity as inverse square law would predict.

Reality check on the DIY PAR sensor also. I have seen things like that done from the aquarium guys from years ago. The problem is that you need to calibrate these things. Which is where the whole thing falls apart. It is also very sensitive to spectrum changes since it's not a true PAR meter, but a lux meter with an added filter to sort of flatten the measurement curve of the sensor a little.
 

GrowLightResearch

Well-Known Member
You can keep repeating this fallacy, but it's simply not true.
Needs citation.

) Inverse square in a reflective tent only applies when you are very close to a point source of light.
Needs citation.

When you offer some proof then I will consider what you have to say. But when science does not agree with you and I have made measurements that prove to me you are wrong, I'm not inclined to take your word. Since there is no evidence to support you claims our discussion is over.

Your formula and measurements for inverse square are simply wrong too:
I will give you it does appear that way. That is exactly why my first attempt had too much error. If I take the entire strip as a point source it will not work out well. I had to look at each individual LED as a point source and sum their flux.

The format of the ISL formula I use is E2=(d1/d2)² x E1
Where
D1 = 3.93
D2 = 4.71
E1 = 440
pow(x,y) = x to the power of y

I used basic right angle trigonometry to calculate the angle and got the flux at that angle from the datasheet's spacial distribution graph (stored in the variable $intensity). This experiment worked so well I now create an $intensity table containing the flux for each angle from the spacial distribution (where 0° = 1) for every LED I purchase.

$lux = pow($ref/$distance,2) * $intensity[$angle];

It's inverse square times $intensity[$angle] (i.e. spacial distribution characteristic ) as 15 of the 16 LEDs are at an angle so the distance between the spectrometer's sensor and 15 of the LEDs are not changing as much as the height. Also as the angle changes the flux changes as shown in the datasheet's spacial distribution. Only the LED directly above the sensor is going to follow the height.

I ran that experiment Aug. 18, 2016. I have since improved the accuracy. In this experiment Distance 1 is 3.93". I should have, and now do, calculate a unique Distance 1 for each LED.

Spacial distribution characteristics are key to ideal height and uniformity. That little snippet of code I used here is the key to finding the ideal fixture height for ideal uniformity.

So I guess you are the "scientist" who got kicked out of his house because it was in such a crappy state. And proud of it.
You sure got that wrong too. I do not know how your mind functions most of the time, smh. Even if it were true, it is a really fucked thing to say to someone. You get it wrong too often. Like your mind is a sieve. Your ability to recollect things is error prone. Sometimes I think you must just make this shit up or just pull it out of your ass. Like it does not matter if what you say is true just as long as you can say I'm wrong. You do have the ability to say things erroneously with very strong conviction. It makes you sound like you know about what you are saying and others may believe it to be true. Like your interpretation of the inverse square. I think that was a jumble of describing ISL and a Lambertian surface.

Try to fact check your nonsense. And use citations when saying I'm the one that is wrong.


Good bye.
 
Last edited:

GrowLightResearch

Well-Known Member
Even your two "favorite enemies" have done a lot of good things for the community in the past, even if it does not look like that for you.
I really want to see how you build the PAR meter, even if it costs $50 at the end. It though belongs more in the field of electrical engineering, but nonetheless it is extremely interesting!!!

Yes I agree. I am really upset. I order the development kit this past Sunday. It was supposed to arrive Wednesday. Still not here. I also order the $25 one on Sunday. Latest scheduled delivery by FedEx is Thursday. Last tracking said it was in Ocala (very close to me 5-10 miles) bu ton it's way to Orlando over 100 miles away.

Bu the good news is I started studying the data sheet. I had my concerns there wasn't enough wavelengths being measured to be accurate enough (for me). But eah photodiodes covers a 40nm width. The entire PAR bandwidth is covered. Then I checked the conversion factors for the 40nm bandwidths worried that the formulas for calculating PAR from radiometric watts would vary too much and introduce more error. But, no! This is going to be so cool!. It is a true spectrometer and can be built for $25. For a few bucks more it will be kick ass.

PhotodiodeArrayVBGYOR-SPDwaveformscr.jpg

Each photodiode covers a 40mn wide stretch of bandwidth,
If I can and believe I can, lower the gain on each sensor and get the SPD curve under the photodiodes 50% point, called Full Width Half Max (FWHM), I should be able to count every PAR photon that reaches the sensor.

There is an open source app. I was hoping they were calculation PAR and Lux. But no. Crap I would not use so I will start the app from scratch. . But the good news is I have written hundreds of apps and this one should be a piece of cake. I'll do a quick app, I figure a day at most to get a basic app running. but where I'm going with it is to improve the accuracy by calibrating it with my spectrometer so that is traceable to NIST standards, even if unofficially. This ALS chip is kick ass, its own calibration they say is for life, the sensor will never need to be calibrated. My calibration is for the software algorithms I will design. My goal is to make it as accurate as the spectrometers costing $1000+. The $25 board needs a USB ort rather then being connect to a dev board.

Even your two "favorite enemies" have done a lot of good things for the community in the past,
I understand. I can tell they actually know stuff they but either haven't used it recently or were a little confused or both. This LED and optical measurement stuff is very complicated. I spent a good percentage of my time for two years trying to get it right. I have read a shit load of books. But still I had to apply what I learned to a real life issue.

The problem I have with these guys is they say I wrong no matter what. They never have anything to back up what they say, the just say I know this and you are wrong. I'm sure they have garnered some respect before I came along and others believe them and their bullshit.. And this most recent remark from @CobKits with him laughing at me. In the past I have always cut @CobKits some slack because he is a vendor and I do not like to interfere with someone's livelihood. I had commented he was giving bad advise, so he too is now always taking a shot at me. Now the gloves are off.
 

wietefras

Well-Known Member
You can keep repeating this fallacy, but it's simply not true.
Needs citation.

https://www.apogeeinstruments.co.uk/conversion-ppfd-to-lux/

But when science does not agree with you and I have made measurements that prove you are wrong I'm not inclined to take your word. Since there is no evidence to support you claims our discussion is over.
You have give us measurements which clearly demonstrated without a doubt that inverse square does not apply to the height of the fixture over the plants. So you have proven yourself wholly and utterly incorrect on your original claim. Or did you get 110umol/s/m2 at 7.93?

So yes our discussion is over. In fact there never was any discussion. This is just the lot of us trying to get some understanding into your stubborn head.

The fact that inverse square is used in the calculation of how the light spreads from a point source, does not mean that your original claim has any validity. You claimed that you need to hang the light as close as possible to the plants because of "inverse square law". Which is bullshit because of reflection and overlap of multiple light sources.

spacial distribution characteristics is also one of the many flaws in the spreadsheet LER QER scenario.
Ah so you do know that it's is possible. LOL. For fucks sake though. Don't drag more bullshit into this. You are also 100% wrong on this claim, but it really is becoming too much. I don't have time to correct all your misunderstood shit.

Anyway, back to spacial distribution. Indeed your calculation is completely useless on that too. Your calculations are simply a textbook example for calculation of a point light source. You "forgot" to add in reflection.

The fact that that didn't matter that much to your calculated values (at least after you manually corrected it all to sort of "match") should have alerted you to the fact that your test was crap. In fact that bend in the line where reflection starts to kick in should also have been a clear indication that something was wrong. That's where the reflection starts to kick in.

The problem is that you only go for confirmation bias and let the cognitive dissonance take care of the rest.

NoFucksGiven_CrapTest.png

See the bottom bit. That's the almost linear light drop off you will get when you increase the height beyond the height for optimal uniformity.

I personally have written much more complete piece of software to do this type of analysis which does include reflectivity. The reflectivity on your walls is so crappy that I had to drop reflection down to 50% to get anything near your measurements.

Which produces light distribution plans (blue indicates area's with insufficient light for adequate uniformity):
At 3.93
Distribution_Strip16_3.93.png

At 6.28:
Distribution_Strip16_6.28.png

So from that's is also painfully clear that the fact that you took a strip instead of a full fixture completely skewed the test. Although it's still 100% clear from your measurements that inverse square does not apply the way you claimed it did.

As nfhiggs and I already explained you need to fill the entire canopy with light like you would when you were actually growing plants in that tent. That's the real world test when the overlap and reflection counteracts the effects of inverse square. Then measure the light intensity at the "canopy level". Double the distance and measure again. The light will not have dropped to 25% of the first measurement, but to something related to the reflectivity of your walls.


Try to fact check your nonsense. And use citations when saying I'm the one that is wrong.
I didn't name facts. I'm simply trying to explain to you how to use your middle school textbook physics


How am I going to give citation on the fact that the light will not disappear into the distance when it gets reflected back? Any idiot should be able to understand this.

Try to get past your middle school basic physics and try to understand the more complex situation of reality where there are multiple light sources and reflection. Your own measurements showed that you are wrong and that the rest of us are right. Height is used for uniformity not for intensity. Sure you want the light as close as possible to the plants, but that's to reduce wall losses and not because of inverse square law. You also don't want them too close because then you get poor uniformity with hot spots and dark spots.

Try a better example in your "simulator". Create a fixture of say 100 miles wide and 100 miles deep with a matrix of SMDs uniformly distributed .7" from each other in x and y directions. If you calculate (or measure) the light intensity in the middle of this at 1 foot and at 10 foot, the light intensity will be almost identical. That's why green houses do fine with their lights 4 or 5 meter up in the air. You only loose light at the edges of the grow area where it will disappear on the walls due to less than 100% reflection.

Or alternatively imagine how light travels through glass fiber for kilometers. If Inverse square applies it would run out in mere meters. It doesn't because the light bounces off the walls of the fiber.
 

wietefras

Well-Known Member
The problem I have with these guys is they say I wrong no matter what.
We only call you out on on the cases where you ARE wrong. The real problem is that you know too little of the matter to understand that you are wrong and why.

The world does not consist of overly simplified text book examples.

You should be thankful that we actually take the time to try and teach you something. Just get over yourself and accept the truth for a change.
 

wietefras

Well-Known Member
Did you understand my reason why I did not get 110 at 7.85"?
The fact that you think you can argue your way out of this is exactly why this is taking so long.

You can call that condescending, but it is what it is. We have tried to explain to no avail and even after you yourself presented us with measurements that actually proved what we told you, you still keep calling us idiots.

This discussion is not that inverse square law is a thing or not, but you specifically claimed that inverse square law applies to the height of the fixture over the canopy. For that to be correct it would mean that the light intensity will have to be cut by four for every time the distance doubles. That clearly does not happen unless you use a point light in a huge black room with no walls.

Inverse square just says that light is spread over 4 times the surface area for doubling the distance to the source. That does not mean light is lost. It's just falling on other plants or on a wall. Which is where the overlap and reflectivity of the walls comes in.

All of which means that in real life the light intensity diminishes by ratio of how much of the light is absorbed by the walls and not by inverse square.

Typically my fixture is never more than 12" above the canopy, more like 4" to 6". So not much reflectance to begin with.
It's not just reflectance, but also overlap.

Try the same test with a whole fixture instead and it will be even more linear. Also, better measure average light intensity instead of just one spot.

ps It was you who who started cursing, name calling, shouting, condescending, deriding and all that. Pretty much from the start. There were 4 or 5 people trying to help you understand this and 2 or 3 already dropped out because of your foul mouthed emotionalism. Don't try to pretend this is somehow my fault. I just treat you back the same way you treat us.
 

wietefras

Well-Known Member
So you do not believe me. Then you must think I am lying about the results of my experiment?
Your measurements quite clearly show that inverse square does not apply to the distance of the fixture to the plant canopy. It proves you are wrong. That's it really. Done deal.

If inverse square applied to the distance of fixture to canopy like you claimed then you would have seen:
3,93 440
4,71 306
5,50 225
6,28 172
7,07 136
7,85 110

The main point to start understanding is that inverse square law does NOT mean that light disappears. It just means it's spread out more.

I guess by now the reality is starting to sink in that you were wrong, so you commence with the goal post shifting. Sad really. Although it's really hilarious that you try to pretend you were right by claiming that overlap is ruining the effect of inverse square law. That is exactly what we told you!

For a whole fixture with many points of lights spread out uniformly over an area and/or closed in by reflective walls (which is the norm for a grow area), inverse square does NOT apply on the light intensity related to the distance. You initially claimed it does and we told you (correctly) that you were wrong.

So the truths obviously still stand:
1) You adjust height of the fixture for desired uniformity. You do not adjust the height of the fixture to some spot value on a light meter.
2) Increasing distance between light fixture and canopy does diminish (average) light intensity, but only because of increasing wall losses. This is a more inverse linear relation and not comparable to inverse square
3) It doesn't matter if people use PAR or lux to measure relative light intensity for light of a single SPD. Even for burples (although it will be somewhat less accurate for those).

If you are looking for other people who realized inverse square does not apply to fixtures at a normal distance from the canopy:
http://rollitup.org/t/riders-first-foray-into-led-strip-lighting-samsung-h-series-gen3.950197/page-5#post-13864152

ps Are you really that pathetic that you posted someone's e-mail address in an attempt to pretend that you weren't wrong all along? Wow. Just wow. That e-mail doesn't prove anything anyway.
 

GrowLightResearch

Well-Known Member
Your measurements quite clearly show that inverse square does not apply to the distance of the fixture to the plant canopy. It proves you are wrong. That's it really. Done deal.
I calculated the Inverse square for each LED, summed each result then measured with a spectrometer with calibrations traceable to NIST standards and the calculated values matched the measured values. You cannot tell me I am wrong.

You are either just trying to fuck with my head or you are just too stupid to understand.
Either way this is over.
 
Last edited:

nfhiggs

Well-Known Member
I want to see more of @nfhiggs light strip working.

How is the plant under it doing?

Good discussion here, but could do without so much emotionalism.
Here is the photo album for the first plant flowered under it:
https://imgur.com/a/i5ncs

Yield was 7.5 ounces.

And here is the second one:
https://imgur.com/a/XPPi4

Both plants did quite well. Second one suffered a bit from high humidity at 7-10 weeks, causing bud rot in three colas. I just harvested her today and its looking like it might be 8 ounces after drying, maybe a bit more.
 

nfhiggs

Well-Known Member
I calculated the Inverse square for each LED, summed each result then measured with a spectrometer with calibrations traceable to NIST standards and the calculated values matched the measured values. You cannot tell me I am wrong.

You are either just trying to fuck with my head or you are just too stupid to understand.
Either way this is over.
In another message (that's gone now) you were asking for other peoples fixture measurements - Graying Geek did some Lux measurements under his 2x4 ft Bridgelux EB strip fixture, using eight 4 ft strips:
https://www.rollitup.org/t/bridgelux-eb-series-on-a-2-x-4-scrog.943414/page-13#post-13841110

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.
 

wietefras

Well-Known Member
I calculated the Inverse square for each LED, summed each result then measured with a spectrometer with calibrations traceable to NIST standards and the calculated values matched the measured values. You cannot tell me I am wrong.
You claimed that inverse square would apply to the whole fixture. In fact you produced a whole herd of insults to enforce how right you were that inverse square did not apply to fixture height.

Distance between light and plant canopy is only relevant for uniformity. You should not try to alter light intensity with the height of the fixture. That's just a waste of photons on the walls. Although some people are under the incorrect impression that inverse square law applies. It doesn't in our case since the light reflects back and/or overlaps with other COBs, strips, smds etc.
Wow. Laws of physics do not apply to you. Hard to counter that. Fantasy land. Delusional. Good luck. Good riddance.
We (at least 4 people) told you over and over that overlap and reflection would counter the inverse square effect and now you claim that we should ignore overlap and only look at light from a single SMD. You have got to be kidding me.

Next step is that we should ignore reflection? Oh no wait, your "grow lights" consist of a single crappy looking led strip. So no reflection needed there when you hang a single strip at 6" over the plants in the middle of a tent. Brilliant. That must sell like hotcakes.

To summarize I measure the PPFD at 12" and at 24" in a tent with reflective sides (Gorilla tent) and my spectrometer measurements match the calculated measurements for the 12" difference using the inverse square law. Does a Gorilla tent not have your mystical reflexive powers? Please 'splain that to me.
So are you saying that you lied here? You saw 25% of the light at 24" from what you saw at 12" right? Or was that also "a bit less and therefore ISL"


Using a Lux Meter? Wrong tool for the job.
For fucks sake man. Stop trying to waste people's money with your ignorance.

Lux meters work perfectly fine for relative visible light intensity comparisons. How can you not understand these basic things and then pretend to be a "grow light researcher".

Malocan measured a whole tent with a spectrometer. He saw 20% decrease in average light intensity when moving from 12" to 24". Not inverse square. Not even close. Stop grasping at straws.

As usual you were 100% wrong. People called you out on your fallacies and you back peddle, blameshift and flat out pretend the truth doesn't hold. For once in your life, try to be a man, get over yourself, admit that you were wrong and learn something.
 
Last edited:

Mullumbimby

Well-Known Member
Here's a question for @GrowLightResearch :
Have you ever grown any Cannabis?
I've grown some. I use a Lux meter to compare the performance of my three different light systems. It works fine.
Two of my lights each have 4 drivers and, on each fixture, I need to tweak my 4 pots to balance the 4 light bars. the meter works fine for this too.
Once I have chosen my light sources (COBs, in my case) the determination of the quality and intensity of the spectrum is measured by the quality, and to a certain extent, the yield of my plants. You have to inhale to fully appreciate how this works. The inhalation also helps me, in a small way, to come to terms with a life surrounded by idiots.
If I've chosen an inappropriate model of light source, a deep knowledge of the PAR or PPF density is almost irrelevant to me, given the choices available to apply corrections (and there are choices).
I don't know for sure, but it may help you to dwell on these items for a while. Perhaps then, you could go away, convinced that we are a bunch of Neanderthals, who will never be able to appreciate the superb level of sophisticated thinking and analysis that you are capable of.
A sort of 'win-win', if you like.
 

nfhiggs

Well-Known Member
Using a Lux Meter? Wrong tool for the job.
You said in you your original post "PAR for mono or Lux for white"...

It is a legitimate measure of intensity - whether its appropriate for determining PAR isn't relevant to that. Please don't backpedal. You asked for some data and I provided some.
 

GrowLightResearch

Well-Known Member
convinced that we are a bunch of Neanderthals
I look at it as the blind leading the blind. No offense,but you are taking bad advice if you believe anything these guys have said.


the determination of the quality and intensity of the spectrum is measured by the quality,
BTW, Lux meters do not measure spectrum. so much for your "quality" spectrum. Lux meters typically measure the number of photons between 500nm and 600nm which is were the majority of human eye response is. It is also where the wavelengths important to photosynthesis are not.
 

GrowLightResearch

Well-Known Member
Malocan measured a whole tent with a spectrometer
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.
 

GrowLightResearch

Well-Known Member
The fact that you think you can argue your way out of this is exactly why this is taking so long.
I do not need to "argue" out of it. My measurement are exactly what they should have been. You would know that if you had the ability to understand optical measurement.

It's not just reflectance, but also overlap.
At least you are keeping my eyeballs fit. You cause my eyes to roll a lot. They almost got stuck in the back of my head with your overlap remark. You know a lot about this stuff. Unfortunately you only know enough to be dangerous.
 
Last edited:
Top