Re: LED Grow Light
 

Inverse square is only for a point source (or similarly compact source of any shape). If you're moving a linear source (like a fluorescent tube) away from an object to be illuminated, then the light intensity is inverse linear as long as the length of the source is much greater than the distance to the object. If you're providing lighting to an entire room, distance is irrelevant: an array of supplemental lights five feet above an expanse of tables in a greenhouse will provide the same amount of light as it would if it were eight feet above the tables.

Re: LED Grow Light
 

hogwash

I provide links to scientific laws and mathematical equations and you make some mushmouth post with weasel words like yeah so if a train leaves a station going MUCH faster...

next

Since the most common flouro lengths are 4ft and 8ft you may want to rethink this statement.
Unless five is MUCH less than four, or eight is MUCH more than eight.


flouro tubes ARE a point source so long as they are horizontal in relation to a horizontal canopy
We (when I say WE I refer to everyone in the world using actual math) are measuring distance from plants to lights.
So long as the canopy and tubes are parallel then the distance will be identical to a point source with the exception of reduced intensity light from for example the opposite end of the tube.
If you apply inverse square then you will see this is very little.
Hell if you apply inverse linear it's plain as well.
Similar to saying that sprinklers will fill a bucket as fast as a hose put directly in the bucket.

No one is trying to find their shoes across the room here so once again your statement holds no water.
We are NOT providing light to the entire room, we are lighting the leaves of a plant which is presumably under the light.
Anyone who has ever seen a reflectorized light can see that your statement cannot apply.


Too reiterate my previous points.
HPS currently rules the indoor plant world with the benefit of radiant heating if you are overwintering in an outbuilding.
The cost of running a 600W will be less than replacing your bananas come spring and they might never realize it's winter.
Flouro lights are OK.
CFLs can be refelctorized and hung individually over each plant or t8 tubes provide good output with low initial investment and easily sourced parts.
BE SURE TO GET THE LIGHTS AS CLOSE AS POSSIBLE WITHOUT SCORCHING.

8ft away is a joke.



I apologize the the forum members if this post seems undeservedly vitriol.
I would be lying if I said I wasn't mad to receive a email telling me that someone with 7 posts came to dilute my comment with mush mouth psuedo science gobbeltygook.

I tried to make the best, most scientifically and empirically supported post in hopes of benefiting everyone here. I've been working on some stuff to post in the wiki come spring, mostly trying to remove all the cannabis related material as I find it inappropriate.
This is proving much harder than my Link-O-Rama idea.



If DSWS would like to post links, research and other information then I would be VERY happy to have been educated, even at the expense of looking like a arse for this post.
Knowledge is far more important than my ego.

Re: LED Grow Light
 

It only takes high-school math to derive the inverse square law.

When you talk about a four-foot-long fluorescent light having to be within inches of a plant, it's inverse linear.

Re: LED Grow Light
 

Like I said..."de ja vu!"

Re: LED Grow Light
 

I'd consider this an ad hominem attack if we were standing in the same room and I'll treat it as such here.

This from a guy who says things like....

:ha: :ha: :ha:

I encourage you to try chewing some.
Be sure to post a pic of your teeth when done.

:lurk:

I'll not have my busy day held hostage by someones refusal to post more than a single sentence reply or attack my intelligence so I'm done with this thread.
I'm even a little sorry I ever tried to help.

Be forewarned folks that if dsws cannot explain, articulate, and post links evidence to back his points then his statements should be disregarded.

See you ladies in the spring.

Re: LED Grow Light
 

I learn something new everyday

Inverse Square Law for Light

Re: LED Grow Light
 

It works for protons and neutrons, so should also work for light.

Re: LED Grow Light
 

I just noticed the edit to #22. If I had seen it before posting #23, I wouldn't have been so short. I expect to get something posted on Monday, either a link or a more detailed explanation of the transition from inverse-linear to inverse-square with increasing distance to a linear source of finite length.

Re: LED Grow Light
 

I'll write this so it will (I hope) be intelligible even if the picture doesn't show. My apologies for the redundancy, if it does.



To find the intensity of light from a linear source, first draw a perpendicular to the source going through the point where you want the intensity. (In the diagram it's on the source, but that's not necessary. If not, you just subtract the term that's off the end instead of adding.) Call the distance from the observer to the source along the perpendicular D, and the lengths of the parts of the source on either side of it L1 and L2. Call the angles from the perpendicular to the ends of the source A1 and A2.

Intensity is proportional to angular area of the source. (Apparent brightness is constant with distance, because both the apparent size and the amount of light from each infinitesimal bit of area decrease as the square of distance.)

The tangent of A1 is L1/D, and the tangent of A2 is L2/D. So A1 is arctangent of L1/D, and A2 is arctangent of L2/D. Width is negligible, so angular width is proportional to 1/D. Intensity is arctan(L1/D)/D + arctan(L2/D)D.

This video shows essentially the same formula, arrived at by taking the inverse square law as given and just doing the calculus: From inverse square to inverse linear - YouTube

So for example, consider a point at six, twelve, and eighteen inches below a 48-inch fluorescent tube, one-fourth of the way along. I'll put the lengths in multiples of 6", and calculate relative intensities compared to 6" below. The relative amounts of light are

6": (arctan(2) + arctan(6))/(arctan(2) + arctan(6)) = 1
12": (arctan(2/2) + arctan(6/2))/2(arctan(2) + arctan(6)) = 0.405
18": (arctan(2/3) + arctan(6/3))/3(arctan(2) + arctan(6)) = 0.225

By inverse-square, it would be 1, 1/4, 1/9.

Note that by going from 18" up to 12" below the light, the brightness increases by a factor of 1.8, whereas inverse square would give 1.75. So it hardly makes a difference when comparing those distances. But when you go from 18" to 6", inverse square would give you a nine-fold increase in brightness, whereas it actually increases only by a factor of 4.4. Inverse square is off by a factor of two.

Re: LED Grow Light
 

The kind of situation I was describing, as the limiting case where the height of the lighting doesn't matter at all, is like the picture on this page:

Food Production

Note that the reflectors only extend level with the lights, to intercept light that would go up and be wasted. They don't block light to distant parts of the greenhouse floor. The array of lights in that greenhouse is high enough to accommodate all the equipment they use, and the full height they want to grow their palm trees to.

If it made an appreciable difference in energy cost, they could have set it up to be able to lower the lights when the plants are small. But it doesn't. The cost of supplemental lighting for a square mile of agricultural greenhouse is the efficiency of your lights times the price of electricity. Losing light upward lowers the efficiency. Reflecting light straight down, with a reflector that absorbs part of the light, when it could have gone sideways and hit a different plant, also lowers the efficiency. Having the light go through a few extra feet of air doesn't.

No one who's setting up a large-scale operation like that would get their engineering off an internet forum, so it's presumably not a situation that anyone here is facing. But limiting cases are often useful for understanding.

Re: LED Grow Light
 

Very nice, now let's move on from grow lights and talk about germinating banana seeds.

Lights is a very contentious issue and has been the source of many a disagreement (goodness knows why). For some reason some people get very hot under the collar about it (again, goodness knows why) so let's put it to bed and why don't you tell us a bit about yourself. Where are you from? What are you growing?

Re: LED Grow Light
 

I'm growing, um, a spider plant and a pothos. I bought a bunch of seeds, including some bananas, and I found this site when trying to find out how to get them to germinate.

Unfortunately, I know diddly about germinating banana seeds. I discuss it at length in http://www.bananas.org/f16/insights-...tml#post210932, but as I say there, it's idle speculation.

The most striking thing to me about the banana seeds is that they're so large. Not peach-pit large, but still, pretty big. I wouldn't expect something that size to survive in the soil for a long time without something finding a way to eat it, or to make it through most animals' digestion. But it looks as though it makes for very vigorous seedlings when they do germinate.

I've been a stay-at-home dad for several years -- long enough that I should call myself unemployed now instead. My degree is in math, but I've learned a fair amount of biology, since that's what my wife works in. I'm in Boston, in a dinky apartment: the plants are stuck indoors without so much as the prospect of a balcony in the spring.

Re: LED Grow Light
 

There is lots of information on germinating banana seeds, all very helpful, but you have to look for it. Good luck with the seeds.

Re: LED Grow Light
 

Thank you dsws for your excellent write up.
This is the best explanation I have seen on that topic and it would make a wonderful addition to the wiki.

I'm coming back with a large edit but nothing critical, just need to soak this in so I can make a coherent response.

Re: LED Grow Light
 

Edit:

Ignore this. I plugged in some smaller values for the distance, and got negative answers. Odds are, I won't figure out what went wrong until after the holidays.

Original post:

The case of a two-dimensional (rectangular) light source is harder. It's more formula than I can type into a calculator without making mistakes. I think I've got it working right on a spreadsheet. I'm going to be away over the holidays, and probably won't get back to playing with this before I go, so I'll post what I've got.

I put labels in the top row, then in the second row I have the length of the source, width of the source, and distance to the source in cells A2, B2, and C2. Then here's the formula for cell D2:

acos(((-B2^2-C2^2)*A2*B2+A2*C2^2*B2+A2*B2*(-A2^2-C2^2))/sqrt(((-B2^2-C2^2)^2+A2^2*C2^2+A2^2*B2^2)*(A2^2*B2^2+C2^2*B2^2+(A2^2+C2^2)^2) ))-PI()/2

That's for a point under one corner of the rectangle. For a point in the interior of the rectangle, just divide the big rectangle into four little rectangles with their corners over the point. I haven't done that yet.

I've only figured out the formula with some geometric reasoning and looking up a theorem or two: I haven't checked it by doing the calculus. I won't absolutely trust it unless I do so, but at least I've checked on the spreadsheet that it approaches inverse square when the distance is big, and constant when the distance is small compared to both dimensions of the light source. And it varies correctly with the area of the light source when the distance is large. So it should be right.

Re: LED Grow Light
 

Did I miss something? I thought you said it was inverse linear, not inverse square.

Re: LED Grow Light
 

Inverse square is when the light comes from a single point. Inverse linear is when the light source is a line, extending forever in both directions.

When a light source of any shape is very far away (compared to its size), it's effectively a point source. So whatever formula I come up with for a rectangle or a line segment or whatever, one of the things to check is that it gets close to inverse square when the distance gets big. That was the factor of 1.75 versus 1.8 in the 48"-fluorescent example: at only 12" to 18" away, it's already almost inverse square. Farther away than that, you can forget about the complicated stuff.

Inverse square for a point source is also used to figure out the other cases. You can get the amount of light from any source by adding up the amount of light from all the points of the source. But adding up infinitely many things can be very difficult sometimes.

In the case with the diagram, the light source is just part of a line. When you're really close to the line it's almost as though the line went on forever. So it's almost inverse linear in that case. When you're really far away from the line, it's almost as though the line was just a point. So then it's almost inverse square.

Re: LED Grow Light
 

I think the mathematical formula is right, but the spreadsheet coding was wrong. Here's another attempt:

=ACOS(-(B2^3*A2+C2^2*A2*B2+A2^3*B2)/SQRT(((B2^2+C2^2)^2+A2^2*C2^2+A2^2*B2^2)*(A2^2*B2^2+C2^2*B2^2+(A 2^2+C2^2)^2)))-PI()/2

I still haven't checked it with calculus, though.

Re: LED Grow Light
 

You can try LED Grow Lights | Apollo LED Grow Lights

Re: LED Grow Light
 

<<< t5-8-12 are diameters

T12 is dinosaur nuff said
t5 is only HO with high ambient temps generated by an enclosure THIS MEANS NOT BETTER
T8 is the most common and cheapest for bulbs, ballasts, etc
T832W is the bulb part # prefix
CFL is cheap and available but harder to reflectorize
HPS is most efficient and a burst and fire hazard if not enclosed
MH is a waste of time with all the cons of HPS and none of the pros

electricity use is defined in watts
a 150w HPS uses the same amount as 150W of fluorescent
the HPS will give slightly more LUMENS PER WATT (this is how you get a real answer)

everything is nearly half price if you go to a local electrical store, not a box store


LED=hype >>>




Unfortunately much of the above info posted here is highly opinionated and not factual. And the closing "LED=hype" remark is just silly and not even close to reality.