Updating the WPG rule (theory)

[jsoong]

What we REALLY want is a Mearsurement of Photosynthetic Flux- That is, the mearsure of Radiant Energy that is usefull to plants- Lumens for Leaves instead of eyes, if you will. I haven't a clue where to look for it- Maybe Hydroponic literature?

Way back in page one of this thread:
http://www.aquabotanic.com/lightcompare.htm
The PAR (photosysthetic active radiation) mentioned is what you are looking for. This measurment makes the most sense to me. After all, we want to measure lumens the plants will use, not what we can see. Unfortunately, this site only gives a small sample of bulbs. I don't know where one can find PAR's for all the bulbs in your database short of measuring each bulb's spectrum....

I've been following this thread to make sure my planned light upgrade does not get into the high light / CO2 territory. I am as confused as ever. If I use the PAR method, it appears I need 8-10x32 T8's to get me close to high light .... but that would be like 6-7 Tong WPG .... now all I need to do is decide which set of criteria to use ....

 

[czcz]

Now if someone can work out an excel spreadsheet to combine the 2 formulas (what light we are actually getting and whats actually needed).

You mean something like Figure 5 of " Lighting as a Function of Tank Size," where the author combines Erik Olsen's formula on the Krib with the analysis of Takashi Amano's tanks?

Others are surely more qualified (and I hope they post), but I think there's at least a couple reasons why this is problematic.
1) These formulas built around Lumens focus on the efficiency of different bulb types. Without knowing Amano's bulb type a comparison is difficult.

2) Erik Olsen's formula estimates surface area by taking volume^(2/3), and fitchfamily building off that really isn't so far removed from lumens/sq in. In this sense Olsen and Ozz's ideas seem similar.

3) We can't use Tong's "Effective wpg" because its not a measure of how many actual watts one needs, but rather how many wpg a tank will behave like, regardless of tank size. Instead we must use something like "Equivalent T12 watts" (Lumens/(58.9 lumens/watt), then compare with Amano's watts (see below).

We may be able to compare directly to lumens/sq in somehow, but it has not occured to me how.

4) The survey data here has folks with a wide range of lighting types. To compare to Amano we likely have to isolate the high light tanks. The questions then are A) Is it a safe assumption to believe Amano is always targeting high light? (I'm not so sure it is) and B) Do we include what Ozz considers "Extremely high light" -- the guys who run insane lighting according to many standards?

In the end, a wise hobbyist would probably plug their tank info into as many modern formulas as possible and then take a best guess from there, like jsoong is doing.

You did raise my interest though, and I'm no expert on statistical analysis, but here's some stuff.

First I removed all samples below 17 lumens/sq in (arbitrary choice) to isolate "High Light" tanks. I then used the fitchfamily formula to calculate watts. Since I do not know Amano's bulb type, I also used lumens for each given tank to find "Equivalent CF watts," "Equivalent T12 watts," and "Equivalent T8 watts," then did a regression for all three individually, as shown in the attached graph.

If we omit ID 2, 4, 8, 28, 34, 43, and 58, the resulting plot is much closer to fitchfamily's. However if we remove those samples we're left with only 11 tanks to play with, and then its a question if thats an acceptable sample size for the comparison. Still interesting though.

As earlier this doesn't take into account the lower lumens/watt of short bulbs (fitchfamily does not make such an adjustment either). Just more stuff to look at.

jsoong, fwiw I think 8 32w T8 over a 70gal has gotta be very high light, no matter what the standard.

edited for units and typos

 

[travis simonson]

Dang Joe *dreams of lumens spinning in my head*

That statistical analysis goes well beyond me. Nice work man

 

[Plantbrain]

Dang Joe *dreams of lumens spinning in my head*

That statistical analysis goes well beyond me. Nice work man

Yes, lots of work, but to what end? What is really gained here in practical application?

You have a number of issue to overcome here with this and it's far from a simple method.

While many have poo pooed on the watt/gal rule, I've never found it to fail after keeping plants as the main focus for 15 years now.

Not once.................

Most of the issues revolve around the aquarist not having enough CO2/nutrients, more light can help in some cases where these are limiting, but generally not.

I've had deep tanks, I've used normal T12's, T5's, PC, MH, MV and probably more different types of lighting than most anyone here.

Incadescent light is another matter, but MH and FL's hold up very well, there is no plant I cannot grow in almost any reasonable sized tank, even some of the deeper ones.

I've grown nice Gloss at 24-28" deep tank with 1.5-2.1 w/gal of normal FL's with a reflector.

Many like to claim it needs high light etc, this is not true.

While all the analysis is interesting, and I've seen folks try and revamp the w/gal rule and make all sorts of claims about it, most of these same folks should put their efforts into CO2/Nutrients and they will no longer have such issues with lighting.

Here's why:

Plants grow do not on light alone.

Now read that again and 5 more times.
Now think about it.

Why might that statement greatly influence what you may think about lighting when I find it(w/Gal) so flexible?

Because plants require light and CO2 and nutrients to grow well.

If the CO2 and nutrrients are in good shape, then the plant can fully maximize the light available.

This has been shown in a number of studies relating CO2 use and light as well as nutrient availability and location.

So if you have not mastered CO2.nutrients very well, and have been able to definitively rule these oyut as potentially limiting factors, then the light issue is confounded by the other interactions like nutrients/CO2.

Given that CO2 is nearly 95% or more of all algae related issues I field, rather than light, the likelyhood of these other issues being overlooked with respect to light is extremely high.

Fun theory, don't get me wrong, but there are other issues going on here that have not been addressed well.

When folks haggle over lighting and other similar issues, they need to be able to rule out these other confounding factors.

You need to be able to fully master each thing/verify it, that influences growth before you can go it and test something adequately.

Most intermediate and newbies don't have a handle on that yet so they run from one topic to the next without learning step wise in search of that silver bullet, therein lies the danger. They get confused and spend time on other things that are often not nearly as relevant.

Neat spreadsheet though.

Regards,
Tom Barr

 

[czcz]

Thanks.

Yes, lots of work, but to what end? What is really gained here in practical application?

In the sense of the above graph not much except eye candy. We can say the fitchfamily analysis of Amano's tanks are much like your tanks and above advice: he doesn't care about high and very high light tanks so much.

In terms of updating the wpg rule, I think the practical application is best in small tanks, large tanks (wpg from lumens/sq in would argue that your using 1.5-2wpg on big tanks and growing glosso low makes sense), helping newbies/intermediate aquarists find the point where CO2 is needed/how much to upgrade their lighting/debating T8 vs CF vs T5 vs etc, and so on.

I agree light is less important than nutrients and CO2, and after messing with my tanks personally more subscribe to the idea of thresholds. Once one is in "high light," more light doesn't really matter so much. Once one is in "med light." they can probably grow "high light plants" if they just focus on nutrients and CO2. We see this with tanks around the forum and with some survey entries and plants grown, for example.

The data analysis is a lot of work, especially for those of us who aren't number crunchers by trade, but a good way to be a nerd at the moment I think. Nutrients and CO2 are easy (because of your work, EI, Chuck's calc, collective experience, and so on) and thinking of them in terms of thresholds is relatively easy. Lighting is expensive and a little analysis is at the least interesting.

The complication of above will hopefully be minimized by the developing calculators. We have at least three working models now. Since we are using databases and such, its much easier to manipulate the real-life tank samples whenever someone has a new idea. Personally I want this data to spread around and stay open so anyone can do the analysis with true and theoretical samples a little more easily, The code for the calcs so far are open. (Thanks Ozz.)

While many have poo pooed on the watt/gal rule, I've never found it to fail after keeping plants as the main focus for 15 years now.

Even on your nanos and small tanks?

The 20gal example at ~5.5wpg for EI is a good one to look at. If one ran this wpg over, say, a 125 gal, your dosing recommendation would increase the P and N to the higher ranges to avoid them bottoming out, no? If they lit with 5.5wpg of MH, wouldn't they be going way overborad?

Thanks again. Much appreciated.

 

[Tong]

While many have poo pooed on the watt/gal rule, I've never found it to fail after keeping plants as the main focus for 15 years now.

Never said it was a bad rule. It works well for most situations and is the simplest approach available. However, as czcz mentioned, many are finding it inadequate for nano tanks.

Take for example a 5.5 gallon with a 16" 14 watt bulb. Strictly using the traditional WPG rule, this is 2.5 WPG. Could you grow a med-high light plant in this? Yes, probably, if you did your CO2 and nutrients right. Would it be a lot easier in a 55 gallon with 2.5 WPG (using traditional rule) with exactly the same CO2, nutrients, etc...? I think so.

Our point is not to tell someone "Hey, you only have x WPG, you cannot grow plant y." But rather it is to point out that 2.5 WPG on a 55 gallon is different from 2.5 WPG on a 5.5 gallon and explain why it may be different, all else being equal...or however that latin phrase goes.

 

[czcz]

So I have some free time this weekend, and if we can get some numbers for different bulb types I'll put together a calc specifically for shorter bulbs for your consideration.

What's the bulb length cut-off point?
What's the volume cut-off point?
(Vote 24" and 20gallons, respectively, to start, then can narrow it down later by comparing data if needed.)

I'll PM Ozz in case he hasn't checked the forums lately.

To Do list. Please adjust or add to this list as you see fit.
1) Samples for CF, T12, and any other bulb type we're interested in below the agreed upon length.

We need length and rated watts and lumens, and it will be easiest to sort the data if the data is in that order. Please also link sources.

(Any disagreement with the T8 data above?)

2) Adjustable Calcs for above. Adjustment of existing survey data.

Can run the data first and maybe we can continue to narrow down where bulb types break.

3) Get more real-life tank info. http://woo.gotdns.com/Aquarium/AddData.php

--

I hope our experts and gurus continue to pick this apart and provide input btw. I wasn't asking questions to be a <beep>

 

[Wizzard~Of~Ozz]

It is interesting what you say about the length of the bulb becoming important.. I would have to look through spec sheets to find out more but I think this would have to do with the ballast construction, also the bulb being shorter it may be able to handle more current / inch then the long bulbs.. It is an interesting point none-the-less. I'll see if I can dig up any information on bulbs output relative to length for the exact same bulb, perhaps it will show more of the efficiency/watt for a shorter bulb.

 

[Floyd R Turbo]

Lighting upgrade

Well, it's been a few years since this thread was busy. I just got done reading through it, and now my brain hurts. Thanks. I am an engineer and I used to work in commercial electrical design, so I HAD to read all of it, so it's partially my fault.

So maybe someone can tell me if I've got this right. Here was my initial setup:

55g freshwater, 3" substrate, all-glass hood w/dual 24" fixtures for 18" bulbs.

I had 2 x 18" 15W Tropic Sun full spectrum T8 5500K lamps (1 in each) giving me a total of 30W of T8 light.

I have 576 sq in of surface area. I want to get to right around the moderate light level, 10 lumens/sq in according to 'the article'. So I need about 6000 lumens, just to make the math easy.

I have purchased a 4' fixture from Menards that can take T12s or T8s. I've just now come across this thread and figured out that I probably should have just popped for the T8 fixture for 3x the price from the LFS. But oh well, I don't have the $ anyways so I'll do with my choice.

Also at Menards, I got 2 Verilux Instant Sun trichromatic F40T12 bulbs:
6280K
lumens not listed
color rendering index (cri) 94.5

So if these go by the standard 59 Lumens/watt, I've got 80W total output which gives 4720 lumens. This gives me 8.2 lumens/sq in.

Switching up to the 4' F32T8s in the new fixture yields 82.8 l/w x 64w = 5300 l, or 9.2 l/sq in

I think with the old fixtures, which had 2 x 18" F15T8s, 82.8 l/w x 30w = 2484 l, or 4.3 l/sq in

Using the old WPG method,

1) Old 18" fixtures: 30W/55g = .54 WPG
2) New 4' w/T12s: 80W/55g = 1.45 WPG
3) New 4' w/T8s: 64W/55g = 1.16 WPG

Adjust those for efficieny, the T8s x 141%

1) 0.54 x 1.41 = 0.76 "WPG"
2) 1.45 x 1.0 = 1.45 WPG
3) 1.16 x 1.41 = 1.64 "WPG"

I think I've got a pretty good grasp of this...don't I? It makes a LOT more sense than the WPG rule.

Forgive me if I missed it, but what about CRI and Temp? I'm assuming that a 5000K and 6500K bulb are NOT the same, can someone explain the difference to me as far as it relates to plant growth/etc?

 

[Purrbox]

The Kelvin Rating only describes how are eyes see the light and has no bearing on whether or not the bulb has peaks at the right points for the best plant growth.

 

[Floyd R Turbo]

The Kelvin Rating only describes how are eyes see the light and has no bearing on whether or not the bulb has peaks at the right points for the best plant growth.

Thanks for the info!!

I guess that leaves me with another question though, is there a proper K rating for plants, or is it only important to get a full spectrum trichromatic bulb?

 

[Cliffz]

wooo
So my 55gal with 2x65w (has 4 bulbs but only have 2 turned on) CF (coralife) would be 1.75wx65=113.75x2(bulbs)=227.50w/55=4.4wpg?

 

[Floyd R Turbo]

I have 576 sq in of surface area. I want to get to right around the moderate light level, 10 lumens/sq in according to 'the article'. So I need about 6000 lumens, just to make the math easy.

Switching up to the 4' F32T8s in the new fixture yields 82.8 l/w x 64w = 5300 l, or 9.2 l/sq in

The bulb I just found are Philiips Natural Sunshine full spectrum listed at 2950 lumens, which makes it 92.1 l/w, so now I have 5900 l over 576 sq in, or 10.24 l/sq in. I reached my goal. Yay!

 

[Thomas.w]

now my brain hurts

so now that you guys have that all figured out how would it apply to LED lights say 32x3 watt LEDs = 96 watts and is 4320 lumens at 6500k and the tank dimensions are 90gal. 48"l X 18"W X 24"D (24" being depth of water). I am looking at switching to LEDs and currently have 2 40 watt F40T12 48" NO Fluorescent bulbs and believe they are 2150 lumens each for a total of 4300 lumens and .88 wpg. The LEDs going by rule of wpg should be 1.06 wpg
I am growing low light plants Crypts, Swords, Anubias, Java fern, Val's. all growing well except Val's they seem to be growing very slow but are growing under my current lights. So I was wondering if the LEDs would be a good upgrade and saw this thread but no info on LEDs. Are there future plans to include the new technologies into your new lighting rule? Just another monkey wrench thrown in lol

 

[Floyd R Turbo]

monkey wrench!

 

[aqua_chem]

Honestly, LEDs are the only place where using a watt as a measurement of light is actually semi-relevant, and even then it's thoroughly confounded by optics, spectrum, tank depths, etc.


I would advise you to ignore anyone pedaling a theory that involves watts and/or doesn't use empirical data, ie a par meter (not lumens), to formulate it. It's 90% bad math and bad science. I would recommend seeing what other people are using to good success and try to emulate that. The good news is that quality LEDs can be turned down if you overshoot your goal or have issues, and most fixtures can be elevated or screened to reduce excess light, so you don't need to be perfect about it, just close.


Oh, and you posted in a thread that was 4 years old. Start a new thread next time so we can better address your question.