Measuring par T5/Halide VS LED

[SamMule]

Everyone knows the age old debate of T5/Halide VS. LED.
I often hear people say 500PAR of halide/T5 is not the same as 500PAR with LEDs, and I have to say, from my limited experience, this seems to be true.
In one tank, I can go well over 500PAR with T5's and the acros love it. However, in the other tank, 400 will torch the same corals.
No, not all things are equal between the tanks, so it is certainly not an apples to apples comparison. However, this got me thinking. What if the problem is the measuring device?

Stay with me here... and correct me if I'm wrong, because I don't KNOW all of this. Just making conjectures from my knowledge of electronics.

LEDs are digital on/off. DC voltage driven. No in-between. 1 second of 430nm light emitted is 100% on time. Or duty cycle, if you will. Square wave.

T5's and LEDs are A/C driven correct? Therefore, the light must cycle on/off multiple times per second emitting light at a varying intensity throughout the cycle in a sine-wave type pattern.

Our PAR meters (also digital, i assume,) must have a set sample rate at which it measures micromoles/second.
What if the sample rate on our PAR meter is low enough that when it samples light from an AC source that it just averages the result?
It seems this would skew the results on the low side, but I dont know how PAR meters work...

Kind of like measuring voltage of a pulse-width modulated circuit with a voltmeter. Sample rate is too slow, so it spits out an average. (I know, not apples to apples, because PWM is DC, but you get the idea!

Again, I don't know how these things work... just asking questions!

 

[Spare time]

I think the reason why the par isn't equal is that par is only part of the equation. Spectrum is equally important and a light that has heavy blue output is going to cause photosaturation with relatively less "par" than a light with that intensity spread out more or less optimized at certain wavelengths. There is some data to support this in a MH vs LED article Dana did on here.

 

[SamMule]

Yeah, I read that one awhile back.

 

[jda]

PAR meters that we use only measure visible waves. 400-700nm. There are more waves than this that are helpful to corals, plants and anything else that is not human eyes. Most LEDs are from 400-700nm and were made to pop a PAR meter with doing well for corals being a secondary issue - this was their watt to par ratio that it total BS. Mercury based lights do have waves outside of 400-700nm and some have energy from 350-850nm. The ranges from 350-400nm and 700-850nm will have un measured benefit if you use a reefing type of PAR meter - a better meter will capture these spectrum. If you have 500 PAR from a typical LED and 500 PAR from a MH, you can bet that they MH will have more usable light than the LED does.

People have erroneously just focused on optimum excitement wavelengths of some proteins and stuff. While they might exite the most at X spectrum, all proteins can still take energy from other wavelengths and some even NEED other waves for energy transfer, color and just accure pure energy. For example, chlorophyl A excites from about 350-450nm, but also again at about 650-750nm. Peridenin needs as much at 600 as it gets near 400. You really need all of this for this to be efficient - blue alone will not give you the most benefit like many LED companies and honks want you to believe.

This is an oversimplified graph, but not a single one of these proteins are used efficiently with just blue... all of them could use some UVA, violet and green with most wanting some yellow and red. Notice how all of them really want to use the super high energy in the UVA range which really only MH and some true/super actinic T5/T8/T12 bulbs provide (there are a slight few LEDs with real UVA light).

AC vs DC does not matter. Hz or PWM can be a thing, but I think that people stopped caring about PWM since nobody every figured out anything about it - it is just an unanswered question. An ignited gas stays ignited even when the current pulses. It appears that most decent Hobby PAR meters can get nearly to the same places as a very expensive one - I have an Apogee 510 and it is close enough to a $4500 LiCor (which samples very well) to believe that it does not have an issue with Hz or PWM. There are some indications that the things that use the light might care and not like PWM... again, a tabled discussion for now.

For me, it seems that if you give a more full spectrum, you can go higher PAR with LEDs. I do believe that things like Emerson effect in which things need red and far red to move energy between the photosystems are indeed true. Also, if you raise the lights up and allow the colors to blend better, this helps too. The two best coral growers that I know under LEDs use nearly all channels on their LEDs and they keep them up high.

Lastly, not all diodes are the same. When LEDs first came out, and the next few generations, there were some white diodes that torched corals - they were horrible. While nobody uses these anymore, I do think that some of the diodes that you see being used for high PAR are just better at it than others. I am not a huge LED dude, but XP-E 5500k that Photon V2 uses are likely better than the "cool white" and "warm white" that Radion since most people can crank the Photon up higher. I wish that more manufacturers would let us know exactly what kind of diodes that they use so that we can study this - they might and I just don't know since I don't care all that much.

 

[SamMule]

Very informative, @jda. Thank you!

 

[jda]

LEDs are digital on/off. DC voltage driven. No in-between. 1 second of 430nm light emitted is 100% on time. Or duty cycle, if you will. Square wave.

LEDs dim through PWM - pulse width modulation. On-off-on-off-on-off very quickly since they cannot lower voltage or current.

They dim by turning them on/off a bunch really quickly so that human eyes cannot tell. It is the hubris of man to think that other things cannot tell either. Again, this was a thing years ago for corals and now nobody cares. In some industries, like terrestrial plants and the hippie lettuce, PWM has been shown to harm chlorophyll - these are studied more than corals are. These are industries who have pulled back from LED significantly lately.

Units being run at/near 100% over corals could also be doing better because of less PWM. Who knows?

 

[SamMule]

I wondered weather it was PWM or voltage regulation.
If LEDs are PWM as you say, and the light emitted from gas burn does not fluctuate, then I can certainly see how measurements with a PAR meter could be skewed.
This would suggest that the LED/Halide (gas) light sources provide a more "steady" source of light than the LED.
Example:
Gas light emits 400par at 100%duty =400PAR
LED emits 1000 PAR at 40% duty =400PAR

Since we measure in light intensity over time, these will both be returned as the same, but are they?


This, to me suggests that the further we get from 100% duty on the LEDs, the more misleading the results will be.

If you try to drink from a garden hose that is running for 10 seconds at 10PSI, you could probably get it done just fine.

However, if i take that same garden hose and run it for 1 second at 100psi, you're... not gonna have a good time.

In both scenarios, the same volume of water was supplied, but only one of them is really usable.

Our PAR meters are more like a bucket.
Both scenarios will end up with the same amount of water in them.

Now, again, I do not know how the biological process of absorbing light works, nor weather the time scales we are talking about here even matter, (gotta be 60hz or more if its too fast for our eyes to see) but it's a thought!

 

[jda]

You could ask LiCor and Apogee how they handle this. This is probably something that Oreo would know or have a handful of links ready.

There is evidence that the light intensity is not biologically the same for many things. There is really nothing on coral. I don't really care what the meter says over what living things say. The nearest comparables say that it matters. Some lighting honks will tell you that you cannot prove that it matters, but the real approach should be to prove that the most similar photosynthetic things that we have where it does matter do not apply to corals. For me, I will just say that I do not know - seen plenty of corals that don't seem to mind at all whereas other certainly do.

You can see the same things with DLI- direct light interval. Within reason, this does not seem to matter. 10 hours at 400 PAR is about the same as 9 hours at 444 or 8 hours at 500 PAR. 1 hour at 4000 does not work nor does 24 hours at 167. The theory does not have a "within reason" part of it, but it does seem to matter.

There are more parallels between coral and plants than people want to admit. I live in Colorado and grow a few banana plants. In the summer, they get all-day full sun - almost unlimited. They thrive. In the winter, they are in my basement under a 5500k MH where they adapt to about 1/10th the light and also thrive, just not as well growing about half as fast. Most corals that can live on 1/10th the light in our tanks also thrive like crazy under full sun. People call these low light corals, but they truly are not... they are quite happy to get lots of light too.

 

[KenRexford]

Interesting discussion. If I am thinking this through correctly, this seems to run contrary to one aspect of what I think is currently the theory on lighting. If PWM is potentially problematic, then the best led lighting should have fewer diodes at maximum intensity rather than more diodes at lower intensity. The former would tend to avoid wild fluctuations; the latter would not. The former might have, say, 600 par spikes modulated to 400 average, but the latter 1200 par spikes modulated to 400 averages. This makes me happy with my T5 hybrid with three AI Prime 16’s rather then more led units run at 50% power for “spread.” It seems like spread via leds would be bad unless the diodes were weaker.

 

[SamMule]

Interesting discussion. If I am thinking this through correctly, this seems to run contrary to one aspect of what I think is currently the theory on lighting. If PWM is potentially problematic, then the best led lighting should have fewer diodes at maximum intensity rather than more diodes at lower intensity. The former would tend to avoid wild fluctuations; the latter would not. The former might have, say, 600 par spikes modulated to 400 average, but the latter 1200 par spikes modulated to 400 averages. This makes me happy with my T5 hybrid with three AI Prime 16’s rather then more led units run at 50% power for “spread.” It seems like spread via leds would be bad unless the diodes were weaker.

I think the idea with using more small diodes over a few large ones is to attain better spectrum blending.

 

[oreo54]

Everyone knows the age old debate of T5/Halide VS. LED


Again, I don't know how these things work... just asking questions!

It is probably not the measuring device with one exception which I'll get to

First it samples at a specific sample rate... say 1000 times/sec.

The PPFD unit is measured in micromoles (one micromole equals 62 quadrillion photons) per square meter per second (µmol/m2/s).

Soo how does it do this.. sample once and multiply by 1000 or sample 1000 times and add?

All of the light types flicker to some degree or another. Magnetic ballasts are like 60-100Hz. Currently mostly only mh. Magnetic t whatevr ballasts are def uncommon in the aquarium lighting.
I do have some old t8's running starters and magnetics though.



Electronic ballasts have their own flicker rate.
Led own rates generally start ( or should start) 500Hz. Some drivers can go into the MHz range. This range is also the area of Dan's, motors, and switching power supplies and ballasts.

The arc on mh lamps can also fluctuate while " burning" creating another type of flicker.

Ocean " flickers" as well.. Caustic lines.

As we have seen, lensing effect of passing waves creates ‘glitter lines’ and the instantaneous amount of light can be highly variable at a depth of two inches in the tide pool.

If I remember correctly this lensing can be in orders of magnitude.

As to other studies ..
I'm sure there are various ones .

Light use efficiency was the highest under irradiation with 1 kHz pulsed LEDs. In conclusion, pulsed LEDs with 75% duty ratio and low frequencies did not show significant inhibition on plant growth, suggesting that pulsed LED irradiation technology has a potential to save energy consumption for producing crops in plant factories.

Now to a sort of exception. Good sensors take in all photons in a close to 180 degree arc ( cosine corrected).
So a laser or a flood light could generate the same # of photons yet be a VERY different distribution.
Let's just say it could be a lensing issue making equal par not so equal spatially.

Food for thought

Last is the Emerson enhancement effect.
Mostly seems to apply at high par levels and

Yes mh's have high IR levels.
The useful cutoff nm level seems to be around 800-ish. So say 700-820 is needed to protect the photo system from too high of par levels.
LEDs ( nor t5's have this afaict though it's relatively hard to get THAT high of par levels with t5 alone) have any abundant >700nm photons.
White led vs 10000k mh

As far as the heat IR, UV (a or b or both) and any suspected nm balance or ratios ????
Or like individual colors from discreet leds?

Soo what is it??
Don't know. Personally I suspect the lenses and possible lack of >700nm diodes at that high of PAR.

Then again.. Don't see any MH's here..

 

[SamMule]

It is probably not the measuring device with one exception which I'll get to

First it samples at a specific sample rate... say 1000 times/sec.


Soo how does it do this.. sample once and multiply by 1000 or sample 1000 times and add?

All of the light types flicker to some degree or another. Magnetic ballasts are like 60-100Hz. Currently mostly only mh. Magnetic t whatevr ballasts are def uncommon in the aquarium lighting.
I do have some old t8's running starters and magnetics though.



Electronic ballasts have their own flicker rate.
Led own rates generally start ( or should start) 500Hz. Some drivers can go into the MHz range. This range is also the area of Dan's, motors, and switching power supplies and ballasts.

The arc on mh lamps can also fluctuate while " burning" creating another type of flicker.

Ocean " flickers" as well.. Caustic lines.

If I remember correctly this lensing can be in orders of magnitude.

As to other studies ..
I'm sure there are various ones .

Now to a sort of exception. Good sensors take in all photons in a close to 180 degree arc ( cosine corrected).
So a laser or a flood light could generate the same # of photons yet be a VERY different distribution.
Let's just say it could be a lensing issue making equal par not so equal spatially.

Food for thought

Last is the Emerson enhancement effect.
Mostly seems to apply at high par levels and

Yes mh's have high IR levels.
The useful cutoff nm level seems to be around 800-ish. So say 700-820 is needed to protect the photo system from too high of par levels.
LEDs ( nor t5's have this afaict though it's relatively hard to get THAT high of par levels with t5 alone) have any abundant >700nm photons.
White led vs 10000k mh

As far as the heat IR, UV (a or b or both) and any suspected nm balance or ratios ????
Or like individual colors from discreet leds?

Soo what is it??
Don't know. Personally I suspect the lenses and possible lack of >700nm diodes at that high of PAR.

Then again.. Don't see any MH's here..

Excellent post! Very informative! Thank you! I suppose that puts my whole hypothesis to bed!
I see what you mean about the UV and IR spectrum. The halide does have a much stronger representation there.

 

[A. grandis]

You could ask LiCor and Apogee how they handle this. This is probably something that Oreo would know or have a handful of links ready.

There is evidence that the light intensity is not biologically the same for many things. There is really nothing on coral. I don't really care what the meter says over what living things say. The nearest comparables say that it matters. Some lighting honks will tell you that you cannot prove that it matters, but the real approach should be to prove that the most similar photosynthetic things that we have where it does matter do not apply to corals. For me, I will just say that I do not know - seen plenty of corals that don't seem to mind at all whereas other certainly do.

You can see the same things with DLI- direct light interval. Within reason, this does not seem to matter. 10 hours at 400 PAR is about the same as 9 hours at 444 or 8 hours at 500 PAR. 1 hour at 4000 does not work nor does 24 hours at 167. The theory does not have a "within reason" part of it, but it does seem to matter.

There are more parallels between coral and plants than people want to admit. I live in Colorado and grow a few banana plants. In the summer, they get all-day full sun - almost unlimited. They thrive. In the winter, they are in my basement under a 5500k MH where they adapt to about 1/10th the light and also thrive, just not as well growing about half as fast. Most corals that can live on 1/10th the light in our tanks also thrive like crazy under full sun. People call these low light corals, but they truly are not... they are quite happy to get lots of light too.

DLI = Daily Light Integral.

Adding to your nice writing... Yes, most of the so called "low light corals" are just corals that CAN adapt to a lower light intensity situation. Each species do have their adaptation ranges and differ in tolerances, but with proper care most will do great under the so called "high PAR values" people like to mention! People need to understand that comparing to natural sunlight, those "high PAR values" aren't that high at all!! Another classic example is when people say that "zoas are low light corals". They aren't corals, by the way, much less low light organisms. The great majority of zoanthids are from very shallow reef zones exposed to very high PAR levels from direct sunlight. The majority of corals in the hobby come from shallow waters too!

 

[A. grandis]

Excellent post! Very informative! Thank you! I suppose that puts my whole hypothesis to bed!
I see what you mean about the UV and IR spectrum. The halide does have a much stronger representation there.

Halides have the representation of real UV and IR there.
Halides also have "stronger representation" in all other wavelengths in between the 400-700, by nature. It's important to know that a 20K lamp has "less white" than a 14K lamp, for example, but they both cover the full spectrum in their own way. The representation of all wavelengths in the spectrum from halides is strong in intensity too. The spectrum comes from one source (arc), like the sun, not from many different diodes trying to blend. The presence of different diodes in order to achieve a visually appealing spectrum to the human eye has it's limits of coverage due to distance between diodes, quantity, lenses, types/quality, "colors"..., in comparison to the metal halide's single arc discharge tube that uses reflectors to distribute the light homogeneously with the correct application.

Just one note here... it's important to remember that the "flickering" we see in the ocean, which is not a flickering, but the glitter lines, has absolutely nothing to do with the different types of real flickering from any type of artificial lamps in any way. Caustic lines work completely different than any of the types of flickering problems we have from artificial lamps due to any intervention of malfunction of magnetic/electronic ballasts, or the normal pulse-width modulation from most LEDs we know about. The same real caustic lines represented by metal halide lighting in our tanks also have nothing to do with flickering problems from any artificial lighting. The caustic lines from LEDs often show a very different and artificial structure of the caustic pattern due to the normal characteristics of diodes.

 

[MartinM]

All of this is interesting, but from my 3 decades of hobby experience, all light seems to do equally well if coverage is equal (assuming you're covering the full spectrum, I don't run the American style 'any color you want, as long as it's blue'). IME the better results come from better coverage rather than the type of light source itself.

From my commercial experience, commercial aquaculture uses sunlight and NSW flow through systems (no, these indoor places are not commercial scale ventures, they're small to medium importers and/or wholesalers that take colonies, cut them up and sell the frags, and they run LEDs because they receive them for free - otherwise they'd be in greenhouses because nothing beats sunlight for both efficacy and cost efficiency). And since they get their LEDs for free and they're focused on growth, these small/medium operations who do grow out colonies and frag them use 2-3x as many fixtures as a regular hobbyist would - again, to achieve the best possible coverage. The only large scale commercial aquaculture I'm aware of using LEDs is in Taiwan, and they exclusively culture flowerpots (G. stokeskii) for hobby and pharmaceutical purposes. This is a well-researched coral that only utilizes a narrow section of light (violet and blue) and that doesn't utilize more than 70 PAR, so LEDs make sense and are also economically viable in their case. (I'm defining "commercial" as exclusive propagation of 10,000+ colonies per year, not frags or importing)

MH is my personal fav for aesthetics, but I use exclusively LEDs now for the convenience. However, I don't use a single 'point source' LED fixture, they're all high spread (Straton, AI Blade) that provide an even coverage. Of course T5's get you that also, but I don't use them because I always keep a lot of clams, and I don't want to block my top down view!