Figuring PPF of Samsung F Series Build

[Buddhafamily456]

Hey everyone I’m new here but I want to thank you guys for all the knowledge you’ve posted on the topic of DIY LEDs. I’m trying to put together an LED build using Samsung F Series strips and COBs for Far Red/Red/and Blue Spectrums.

I’ve been doing a lot of reading and watching a lot of the videos for Grow Mau5 and I understand you will have to find out the PPF in order to find out the PPFD of your grow area with the light.
I seem to be having a hard time on how to find out my PPF, and was wondering if anyone could maybe show me the equation necessary to figure this out. Thanks a lot

 

[pirg420]

If you have a par meter you could take a measurement at the center of each square foot and figure out a pretty good estimate of the ppfd.

 

[Buddhafamily456]

If you have a par meter you could take a measurement at the center of each square foot and figure out a pretty good estimate of the ppfd.

I thought there was some kind of equation

I saw in a video, I think it was posted by GrowMau5.. he used the micromoles per joule of a single ended HPS to figure out PPFD.. he used 1,000(the fixture wattage) x 1.3(micromoles per joule of fixture) and figured that out to be 1300. Saying the fixtures PPF was 1300 .. how could I find the PPF of a fixture I am building using the Samsung F Series strips? I see on the info of the strips it gives a lm/w number and I’m guessing this somehow correlates to the micromoles per joule measurement ... also if I were to be using multiple strips how could this be measured?..

 

[wietefras]

You can divide the lumen values by a conversion lumen to par factor. or multiply by "1/factor".

For that you need to know the "spectral power distribution" (SPD) of the light though. For a 3000K 80 CRI Samsung this is 69. For a 3500K 80 CRI it's 71. Every spectrum has it's own conversion factor to correct the lumen curve to the PAR "curve". Usually it's somewhere around 70 for warm white light.

So say 7000lumen would then translate to 100PPF.

Or similarly, 140lm/W would be 2umol/s/W. Then multiply that 2umol/s/W by the number of watts to get the total PPF.

Once you start adding mono's it becomes a bit more difficult, but technically you could add the charts and create your own SPD. Or just go with the PPF of the strips since those mono's usually don't have much impact anyway.

 

[Buddhafamily456]

Thank you for the help.
I believe I am following what you said up to a point.
So you are saying to divide the lumens to par factor, and the given PAR Factor for example of a 3000k strip would be 69. The lumens provided on the info sheet for a 3000k strip at 1120 mA would be 4335 lumens.
So for example:
4335/69 = 62.8 PPF
Not sure if I have that right, but that’s what I’m understanding. ?

Now I seem to be getting lost on the conversion you are giving for lm/w to u/mol/s/W.
Where am I finding the value of u/mol/s/W to divide by my lm/w?
Or is the standard given value for lm/w to u/mol/s/W just 140 lm/w = 2 u/mol/s/W?
- and from there I have to find my proportionate ratio using that compared to the given lm/w of a 3000k f series strip at 1120 mA being 168 lm/w and the u/mol/s/W being the variable?

You can divide the lumen values by a conversion lumen to par factor. or multiply by "1/factor".

For that you need to know the "spectral power distribution" (SPD) of the light though. For a 3000K 80 CRI Samsung this is 69. For a 3500K 80 CRI it's 71. Every spectrum has it's own conversion factor to correct the lumen curve to the PAR "curve". Usually it's somewhere around 70 for warm white light.

So say 7000lumen would then translate to 100PPF.

Or similarly, 140lm/W would be 2umol/s/W. Then multiply that 2umol/s/W by the number of watts to get the total PPF.

Once you start adding mono's it becomes a bit more difficult, but technically you could add the charts and create your own SPD. Or just go with the PPF of the strips since those mono's usually don't have much impact anyway.

 

[wietefras]

n PAR Factor for example of a 3000k strip would be 69. The lumens provided on the info sheet for a 3000k strip at 1120 mA would be 4335 lumens.
So for example:
4335/69 = 62.8 PPF
Not sure if I have that right, but that’s what I’m understanding. ?

Correct

Now I seem to be getting lost on the conversion you are giving for lm/w to u/mol/s/W.

That's the same conversion factor. I divided both lumen/lux numbers by 70 (as an example).

The 3000k f series strip at 1120 mA being 168 lm/w would give you 2.4umol/s/W (divided by 69)

Clearly this is not super accurate, but then none of it ever will be. PAR meters aren't that accurate either and the datasheets usually have a notice that the quoted numbers can be 7% to 10% off (plus or minus).

Still, it's an indication and that's all you would normally need anyway.

 

[Aolelon]

Here is the number for 3000k samsung lm561c s6 80 cri
Samsung 561C 3000K 80 CRI 0.014505
Multiply lumens by the 0.014505 to get your umol/j / umol/s
According to this thread here
https://www.rollitup.org/t/par-multiplier-thread.928907/#post-13194602

 

[Aolelon]

Of course wietfras method works just as well also

 

[Buddhafamily456]

Of course wietfras method works just as well also

Wow thank you so much.

 

[Buddhafamily456]

Again, thanks so much for taking the time to help me.

So for the first equation you gave me. The lumens/par factor .. the PPF number that you get is indicative of what?
And by that I mean my only understanding for getting a PPF value is from the YouTube video I saw from GrowMau5 where he figures it out using u/mol/sW x wattage(which I’m still not clear about, whether it is PAR Watts or total driver output?)

Also for the second equation I understand and thank you for clarifying it for me, however to find out the total PPF of the fixture am I multiplying the mol/s/W x PAR Watts or total output of the driver.
Also what I can’t fully wrap my head around is, how would the PPF value you find be the same for 1 strip as 10 strips or any other value for strips at that matter?

Correct

That's the same conversion factor. I divided both lumen/lux numbers by 70 (as an example).

The 3000k f series strip at 1120 mA being 168 lm/w would give you 2.4umol/s/W (divided by 69)

Clearly this is not super accurate, but then none of it ever will be. PAR meters aren't that accurate either and the datasheets usually have a notice that the quoted numbers can be 7% to 10% off (plus or minus).

Still, it's an indication and that's all you would normally need anyway.

 

[Buddhafamily456]

Basically I’m getting lost in translation of what my next step is after finding lumens/par factor in the first equation. Or if that’s the final value what is that indicative of?
Forgive me because the only knowledge to find out around how dense your light is in a given area is the equation that I saw in the GrowMau5 video using u/mol/s/W.

 

[wietefras]

Yeah I like dividing better than multiplying by some small fraction. Besides, 69 is much easier to remember than a lot of digits behind the comma.

So for the first equation you gave me. The lumens/par factor .. the PPF number that you get is indicative of what?

Well if you get the lumen from the Samsung engine calculator, you could divide that by 69 and get the total PPF of your light.
http://www.samsung.com/led/support/tools/engine-calculator/

Divide that PPF by the surface area of your grow and you get an indication of the average PPFD.

Say 2100umol/s PPF divided by 3m2 gives you an average PPFD of 700umol/s/m2. Although you should subtract something for wall losses, but it's an indication.

And by that I mean my only understanding for getting a PPF value is from the YouTube video I saw from GrowMau5 where he figures it out using u/mol/sW x wattage(which I’m still not clear about, whether it is PAR Watts or total driver output?)

Another way of getting the PPF is taking the efficacy numbers in lm/W and converting to umol/s/W. Then multiply by total watts. This is watts from the wall. So multiply that 2.44umol/s/W by the total watts to get total PPF (example below).

"PAR watts" is pretty much a useless value for us. It doesn't compensate for spectrum. It's a relic from a few years ago when people thought you needed to calculate PAR W by dividing lumen/LER and then multiply by the QER. When in fact you can simply divide the LER by the QER and use that value go straight from lumen/lux to umol in one calculation.

So it's really much easier and more accurate to simply convert straight from lumen/lux to umol. Otherwise you might as well look at lumen directly and aim for say 55000lumen per m2 instead of 800umol/s.

Also what I can’t fully wrap my head around is, how would the PPF value you find be the same for 1 strip as 10 strips or any other value for strips at that matter?

It is not. If one strip uses 1.12A * 46V = 51.5W and produces 2.44umol/s/W * 51.5W = 126umol/s, then 10 of those would produce 1260umol/s (and consume 515W).

 

[Buddhafamily456]

Yeah I like dividing better than multiplying by some small fraction. Besides, 69 is much easier to remember than a lot of digits behind the comma.

Well if you get the lumen from the Samsung engine calculator, you could divide that by 69 and get the total PPF of your light

Divide that PPF by the surface area of your grow and you get an indication of the average PPFD.

Say 2100umol/s PPF divided by 3m2 gives you an average PPFD of 700umol/s/m2. Although you should subtract something for wall losses, but it's an indication.

So just for example, if my fixture output was

Flux = 128264 lm
Then I would have 128264 lm coming from my fixture.

This would be 3500k so I would use 71 as my par fActor. 128264/71= 1806 umol/s
To get my PPFD in a 4x4 area...
4x4 = 2304 in
128264/2304 = 55.67?
... i don’t understand where I’m going wrong?

Yeah I like dividing better than multiplying by some small fraction. Besides, 69 is much easier to remember than a lot of digits behind the comma.

Well if you get the lumen from the Samsung engine calculator, you could divide that by 69 and get the total PPF of your light.

Divide that PPF by the surface area of your grow and you get an indication of the average PPFD.

Say 2100umol/s PPF divided by 3m2 gives you an average PPFD of 700umol/s/m2. Although you should subtract something for wall losses, but it's an indication.


Another way of getting the PPF is taking the efficacy numbers in lm/W and converting to umol/s/W. Then multiply by total watts. This is watts from the wall. So multiply that 2.44umol/s/W by the total watts to get total PPF (example below).

"PAR watts" is pretty much a useless value for us. It doesn't compensate for spectrum. It's a relic from a few years ago when people thought you needed to calculate PAR W by dividing lumen/LER and then multiply by the QER. When in fact you can simply divide the LER by the QER and use that value go straight from lumen/lux to umol in one calculation.

So it's really much easier and more accurate to simply convert straight from lumen/lux to umol. Otherwise you might as well look at lumen directly and aim for say 55000lumen per m2 instead of 800umol/s.

It is not. If one strip uses 1.12A * 46V = 51.5W and produces 2.44umol/s/W * 51.5W = 126umol/s, then 10 of those would produce 1260umol/s (and consume 515W).

 

[nfhiggs]

4x4 = 2304 in

Its square meters you need to get to for the PPFD. 4x4 feet is 1.486 square meters. 1806 uMol/s divided by 1.486 = 1215 uMols PPFD. Quite a bit more than you need - most of us shoot for about 800.

 

[Buddhafamily456]

Its square meters you need to get to for the PPFD. 4x4 feet is 1.486 square meters. 1806 uMol/s divided by 1.486 = 1215 uMols PPFD. Quite a bit more than you need - most of us shoot for about 800.

Ahhh! Haha damn thanks man, I appreciate that.

My next question is I was planning on supplementing red/far red/ and blue light with cobs, I don’t really understand which ones to use to cover the area I’m using.
Could you maybe help me out by explaining basically how I would go about choosing a cob model to use and make sure I’m getting enough coverage of those light spectrums thru out my area ?

 

[Randomblame]

Bridgelux Decor Series has a 1750°k COB (for marbled meat), these have a nice blue/red/far-red spectrum and you don't need to use 3 different chinese blue, red and far-red COB's.
With cheap Chinese COB's, you would also loose some system efficiency, because these COB's may only come to 30-35% efficiency... when it comes high.

Timber Growlights for example use latest gen. 7 Vero29's in 1750°k in some of their growlights. @PSUAGRO has a nice thread where you can see them in action. Looks like growing on mars, lol!
Good brand red and blue COB's are hard to find. Citizen has a few but it's a PITA to find them. Maybe cobkits could help, if you're still interested. But there is no far-red COB available besides the generic ones.

Below is a screenie of the 1750°k spectrum. Lot's of blue and deep-red and plenty of far-red and these COB's are at least 50% efficient at nom. current. IMO much better as to add lot's of different no name COB's. But don't ask me for a conversion factor. Should be anywhere between 55 and 60.. Digikey should have Vero10, 18 and 29, search by part-number found in the datasheets.
You may find more information on the efficiency of this spectrum on Timber's website, can't say much, but it's maybe worth to visit it.




You already have very efficient stripes and you should only use the very efficient COB's to keep the 2,4μmol/j as good as possible. 8 Vero18c at 525mA for instance would be ~140w on an HLG-120H-C1050B (4s2p circuit) and should provide between 2.2-2.4μmol / j.

 

[Buddhafamily456]

Bridgelux Decor Series has a 1750°k COB (for marbled meat), these have a nice blue/red/far-red spectrum and you don't need to use 3 different chinese blue, red and far-red COB's.
With cheap Chinese COB's, you would also loose some system efficiency, because these COB's may only come to 30-35% efficiency... when it comes high.

Timber Growlights for example use latest gen. 7 Vero29's in 1750°k in some of their growlights. @PSUAGRO has a nice thread where you can see them in action. Looks like growing on mars, lol!
Good brand red and blue COB's are hard to find. Citizen has a few but it's a PITA to find them. Maybe cobkits could help, if you're still interested. But there is no far-red COB available besides the generic ones.

Below is a screenie of the 1750°k spectrum. Lot's of blue and deep-red and plenty of far-red and these COB's are at least 50% efficient at nom. current. IMO much better as to add lot's of different no name COB's. But don't ask me for a conversion factor. Should be anywhere between 55 and 60.. Digikey should have Vero10, 18 and 29, search by part-number found in the datasheets.
You may find more information on the efficiency of this spectrum on Timber's website, can't say much, but it's maybe worth to visit it.


View attachment 4147707

You already have very efficient stripes and you should only use the very efficient COB's to keep the 2,4μmol/j as good as possible. 8 Vero18c at 525mA for instance would be ~140w on an HLG-120H-C1050B (4s2p circuit) and should provide between 2.2-2.4μmol / j.

Thank you for showing me these.
So before I knew that these were available I had planned on making a series circuit of red, far red, and royal blue small cree cob LEDs from rapid LED
I was gonna use
cree xp-e photo red (660 nm)
Cree xp-e far red (720-740 nm)
And either
Cree xp-g3 royal blue led(440-455 nm)
Or
Semiled royal blue led(440-455 nm)

I was basically gonna make a strip out of them using them all in series and tying them into the parallel circuit I was making with the f-series strips.

So you think it’s a better idea to scrap that thought and go with these 1750k strips instead.

 

[Buddhafamily456]

With the light I am building I want to make sure I’m giving the best spectrum for grow quality.

What would be better 3000k or 3500k?

 

[Buddhafamily456]

So for my build I’m going to be using

25 Samsung F Series Strips
24 fV

 

[nfhiggs]

So for my build I’m going to be using

25 Samsung F Series Strips
24 fV

I'm assuming the 22", 25W strips - I would do 24 so you can split them evenly between two drivers - that would also allow you to put 6 strips over each 2x2 quadrant in a 4x4 tent. OTOH, If you're talking about the 44" 50W strips, that's way overkill.