Help using knna's and Beta Test Teams Data Sets

[Ranger]

This thread will be for anyone needing help with knna's and Beta Test Teams data materials and spread sheets to compare and utilize bulbs and such.

Hopefully the Moderators can move a few posts from BTT's Threads on several issues. Here are links to various threads where data can be found along with several spread sheets.

Cannabis absorptance spectra:Calculated and compared.
https://www.icmag.com/ic/showthread.php?t=293045

[FONT=Arial, Helvetica, sans-serif]Action spectra and RQEs weighted with Cannabis absorptance[/FONT][FONT=Arial, Helvetica, sans-serif] [/FONT]
https://www.icmag.com/ic/showthread.php?t=295933

[FONT=Arial, Helvetica, sans-serif]Various approximate RQE of Cannabis (graphed)[/FONT]

https://www.icmag.com/ic/showthread.php?p=6718314

Link to knna's last spread sheet and info on using it.

http://buymarijuanaseeds.com/community/threads/bulb-analyzer-tool-actualized.117933/

I made this thread so anyone wanting to learn how to utilize this data can perhaps have an easier time doing so for all those whom like me needs a little help.

I'll be adding more later as we get some posts moved.

 

[Beta Test Team]

Hi Ranger, if you want to ask a mod to move the posts that's fine with me, or I can, just let me know.

For those reading, below are the posts we're referring to; the main discussion Ranger and I are having is how to calculate the lux to PPF conversion factor for a specific lamp.

This topic is useful when using a lux meter to measure irradiance, because we can convert lux to umol (photons), so a grower doesn't need to spend $1,000 on a good quantum sensor, instead, they can spend around $150 on a good lux meter - or around $350 on a good NIST certified lux meter - and then convert lux to PPF. For our suggested lux meter, see this post:
https://www.icmag.com/ic/showpost.php?p=6570166&postcount=75

Here are the posts we are going to ask to be moved:
https://www.icmag.com/ic/showpost.php?p=6846220&postcount=168
https://www.icmag.com/ic/showpost.php?p=6847142&postcount=169
https://www.icmag.com/ic/showpost.php?p=6847975&postcount=170
https://www.icmag.com/ic/showpost.php?p=6849400&postcount=171
https://www.icmag.com/ic/showpost.php?p=6850233&postcount=173
https://www.icmag.com/ic/showpost.php?p=6850877&postcount=174
https://www.icmag.com/ic/showpost.php?p=6850901&postcount=175
https://www.icmag.com/ic/showpost.php?p=6850941&postcount=176

 

[Beta Test Team]

So just to start out with a bit of context, our spreadsheet called HISSA ("Harvest Irradiance Spectral-System Analyzer") won't be publicly released until we can afford to have 3rd party SPDs created for various lamps. See this thread for the topic of 3rd party SPD measurement:
https://www.icmag.com/ic/showthread.php?t=298833

knna's spreadsheet "BA_models" (the Microsoft Excel version) is what people should use for now, until HISSA is released to the wild. To that end I have just made a few bug fixes to knna's most recent spreadsheet (from 8-27-2008), which I uploaded to our website (this is the version everyone should use):
https://csrg.info/wp-content/uploads/2015/03/BA_Models_xls-bug-fixes-by-BTT-3-15-2015.xlsx


The main reason Ranger was asking me about knna's spreadsheet is to calculate the radiant PPF/Klm ("Klm" = kilolumen) conversion factor. However, knna's spreadsheet does much more than that single calculation.

I hid the two sheets in knna's spreadsheet titled "5 nm" and "10 nm" because we only want to use the "1 nm" sheet for most SPD graph of lamps' growers use. This increases accuracy a lot.

Also, the value knna teremed "PUR" (from K. Inada's work) is not really worthwhile, so in my opinion it can be ignored. The better value is YPF (from K. McCree's work).

Below are the bug fixes I made to knna's spreadsheet. He made a few errors, one of which will skew the resulting weighted irradiance data knna called "PYF." That bug is important to fix, because from knna's spreadsheet the only two important values for radiant energy are "PPF" and "PYF." The former, PPF, is unweighted, the latter "PYF" is weighted to take into account Cannabis absorptace and higher plant photosynthesis action spectrum:

1) Renamed "PYF" to "radiant YPF," the latter is the correct term which stands for "Yield Photon Flux," the former is not correct. Radiant YPF is found by multiplying RQE by "radiant PPF." NOTE: the value "radiant YPF/radiant PPF" is used to find YPF from PPF, just like finding PPF from lux (i.e. multiply PPF by the "radiant YPF/radiant PPF" ratio to find YPF).

2) Renamed "PPF" to "radiant PPF," because "PPF" is umol/s/m2 in PAR range, while "radiant PPF" is umol/s in PAR range; the former is irradiance while the latter is radiance.

3) "McCree" RQE ("relative quantum efficiency" of photosynthesis) data was corrected to only include C3 plants (knna's version includes C3 and C4), as well as to properly remove McCree's absorptance data and replace it with Cannabis absorptance data, then normalized to 1.

4) Radiant YPF equation was corrected to multiply RQE by radiant PPF.

5) "Radiant PPF (400-700)" (cell D13) was corrected so it's using "PAR W" from 400-700 nm, not radiant watts knna lists as "W (380-780)" that it was using.

 

[Beta Test Team]

The main value of interest to Ranger that started our discussion is "radiant PPF/Klm" ratio, which is the same thing as the multiplication conversion factor to find PPF from lux.

So for example, if a lux meter shows 145,000 lux, to find PPF a grower would first divide the lux reading by 1,000 (if it's not reported as Klux already), then multiply that resultant value by the radiant PPF/Klm value from knna's spreadsheet.

However, Dr. Bugbee and student have calculated PPF/Klux values for many lamps, so for example, if you're using Philips MasterColor Elite Agro 930 T12 315W CDM lamp, you can find the PPF/Klux ratio in the following PDF. Using that value works the same as the value from knna's spreadsheet: divide the lux value from the meter by 1,000 (if it's not reported as Klux already), and then multiply by the conversion factor (which is the same thing as the PPF/Klux ratio):
http://cpl.usu.edu/files/publications/poster/pub__6740181.pdf

For example, in the PDF above (from Dr. Bugbee) the PPF/Klux value for my favorite lamp, the Philips MasterColor Elite 942 T12 315W CDM, is 14.3. Therefore, if the lux reading from the meter is 145,000, I would divide 145,000 by 1,000, so, 145, and then multiply that by the ratio (14.3), giving a value of about 2,074 PPF (i.e. umol/s/m2 from 400-700 nm).

 

[Beta Test Team]

So that's about all I have to add at this point. But I'm happy to answer questions when I'm able. So if people have questions about using knna's spreadsheet, or about the processes of SPD normalization and digitization, fire away.

 

[Ranger]

Ok I have some data to add but first I want to thank Beta for all the work he has done and is doing to aid us all in utilizing this data and making for a better grow for anyone interested.

With this spread sheet work book you can figure out several important parameters for your lighting and grow room needs.

I had trouble with using it so I asked Beta to help me out and he did so much more. What we need to do is use the spread sheet in Beta's post above. Also here

https://mega.co.nz/#!34ElAIyI!3xd2NiHU09VLOxB-2w06LsloKJLK8SS9gEfCmKSIwBE


Once you download this file you have the latest updated version, by Beta Test Team and then all you need is also the spread sheets titled. "Hand Digitizer.zip" or "hand digitizer.xls.zip" if you prefer, both can be found in the OP post on this page.

http://buymarijuanaseeds.com/community/threads/bulb-analyzer-tool-actualized.117933/

Disregard the other zip files titled "BA Models" since we have and updated version from BTT above.

Once you have these installed you can use the guide that is part of the thread above to digitize your SPD chart for the bulb you are interested in. The guide on that page is very informative and if you have any questions on using it please feel free to ask.

Here is a sample SPD from a Solis-Tek 6k 400 watt bulb.

These are the results I got using this method and I'll wait for BTT to go over it and see if it looks correct and also to let me know how best to copy the recorded figures and apply them to the forum.

 

[Ranger]

So with my chart I came up with a Radiant PPF/Klm of 14.9 so I'll use Beta's equasion and multiply 14.9 x my Lux reading of 60Klux @ 12 inches above meter (canopy). So 14.9 x 60 would be 894 PPF (umol/s/m2 from 380-780nm).

PPFD and PPF are the same thing btw.

At 12" -894 PPF
At 16" - 625 PPF
At 20" - 461 PPF
At 24" - 447 PPF

So as you can see it's quite useful and now I can use this figure to calculate the DLI (Daily Light Integral) and hopefully deliver just right amount of photons to my plants per day/night cycle.

[FONT=Arial, Helvetica, sans-serif]([/FONT]DLI can be calculated this way- PPFD X (Hours of light per day) x (.0036)

So for a 12 hour day and 12 hour night @ 12 inches above canopy (What many use for flower) I would have a DLI of 38.62 and as Beta has mentioned a DLI of 30-55 is good for cannabis in all growth stages.

Now to recap and remember this DLI is good for 1 meter square of canopy space @ 12 inches above canopy. Just adjust your figures for raising or lowering your lights.

One of the major reasons I have done the work to figure this out is just how detrimental too little or too much DLI can be to Cannabis. Too much DLI being the worse of the two btw.

 

[Beta Test Team]

Ok I have some data to add but first I want to thank Beta for all the work he has done and is doing to aid us all in utilizing this data and making for a better grow for anyone interested.

You're welcome, glad to help. I wish knna still posted because there are things about his spreadsheet I don't think I understand fully, mostly due to language barriers (he's a native Spanish speaker, I think). As well as some of his methods I would like to ask him about.

Nevertheless, knna did an amazing job on his spreadsheet and it was really great of him to share it for free. While there were some bugs in his work (most of which I fixed in the version I posted), his work is unlike anything I've seen before available to the general public.

I have a good idea how much time, effort, and thought knna put into his spreadsheet, and it's quite amazing. Much respect goes to knna.

However, I think it's important to also give lots of credit to the people who did the groundwork knna used for his spreadsheet, most notably is Dr. Ivo Busko, as well as Dr. Sager, Smith, Edwards, and Cyr. Without whom I suspect knna (and ourselves, as well) wouldn't have been able to create what he created (and what we're creating). This is the one thing that kind of bothers me about knna's work: he didn't credit the scientists who came up with the methods he used in his spreadsheet, as found here:

Dr. Ivo Busko's groundbreaking work from the year 1999:
http://www.discusnews.com/article/cat-04/lightcompare.shtml#S-3

Dr. Sager, Smith, Edwards, and Cyr's work from the year 1988:
https://mega.co.nz/#!m40HFQbB!wtujIooa14CZOUy7RS2inQBX2lQU2-qgoSmsyr9dwXM

Here is a sample SPD from a Solis-Tek 6k 400 watt bulb.

View Image

These are the results I got using this method and I'll wait for BTT to go over it and see if it looks correct and also to let me know how best to copy the recorded figures and apply them to the forum.

View attachment 308763

Those data look pretty accurate and correct, great job. I do like those Soilstek MH lamps a lot, I just wish they have longer useful life span.

To post a full size of your screenshot, visit the screenshot in your gallery, and copy the link text under "BB Code," then past it into a post.

 

[Beta Test Team]

So with my chart I came up with a Radiant PPF/Klm of 14.9 so I'll use Beta's equasion and multiply 14.9 x my Lux reading of 60Klux @ 12 inches above meter (canopy). So 14.9 x 60 would be 894 PPF (umol/s/m2 from 380-780nm).

PPFD and PPF are the same thing btw.

At 12" -894 PPF
At 16" - 625 PPF
At 20" - 461 PPF
At 24" - 447 PPF

So as you can see it's quite useful and now I can use this figure to calculate the DLI (Daily Light Integral) and hopefully deliver just right amount of photons to my plants per day/night cycle.

[FONT=Arial, Helvetica, sans-serif]([/FONT]DLI can be calculated this way- PPFD X (Hours of light per day) x (.0036)

So for a 12 hour day and 12 hour night @ 12 inches above canopy (What many use for flower) I would have a DLI of 38.62 and as Beta has mentioned a DLI of 30-55 is good for cannabis in all growth stages.

Now to recap and remember this DLI is good for 1 meter square of canopy space @ 12 inches above canopy. Just adjust your figures for raising or lowering your lights.

One of the major reasons I have done the work to figure this out is just how detrimental too little or too much DLI can be to Cannabis. Too much DLI being the worse of the two btw.

That all looks great, good job.

However, just to point out, PPF is measured from 400-700 nm, while lumens (in knna's spreadsheet) is measured from 380-780 nm. The conversion factor accounts for this waveband mismatch. Also, lumens are really measured from 390-830 nm, not 380-780 nm, at least with the most recent (2006) CIE luminous efficiency functions (I don't think knna is using the 2006 CIE functions).

 

[Ranger]

That all looks great, good job.

However, just to point out, PPF is measured from 400-700 nm, while lumens (in knna's spreadsheet) is measured from 380-780 nm. The conversion factor accounts for this waveband mismatch. Also, lumens are really measured from 390-830 nm, not 380-780 nm, at least with the most recent (2006) CIE luminous efficiency functions (I don't think knna is using the 2006 CIE functions).

Hey thanks for looking over the data BTT and I was wondering about that also. As can be seen my Solis Tek bulb has a SPD from 340 to 750 and was not sure how to proceed.

What I did was set it up from the X axis at 380nm and started from there and just "approximated" anything beyond the 750nm mark as best I could.

The Ushio bulb I'm working on now is even less accurate due to the very basic SPD published by the manufacturer. I'll work on posting that data as well. Also for the Ushio bulb they do not post an initial lumen number so I'm using an average of 400 watt sodium bulbs of 55Klumens.

Would be great if others posted what they have found with bulbs that haven't been tested as yet. Like what knna stated back then so anyone can just look it up I suppose in possibly your new program (HISSA) which I'm looking forward to.

 

[Beta Test Team]

Hey thanks for looking over the data BTT and I was wondering about that also. As can be seen my Solis Tek bulb has a SPD from 340 to 750 and was not sure how to proceed.

What I did was set it up from the X axis at 380nm and started from there and just "approximated" anything beyond the 750nm mark as best I could.

It may have been better to just use the defined waveband for the SPD, rather than guessing past 750 nm. The reason is past about 750 nm the photopic curve (representing luminous efficiency function) is very weak, that is, there's little use of greater than 750 nm by human eyes. So if you only included the defined waveband your results would likely be not different much from you current method of guessing at relative spectral radiation past 750 nm. For example, here's a photopic curve where you can see there's little action greater than about 700 nm (I'm not sure what version of CIE luminous efficiency functions were used to create this curve, i.e. if it's the 2006 version, or earlier versions):

​

The Ushio bulb I'm working on now is even less accurate due to the very basic SPD published by the manufacturer. I'll work on posting that data as well. Also for the Ushio bulb they do not post an initial lumen number so I'm using an average of 400 watt sodium bulbs of 55Klumens.

Those DE SPD they have are low resolution, so they're pretty worthles. They look like they're made with 2 nm or 5 nm stepsize (i.e. measuring irradiance over a 2 nm to 5 nm waveband, rather than every 1 nm).

You could find the lumnes for the lamp from the listed radiant PPF values (which they incorrectly call "PAR") with knna's spreadsheet, similarly to how you found PPF from lux. But if you send them an email they may have those data they can send you (lumen data, that is).

Would be great if others posted what they have found with bulbs that haven't been tested as yet. Like what knna stated back then so anyone can just look it up I suppose in possibly your new program (HISSA) which I'm looking forward to.

Yes, I agree, it would be nice to have users post that data.

In terms of our work with HISSA, we intend to have 3rd party SPD created for all lamps of interest, and in doing so the relative values per nm are provided from the lab, so there's no need to normalize and digitize the SPD, because we already have the raw data used to create the SPD graph. This method is much more accurate than normalizing and digitizing published SPDs, like what you're doing now. The down side is it's about $200 per lamp to have 3rd party lab create SPDs for us, and we need to test at least 10 lamps to get the price point (for single tests it's $280).

 

[Beta Test Team]

Ranger, about HiLux Gro lamps from Ushio, I found these lumen data. I'm not sure if this is correct, but you could likely send an email to Ushio and get a response pretty quickly:

http://pentairaes.com/hps-400-w.html
HHP109: HPS, 1,000 W, 146,000 lumens
HHP110: HPS, 600 W, 97,000 lumens
HHPSB109: HPS/MH, 1,000 W, 117,000 lumens
HMHC100: MH Conversion, 1,000 W, 109,000 lumens
HMHC114: MH Conversion, 600 W, 55,000 lumens
HMHC123: MH Conversion, 400 W, 39,000 lumens
HHP123: HPS, 400 W, 56,000 lumens

 

[Ranger]

It may have been better to just use the defined waveband for the SPD, rather than guessing past 750 nm. The reason is past about 750 nm the photopic curve (representing luminous efficiency function) is very weak, that is, there's little use of greater than 750 nm by human eyes. So if you only included the defined waveband your results would likely be not different much from you current method of guessing at relative spectral radiation past 750 nm. For example, here's a photopic curve where you can see there's little action greater than about 700 nm (I'm not sure what version of CIE luminous efficiency functions were used to create this curve, i.e. if it's the 2006 version, or earlier versions):


Those DE SPD they have are low resolution, so they're pretty worthles. They look like they're made with 2 nm or 5 nm stepsize (i.e. measuring irradiance over a 2 nm to 5 nm waveband, rather than every 1 nm).

You could find the lumnes for the lamp from the listed radiant PPF values (which they incorrectly call "PAR") with knna's spreadsheet, similarly to how you found PPF from lux. But if you send them an email they may have those data they can send you (lumen data, that is).


Yes, I agree, it would be nice to have users post that data.

In terms of our work with HISSA, we intend to have 3rd party SPD created for all lamps of interest, and in doing so the relative values per nm are provided from the lab, so there's no need to normalize and digitize the SPD, because we already have the raw data used to create the SPD graph. This method is much more accurate than normalizing and digitizing published SPDs, like what you're doing now. The down side is it's about $200 per lamp to have 3rd party lab create SPDs for us, and we need to test at least 10 lamps to get the price point (for single tests it's $280).

Yeah that's pretty much what I did was just input the lowest possible numbers for the 750-780 nm range since my lamp SPD was on a steady decline at that point.

Thanks for the lumen input amigo so 55Klumen was a reasonable guess and from what I gather "most" 400 watt HPS bulbs seem to fall there.

I'm digging those 315 CDM bulbs you guys are rocking btw. What kind of price is a setup bulb and ballast?

Wow that is expensive to get bulbs tested, didn't realize that. Your program should real nice when completed.

 

[Beta Test Team]

I'm digging those 315 CDM bulbs you guys are rocking btw. What kind of price is a setup bulb and ballast?

Price depends upon what you want to buy. For example, Greenbeams (lamp + state-of-the-art reflector + ballast) is about $420 each. Then there's also the dPapillion (one 315W CDM) or Double dPapillion (two 315W CDM), but I'm not sure of their price.

For other setups I would suggest asking rives or Jhhnn. Check out rives's great thread on this topic:
https://www.icmag.com/ic/showthread.php?t=299165

 

[tomate]

knna's spreadsheet "BA_models" (the Microsoft Excel version) is what people should use for now, until HISSA is released to the wild. To that end I have just made a few bug fixes to knna's most recent spreadsheet (from 8-27-2008), which I uploaded to our website (this is the version everyone should use):
https://csrg.info/wp-content/uploads/2015/03/BA_Models_xls-bug-fixes-by-BTT-3-15-2015.xlsx

Can someone re-upload this spreadsheet?