Math behind

[Rahz]

All other things being equal the Fluence has a better spectrum. If you're wanting to see it as data, you would digitize both spectrums as well as the Mcree curve. Use the Mcree values as multipliers for the spectral data.

Here's a zip file from my first attempt. I went with 5nm segments but you can see the formulas used in the cells to get an idea of how to apply the data.

As far as digitizing, there are various graph digitizers. One that's relatively easy to use is GetData.

 

[Rahz]

Also, it could be less trouble to just use high CRI, rather than using something like 3500K 80CRI + 660nm red. You can see exact differences by digitizing and laying out out in a spreadsheet.

Personally, I think 3000K 90CRI is an excellent spectrum. Blue levels are where I like them and the red peak is at 630 with similar fall off on both sides = most amount of 580-680 available from white phosphor. It out-performs 2700K 90CRI on paper with and without factoring PAR readings.

 

[Rocket Soul]

As far as digitizing, there are various graph digitizers. One that's relatively easy to use is GetData.

Been waiting for info on how to plot the spectrums. Does GetData work for mac? Cause im assuming theres theres for android. And is the process easy, as in feed into to the program and done or do you have to plot it point per point. Thx so much for the info!

 

[Rahz]

Don't know about mac, but it's fairly easy. If it's the prog I think it is it will auto-identify the curve. If it has trouble you will have to clean up the image with an image editor of some sort.

 

[sethimus]

Also, it could be less trouble to just use high CRI, rather than using something like 3500K 80CRI + 660nm red. You can see exact differences by digitizing and laying out out in a spreadsheet.

Personally, I think 3000K 90CRI is an excellent spectrum. Blue levels are where I like them and the red peak is at 630 with similar fall off on both sides = most amount of 580-680 available from white phosphor. It out-performs 2700K 90CRI on paper with and without factoring PAR readings.

afaik samsung doesn't produce 3000k 90cri stripes

 

[Rocket Soul]

afaik samsung doesn't produce 3000k 90cri stripes

Afaik know only 90 cri strip solution is bridegelux vesta strips at 2700/5000k 90 cri, dual channels. Not as good as sammie diodes, about same efficiency as brudgelux eb gen 1.
Its surpring i never seen anyone try them here. Somone tried them on ledgardener but very little info. But probably a convo for another thresd, this thread is already messed up since the nofucks2give fiasco.
I wonder what happened to that dude. But maybe better not mentioning. If you say his name 3 times he'll be back under a new name.
This place for was like whackamole for a good while

Edit: did a search and the accounts have been deleted from riu. Thx mods, for once...

 

[Rocket Soul]

Don't know about mac, but it's fairly easy. If it's the prog I think it is it will auto-identify the curve. If it has trouble you will have to clean up the image with an image editor of some sort.

Ill give it try when i get some time. Theres a few specialty high cri spectrums that ive been wanting to check out properly.

 

[wietefras]

I use Engauge Digitizer which produces a 1nm export.

In this case the trouble is not so much the digitizing though, but if you want to calculate how much watts per type of led then you need a bit more work to normalize the data

 

[sethimus]

i asked my way smarter cousin, he did a rough calculation for me with my botched digitizing job, looks like that additional 660nm area of the phytospec indoor spectrum is about 7% more than the vanilla 3500k from samsung, less than i expected.

 

[sethimus]

so the 660nm part of the spydr 2 should be about 120µmol/s ppf

 

[Dave455]

i asked my way smarter cousin, he did a rough calculation for me with my botched digitizing job, looks like that additional 660nm area of the phytospec indoor spectrum is about 7% more than the vanilla 3500k from samsung, less than i expected.

View attachment 4234032

View attachment 4234033

really what kind of yield improvement ?

 

[PilouPilou]

(408 lm301b + 8 osram 660nm per bar)

Hi!

408 lm301B to make around 100W per bar it must be mega efficient!

 

[sethimus]

really what kind of yield improvement ?

?

did you even understand what i wanted to find out?

 

[mr. Amsterdam]

Afaik know only 90 cri strip solution is bridegelux vesta strips at 2700/5000k 90 cri, dual channels. Not as good as sammie diodes, about same efficiency as brudgelux eb gen 1.
Its surpring i never seen anyone try them here. Somone tried them on ledgardener but very little info. But probably a convo for another thresd, this thread is already messed up since the nofucks2give fiasco.
I wonder what happened to that dude. But maybe better not mentioning. If you say his name 3 times he'll be back under a new name.
This place for was like whackamole for a good while

Edit: did a search and the accounts have been deleted from riu. Thx mods, for once...

Bridgelux is coming with 90 CRI options for there EB series gen 3 strips mid december.

But Greengenes and Growmau5 did say, at the latest 'tech talk', that using 80 or even 70 CRI whites with added 660nm reds are more efficient because of the thinner phosphor layer that is letting through more photons then 90 CRI.

 

[Rahz]

The XP-E photo-red is 2.34 PPF (top bin) at .735 watts. Output of .425 radiant watts. This indicates 57.8% efficiency and 3.18 umol/j. Next bin down is 3 umol/j then 2.82 and 2.64. Pretty good deal and could potentially unseat cobs, but there is also cost to consider. The best of both worlds would be XP-E pucks. Pair them up 80w to 20w with the blue Citizen cobs and you have the ultimate blurple lamp, could easily achieve 3.2 umol/j output.

To reach that efficiency with Vero29C 3000K 90CRI you would need to run them at 350ma (22.4 watts each). That's $1.29 per output watt assuming $29 Vero cost.

If we assume a few percent enhancement from spectrum, it would put the virtual price at $1.33 ballpark figure, so a 3.2 umol/j blurple for less than $1.33 per output watt would be a good deal. That's just emitter cost, with everything else being equal.

 

[Rocket Soul]

Bridgelux is coming with 90 CRI options for there EB series gen 3 strips mid december.

But Greengenes and Growmau5 did say, at the latest 'tech talk', that using 80 or even 70 CRI whites with added 660nm reds are more efficient because of the thinner phosphor layer that is letting through more photons then 90 CRI.

Awesome! You have a link for this? As in bridgelux eb gen3?

 

[mr. Amsterdam]

Awesome! You have a link for this? As in bridgelux eb gen3?

No but when my rep at bridgelux sends it to me i will post it.

The eb gen 3 series will come in 2700K, do you guys think this will be similar like 3000K or 3500k with added red 660nm?

This are dry yield and stretch charts with different color temperature leds from the Micro Grow Blog which Wietefras had mentioned earlier.
Here you can see what's the best spectrum.

 

[Rocket Soul]

No but when my rep at bridgelux sends it to me i will post it.

The eb gen 3 series will come in 2700K, do you guys think this will be similar like 3000K or 3500k with added red 660nm?

This are dry yield and stretch charts with different color temperature leds from the Micro Grow Blog which Wietefras had mentioned earlier.
Here you can see what's the best spectrum.

View attachment 4239623

View attachment 4239624

I opened a thread about the new gen 3 strips here:
https://www.rollitup.org/t/bridgelux-eb-strips-gen-3.980457/

We should try to take or conversation there instead as this thread is about math/spectrum etc.

 

[GD_Cronista]

Afaik know only 90 cri strip solution is bridegelux vesta strips at 2700/5000k 90 cri, dual channels. Not as good as sammie diodes, about same efficiency as brudgelux eb gen 1.
Its surpring i never seen anyone try them here. Somone tried them on ledgardener but very little info. But probably a convo for another thresd, this thread is already messed up since the nofucks2give fiasco.
I wonder what happened to that dude. But maybe better not mentioning. If you say his name 3 times he'll be back under a new name.
This place for was like whackamole for a good while

Edit: did a search and the accounts have been deleted from riu. Thx mods, for once...

You can say wathever you want about NoFucks2Give, but I have learned a lot from him. I'm actually sad that he is not posting anymore, I hope he is still alive (I think he is around 65yo now).

Almost everyone was giving him a hard time in every post he made, when he was clearly a smart and knowledgeable man.

I did not read this complete thread yet (but I will), so I don't know if if he screwed up in this concrete instance. But overall, across all the things that I have read from him, he seemed to be pretty much on point every time; while also sharing lots of information, pdfs, interesting graphs and images... It's thanks to him (and also Rahz, I must admit) that I can make an educated guess about what lights I should buy.

So, if by any chance you read this, NoFucks2Give, thank you A LOT. You were (and still are, because I just started to learn about LEDs like a week ago) a fountain of knowledge to me.

And yes, I made an account to say this, because I felt I had to (well, I made to account to be able to see all the images that he was sharing, to be honest).

Excuse me for my english and have a nice day everyone.

PS: I noticed I needed to say NoFucks2Give one more time. So maybe, as you said, he will be back.

 

[wietefras]

You can say wathever you want about NoFucks2Give, but I have learned a lot from him.

Just be careful that pretty much everything he said was wrong. So if you are just starting out and you think you got great information from him, you will have started on the wrong path.