Discussion in 'LED and other Lighting' started by Prawn Connery, Nov 11, 2017.
Yep. Sorry. You're right - I get it. I was going on your penultimate post.
I have a good news and possible solution for your "time problem"..
F-Series single row strips are available today! 384 3k/2ft. strips in stock! Also other spectrums(3,5, 4 and 5k) are again available. Hope you have not ordered elsewhere yet.
I'm actually looking at the new Q strips with LM301Bs. I will need about 3x as more compared to F Series double row, but there's not a massive price difference once you bulk buy 25+. I'm thinking two frames, 12-14 strips each (6p/7p 2s = 48v), each driven by a HLG-240H-48A for a total of about 480w.
What do you think? I'll have more time tomorrow to have a proper look - just an idea at this stage.
EDIT: I think the LM301B 3000K 80CRI curve looks a bit better than the LM561C at first glance. Slightly more efficient, too.
I already referred to the sheet alesh made to transform SPD to QER and LER and I gave actual numbers plus an explanation why the "actual numbers" mean jack shit.
Perhaps read the whole discussion instead of just one post?
BTW, it's not just LER you also need the QER. Or go one step further and calculate the YPF corrected for McCree (or whatever curve you believe in).
Anyway, it's this pretend accuracy which is actually the bane of this forum. People think they can calculate everything to dozens of decimals when the base data is already only accurate within 10%.
The point still is that the accuracy you assume is there, is actually nonexistent.
I already told you there is less than 2% difference between those SPDs. That's well within the actual (in)accuracy of the datasheets and therefore can be ignored.
If you have something with an actually different SPD then it takes 5 minutes (if not less) to digitize a chart and run it through the calculations alesh posted.
3x more strips but increased photons from more angles illuminating the girls up and down !!!!
You referenced a spread sheet...that is it. You didn't supply the spread sheet...or even apply to the questions in this thread your arguing. You could have at least linked it if you weren't going to actually use it and make some calculations to validate you guesses, but you didn't. Then you went off on a reductio ad absurdum in that the data sheets aren't accurate enough to do anything with...is that your way out of doign actual calculations??? Pure pretentious ambiguity to deflect from the lack of actual numbers and support in your claims. It's simple...do the calcs and stop talking out your....
Here are all your post, you tell me where you made any actual calculations, what the LER is that you arguing about...and so on. Point out one factual based example you made, one calculation, one hard fact number...I'll wait...
As you can see...you have not used what you asserted(the spreadsheet). You have not supplied any example calculations or actual calculation that you speak about.
And to top it off, you don't need QER, but as I said you could go that far with it. Radiant watts is the even playing field that all chips, even with different or shifted SPD's can be compared on. But you have calculated, supplied, or corrected any of them. You just hypocritically comment. Simple as that. Feel free to supply some actual data driven content for your post. Do the calculations, show them...even just a link to have alesh do them. But you gave nothing, claimed that you did...and have now been called on it and exposed using your own post.
I gave you the out to just do calcs and shut him up(and now me). But you didn't present anything in your support.
You get it???...now do some calcs or go back under your bridge.
I have no dog in this fight except the facts. I called you both out. So get over your ego and support your assertions with actual calculations.
I'm not arguing, because what you have stated - to my limited understanding - makes perfect sense.
The point I will make is that I never pressed the validity of the SPD charts I posted other than to use them to ask a rhetorical question: how can you compare advertised lumens per watt between two "same-colour" diodes when they are not exactly the same colour (in spectral terms)?
I provided a seat-of-the-pants figure in terms of relative percentages of 450nm vs 550nm purely in response to this (because I thought his method was flawed):
I don't want to sink the boot into a guy who's down, but I don't think any of us know where the fuck that 2% figure comes from. He hasn't shown how he calculated it - or even guesstimated it.
If he can show how he got that figure, and if the method is sound, I'll gladly STFU.
In addendum, no-one is arguing there aren't tolerances applied to datasheets, but you have to start with something. As has been pointed out.
Just crunched some numbers and the Q Series are quite a bit more than the F Series. I'd need 4x strips to equal one F Series double.
The following are all based on 2-foot Samsung 3000K strips, availability and bulk discounts. Wattage is 3v x 200ma (F/H Series) and 2.91v x 180ma (Q Series) x diodes to keep it simple. Two light frames means halve the number of strips for each frame, each driven by 240w (or possibly 320w). Odd numbers are spares.
F double x 10 = $243 = 864w
.................x 8 = $220 = 691w
F single x 25 = $280 = 1080w
..............x 16 = $230 = 691w
F single x 25 = $253 = 1080w
..............x 16 = $200 = 691w
..............x 12 = $150 = 518w
H single x 25 = $241 = 720w
..............x 20 = $243 = 576w
H single x 25 = $215 = 720w
..............x 20 = $210 = 576w
Q single x 25 = $247 = 524w
..............x 24 = $247 = 503w
Best value for my application seems to be the F Series singles from Arrow. $150 buys 12 strips for 518w, which is about the same as 24 or 25x Q Series for $247 from Digikey.
See it from this point.
A 3k Q-Series strips at 450mA needs 9,85w and put's out 187lm/w for 1840mA total(1st screenie).
A 3k F-Series strips at 525mA needs 11,81w but driven low it put's out 187,9lm/w for 2217lm total(screenie 2).
That means F-Series trumps Q-Series in all aspects.
- You can get more watt's,
- more lumens,
- better effiency and
- more diodes, but
- still a better price ..
..from F-Series strips when driven @525mA(58,34mA per diode) and even better when running at the same current.
Only above 62,5mA per diode the Q-Series is more efficient. And the minor differences in spectrum has more or less no effect, both are 3k spectrums, both peaks are in the same range.
These numbers are just copied from another thread:
LM561C - A1/S6 - 3000K @65mA
VF: 2.9V (max)
Flux: 34.5 lm/W (min)
QER: 321.6 lm/W
LER: 4.86 µmol/J
-> 2.766 µmol/J
LM301B - A1/SK - 3000K @65mA
VF: 2.9V (max)
Flux: 36 lm/W (min)
QER: 331.83 lm/W(+3%)
LER: 4.816 µmol/J(-1%)
-> 2.772 µmol/J
I already told you twice.
:edit: You could also just scale the charts to see how little they differ.
You come in here with your pretend know it all nonsense and yet you miss half the story (the spreadsheet and what it does is clearly mentioned in what you quoted) and even what you say is only half accurate
You said some dumb things. Deal with it. Stop crying and pretending. If you want people to help you understand things then you shouldn't be such an ass to them to begin with.
BTW You do need QER for something useful and for true accuracy. I'm not going to look up a resource that's here on the site. "alesh + math" should get you along. If not, tough luck.
Either way, only a blind man would think those two SPD's look different. Would also need to be someone who doesn't understand at all how these things work. It's pretty much the same coating, but one has more blue. What could that mean? Perhaps just a little less of that same coating?
This is the referenced spreadsheet: http://www.rollitup.org/t/math-behind.868988/
I don't know where to start . . . because you clearly don't know what you're looking at in those graphs. I'm just hoping you have a basic understanding of maths so that you can comprehend where you've gone wrong in trying to stretch a few lines to fit an argument.
The original Cree graph shows there is about 33% more 450nm than 600nm (100% relative to 75%).
Your graph shows 25% more (125% relative to 100%).
The original Cree graph shows there is about 426% more 450nm than 480nm (100% relative to 19%)
Your graph shows 468% more (125% relative to 22%).
The original Cree graph shows there is about 455% more 450nm than 680nm (100% relative to 18%)
Your graph shows 495% more (125% to 21%).
Do you see where this is going? You have just interpolated a couple of lines that have changed the entire spectral relationship of the original graph by shifting the scale.
It doesn't work that way.
Honestly, you cannot see from that combined chart how almost identical those two charts are? Wow. Just wow.
It actually does since it's just a "relative" chart.
The only difference is some blue because one of the two has slightly less coating.
Either way there is only 2% difference in lumen to PAR conversion rates. So insiginificant.
So in reality they will not be the same color temperature, but that's another thing where stuff varies and longing for multi decimal "accuracy" is pointless.
So you come back with ZERO calculations or actual figures.
This has nothing todo with me yet you attempt to make it personal(and failed) beau sue of you lack of actual substance on the subject. It has todo with your childish arguments with no support. And you just continued it without any support.
Stop being a kid yelling with your words. And be an adult that shows and proves things with facts and numbers. Or just shut the duck up cause you can’t support you claims. Pick one.
"Wow. Just wow" is not an argument.
Yes, it is a relative chart - that is why it is called a Relative Intensity vs Wavelength chart.
You cannot change the scale of the graph without changing the relative values. I already proved that to you in mathematical terms. If you are arguing the maths, then please point out where I am wrong . . .
You also cannot compare peaks from one graph to another until you attribute a value to each peak being the total radiant energy divided by the total area under the curve. (Which is what Greengenes is saying.)
The more area under the curve the graph shows, the less relative value is attributed to each % peak.
So by shifting the scale, you've just added more area under the curve without changing the relative values of each peak!
Mate, if you are going to keep stating "there is only 2% difference in lumen to PAR conversion rates", you really need to provide a basis and methodology for that calculation.
But somehow, I don't think you can or will. Because I suspect you do not fully understand the concept of "relativity" as proven by your willingness to doctor a graph without comprehending the underlying mathematical relationship.
We have to stop meeting like this.
It’s pretty simple. If he wants to be a loud mouth want to be knownit all...use actual facts and calculations that can’t be argued. It really is that simple. And the fact that he vaguely references a thread about calculations, but fights so hard to not do them, just shows he has know real understanding and won’t actually apply it...cause he doesn’t know how. Which is all fine and dandy. But when he comes out like he is the shit, but in reality can’t even do or present the most basic of calicaltiosn(hell even a link)...you see where my issues with mr no math come from.
He goes around doing this all the time attempting to look smart and bash others without any actual hard facts. If he has support, then great, but he doesn’t. He has been called on it and is showing his true colors with the lack of valid responses.
Separate names with a comma.