OneHitDone
Well-Known Member
Be sure to share with us how lettuce develops under this spectrum
PAR test of cree vs citi vs cutter
- Thread starter BOBBY_G
- Start date
true understand all of what you said.
But don't we want to know how it performs under realistic operating conditions ? Otherwise its just theoretical potential.
JorgeGonzales
Well-Known Member
Sure, it's just not this test. Go back and look at SupraSPL's tests. I was afraid I'd be embarassed and that's not how his spreadsheets were measured, but it was. Whew. 75F pulsed.
JorgeGonzales
Well-Known Member
Besides we have this data which we can apply with reasonable certainty:
It's a lot harder to make these charts, but current droop is relatively easy, and numbers seem chronically pessimistic with the manufacturers. Cree, Bridgelux, and Citizen all seem to do better than the tools, data sheets, and simulators indicate. How much better is the question.
Still waiting on that sphere data from PLC.
It's a lot harder to make these charts, but current droop is relatively easy, and numbers seem chronically pessimistic with the manufacturers. Cree, Bridgelux, and Citizen all seem to do better than the tools, data sheets, and simulators indicate. How much better is the question.
Still waiting on that sphere data from PLC.
ttystikk
Well-Known Member
Lol you have hit the nail on the head!
Everyone likes to brag about their light's performance at Tj=25C=77F, but who actually achieves it? Even active cooling is top of the grow tent ambient plus, and passive is hotter than that.
Show me the guy running his chips at 25C or below. I wanna meet him.
JorgeGonzales
Well-Known Member
But we aren't chasing performance numbers at 25C, we are trying to plot current droop curves.
So, yeah. And the only guys bragging are running Bridgelux. -ducks-
Seriously though Tj=Tc=25C makes it much harder to compare their numbers to everybody elses, and certainly causes confusion around here Christ knows. Luckily they perform well.
BOBBY_G
Well-Known Member
ttystikk
Well-Known Member
I mean this with the utmost respect; speak for yourself.
JorgeGonzales
Well-Known Member
Ahem https://www.rollitup.org/t/cob-efficiency-spreadsheets.865238/page-20#post-12684377
So now you are arguing with the guy who convinced you in the first place that Tc=Tj=25C is a silly metric. Like, last week.
And I was referring to Bobby's tests, because I know what he is trying to measure. Thus the "we".
JorgeGonzales
Well-Known Member
Anyway, for 50C, a good repeatable test would be Ta=Tj=Tc=50C, let temps equalize and do a quick pulsed measurement at different currents.
Measuring the case or Tj is much more difficult to do accurately than the ambient temperature.
Measuring the case or Tj is much more difficult to do accurately than the ambient temperature.
tomate
Well-Known Member
Won't disagree on that, you are absolutely right.
Then tell us how BOBBY_G can achive those steady state conditions at 50°C with different cobs at different amperages.
That would be far from easy.
And again, we don't need to measure under operating/real life conditions to make a simple comparsion. This would only make everything more complicated and harder to implement.
And you can still extrapolate the data using the case temperature charts from the data sheets.
Last edited:
bad assumption. just one example thermal resistance isn't linear.
ttystikk
Well-Known Member
Ulbricht Sphere.
Ok, last try. If you want to see PAR measurements under real life operating conditions, then you are wrong here.
Greengenes707
Well-Known Member
I have pulsed measurements of 25c and 50c for 3070 and 3590. I'll post when I get home to my computer today.
Hint...monitor you Tc when testing...modified holders.
Hint...monitor you Tc when testing...modified holders.
JorgeGonzales
Well-Known Member
It's not, but it's yet another variable linked to temperature, along with forward voltage, and best controlled.
Nice that you get this though. Heat sink discussions can be difficult around here. Wattage specs are arbitrary, and C/W discussions hard without knowing at what temp they are quoted at. Usually higher than we run.
But moving from 25C to 50C might result in a flux drop of 10%, and thermal resistance is a fraction of that, and the amount it varies by temp is a fraction of that.
JorgeGonzales
Well-Known Member
Also, I want to point out that what I'm talking about isn't theoretical for me. I've spent way too much time comparing different solid state devices to one another, to match this or that parameter, and if there's one thing I've learned is that it's a hell of a lot easier to normalize comparisons to your ambient temperature, than measure your Tc or Tj and try to normalize to that.
BOBBY_G
Well-Known Member
The citis use the anode as the measuring point. i was thinking of soldering a type k in with that connection
http://kicthermal.com/technology-information/conference-papers/261-a-comparison-of-methods-for-attaching-thermocouples-to-printed-circuit-boards-for-thermal-profiling-3/
JorgeGonzales
Well-Known Member
Just pulse everything at the same ambient temp like Supra. For manufacturer to manufacturer
comparisons, I mean. A real world test is definitely warranted as well.
Remember, and I should have said this before when people brought up 25C, the end game here isn't figurimg out PPFD/W at 12" and 25C, it's finding the current droop curve SO WE CAN HAVE REAL EFFICIENCY NUMBERS for these new cobs.
Broke out the caps yo.
ttystikk
Well-Known Member
If you want to use current droop data, why wouldn't you want to use temperature droop data as well, for the same reason?