Quick guess - Cometa has a low temp threshold, and the Osram is lower Vf than a XPL HI, so, the amps goes way up - probably too much for the LED anyway, and therefore the heat jumps way up fast, tripping the temp threshold which the XPL HI either can't do or takes longer to do.
Other guess: I have had problems with the 2mm white flat in combination with FET+7135 drivers (like in the Cometa). Perhaps it has something to do with the extreme low voltage (lower than any white led before). It may even have killed the 7135 chip but that may also have been accidental.
Having problems with 7135s or linears? Switching full regulation is a O:) way. You may also want to try a low current cell or avoid charging cells to the max. Try and measure first, avoid problems.
Did anyone measured an effective max current value for the CSLPM1.TG in a C8?
You must see it as this: in a reflector light every spot on the light source (the die) is projected smeared out in a circle around the center of the hotspot. So the center of the hotspot receives light from the center of the die, and any circle at a given distance from the center of the hotspot receives the light that comes from the corresponding circle around the center of the die, but smeared/evened out. So the shape of the light source is irrelevant, it will be projected as an sum of evenly illuminated “circles”. This is sort of what happens to the hotspot, but the corona can do things a bit different it seems that in contrast is a bit more dependent on the die shape, like showing “petals” when the die is square.
This is the best I can explain this with the limited english that I know, explaining technical things in a non-native language is pretty challenging.
Yeah, that’s the idea…
I burnt some paracord and I placed the washer right in the black fumes to allow a thick coating to depose on it.
My wife hated me for doing that in the living room of course !
I modded my Emisar D4 with 2 1mm^2 and 2 2mm^2 white flats.
These emitters’ voltages make them perfect to put in parallel with each other. Emitters in parallel always have the same forward voltage, so one can predict the relative currents in each emitter by looking at their voltage vs current plots. From djozz’s measurements: if, say, 6A goes through the 2mm^2 LED then 3A will flow through the 1mm^2 one.
It took me a couple tries to get all 4 emitters centered with no shorts. It pulls 19.5A with a VTC5A measured using a clamp meter. I measured 46kcd.
The Cometa takes a standard 17mm driver, so the L4P LD-4A fits without modification
The sweet spot for the 1mm² WF is 4,5A while the 2mm² version should be around 7A.
I’ve swapped a 2mm² in mine 2 days ago and adjusted the current to 6A and get a nice 220kCd and a funny wide screen beam.
Dang man! I never got a Cometa because I already had 3 Jaxman Z1’s, but they take a 22mm driver. I need to find a 22mm driver set at 4.5A with a bleeder for an illumination tail cap. Maybe Led4Power will come out with a larger diameter driver…
From what I see in the published data by luminarium iaculator, CSLPM1.TG maxes at ≈10A. Considering the CSLNM1.TG maxed at ≈5.75A in led4power's tests, for the same relationship the 2mm² could be set at 10A × (4.5 / 5.75) = ≈7.826A.