Man thatās unfortunate. Really wish thereās a way you can choose the tint bin from Samsung instead of receiving these random greenish ones. This emitter has so much potential. Thanks for testing.
Chiming in after slicing the dome off two more LH351D 4000K 90CRI leds to report that the slicing results are very consistent: the tint ends up around 3500K and below the BBL. Judged by eye it is an extremely good rosy tint.
For slicing, this time I used a brand new Wilkinson oldschool razorblade, in three slices: first the top of the dome, then close to the final height, then the last thin slice to just above the phosfor. No sanding.
These are measurements of the hotspot of my Convoy T2 now with sliced LH351D 4000K 90CRI. (FET driver, purple Efest 14500, 860lm at start, 710 lumen at 30 seconds) :
Thats great news!
Did you also measure before shaving?
The LH351D is usually rosy at high currents. At low currents is is usually yellow, above the BBL. Can you maybe measure one at different, fixed current with your power supply before and after shaving? Maybe also include the Luminus SST-20 4000K 95CRI. That would be really nice.
I have never seen any measurements showing the LH351D being rosy. All the data Iāve seen shows it at above the BBL at high currents. Djozzās shaved dome measurements are the first to show the LH351D 4000k can be made to go below the BBL.
Personally being a tint snob, I canāt stand the yellow green on the LH351D. It is similar to the Nichia 219C, which I never liked. Once you compare it with the E21A, 219B 9080 or Cree 5A or 5D tint, you will see how green the LH351D is.
BlueSwordM, thatās interesting. Thatās not what Iāve observed. Maybe there are certain batches on the market with below the BBL tint but I havenāt seen any. Can you post a test showing the negative duv? That would be beneficial to alot of people here who like the performance of this emitter but donāt like the above the BBL tint.
Itās at least right on it at 3A (Maukka posted his measurements here). The tint change is very obvious as you increase/decrease the current.
Djozzās measurement of a single shaved LED only tells us what is possible with some samples of this LED. It doesnāt show us if there is an effect (tint shift) that can be expected with all samples.
Aha, three stages slicing.
Thanks for the tip. :+1:
I can confirm though, that the tint of this LED gets warmer and seemingly more to the red side when you slice the dome.
Very nice result here too. Bit of a messy slice though, but under a TIR you wonāt notice that.
I've sliced some XHP50A emitters, at ā1.1mm without sanding too as it kills the bond wires rather easily. In 3 or 4 slice bites from different angles. Of course ends up looking a bit messy, but that is unnoticeable even below an aspheric at focal lenght distance ābatsignalā :-D mode. I do notice slight barrel distortion at the emitter projection edges and dome remainings, but looks good other than that. I like the result of this method, there's noticeable intensity gain and probably the output loss is not as dramatic as with more radical dedoming ways.