I would like to show you my latest custom product. A GU10 spotlight, active power 4W. It is based on Bridgelux THRIVE 98 CRI 2835 LEDs (the light spectrum is as close as possible to the sun, without excess of blue light). LEDs datasheet
Housing made of aluminium, color black. Optics textured reflector. Wide angle of total emission approx. 120 * Covering glass lens made of quartz glass
A modified Dark Energy power supply, high efficiency of ~90%, no flickering. AC input voltage range 85 - 264 V AC freq: 50 - 60Hz Main 450V filtering capacitor changed to KEMET ESL series (lifespan 10000h at 105*C). For comparison, good quality bulbs PSBs usually use 3000h capacitors. Added fast fuse for safety.
Available CCTs: 2700,3000,3500,4000,5000,5700K.
I did not count it yet. Current regulation is set exactly 100mA for 4pcs LED in series. Below is light parameters for each CCT. Final luminous flux i 4x the single value. Ext for 4000K it is 4x 96lm = 384lm@85C BXEN is mid-power LED, max current according to DS is 120mA
There are no white LEDs without any spikes of base emitter. In real measurements there are always differences (comparing to datasheet). Below theoretical spectrum diagrams for Sunlike, THRIVE and Citizen.
The Thrive spectrum shown here does not exhibit the huge 450nm spike that is present in your spectrum measurement from earlier. This difference is more than can be explained by sample variance. Any chance you were sent wrong emitters?
I was also surprised by that. I don’t know exact condition of this measurement. It was made by my customer and I suppose there could be some additional light from other LEd source indicated in that blue spike. THRIVE emitters I use bought from official Bridgelux distribution.
Here’s TM-30 for a 3000K Thrive 1W 2835 emitter I measured a while back. It still has a bit of a blue spike, but seems more in line with the datasheet.