Got the board with 3x 5000K Optisolis from Clemence. Well, this is it.
I’ve been unable to rigorously test the nuances of the tint, for I powered them off a desktop power supply and the setup is inconvenient for careful observations. But what I did observe was the lack of any tint shift across the beam whatsoever, up to 180deg off axis. While the e21 were already much better in this respect than domed nichias, the Optisolis are just perfect, different league. For me this alone would seal the deal.
So looks like this is a long overdue upgrade for the old king - the N083B
Having said that, the 6500K I got in the same package are of the cleanest tint among the e21s that I have accumulated. I have mixed and observed 3000K, 4000K, 4500K, 5000K, 5700K, and 6500K and among them I like the 3000K and 6500K the most. Next would be the 5700K. But the 6500K is the first 5000+ tint that I have ever so admired.
Thanks a lot Clemence. It ’s taken a lot of time to find the next gen perfect tint no shift LED and now you brought it here.
Most of the people like us who subscribed to this thread must be tint snobs. But now I just found a tint-shift snobs :laughing:
By the way, if you look my “coconut milk” pictures closer, you’ll see that this Optisolis also produce very consistent tint across the full current range. The E21A produces somewhat greenish tint at 700mA and turned pink at max.
In fact, if peripheral vision is important for you then the sharp edge of the beam is already a distraction and if there is also a color artifact there, it would itch a lot. I think that the problem here is that while the brain can adapt to a “wrong” tint pretty easily and it happens all the time every day, it would struggle when the eye sees both “wrong” and “reference” tints at the same time.
Hi Clemence, I wonder if an optisolis can be soldered to the +/- pads of the triple XP board that you are offering (provided the central pad is covered by a ting electrical insulator)?
It’s easier and more effective to insulate the base of the DTP board. Cooling properties of DTP board is already way too much for these 757 LED. But it will not if you insulate the thermal pad contact of the LED. This method suitable for parallel configuration only
Simply use any thin sticker (0,1-0,15mm) in the base of the MCPCB and add thermal paste to it. The sticker is there to separate the MCPCB from the base. Or you can use diamond/SiC/silica powder mixed in the thermal paste. The latter method requires you to carefully tighten the MCPCB as they’re sharp and possibly embedded in the soft MCPCB rather than separating it from the base.
If your LED shelf already anodized you don’t need any separator at the base. You only need to insulate the fastening screws (if there’s any) using dielectric washer (nylon, fiber glass, etc…).
Well, since the optisolis are low power leds to begin with, my plan was to use arctic alumina on the central pad. To put a thick film between the led and the pad first, solder the led, remove the film an pour the alumina to the slot. Or even glue the led and solder afterwards. When I messed with the epoxy potting in an e01 I learned that it restored its properties after being severely overheated.
So the question is if an optisolis geometrically fits the +\ pads of the board?
EDIT: was too slow to post :). Thank you for the info! I Always use wide frosted triples/quads everyehere, in an emissar d4 i even used an additional diffuser to that.
Thank you, that was very informative. I am trying to choose a very few optimal led/optics configurations. So far it’s been two: an aleph mule with the 3x optisolis with the h17f driver (for its luci2, not for fet) and an aspheric with 2x e21s and the rgbw driver as modded by djozz. Completed testing yesterday
Btw I am still stunned by that 2000k e21. Would have never expected that 2000k tint would become my favoite one
I happen to have a few of the aluminium dsch boards that MustSimon uses, never thought that they would be good for anything given that I have the DTP version too. Contemplating another Optisolis build….
Got my aleph LE with the 3x 5000K Optisolis and H17F driver. To say that I am very happy would be an huge understatement. This is the best tint I’ve ever seen and no tint shift at all. It is not only better than that of the 083B, I can easily get 10x lumens which is critical to me.
Having the three LEDs along the line is bit weird but I am going to apply the DC Fix. In a single LED mule the DC Fix would leave some irregularities due to the filter structure being comparable to the LED die in size, but here the 6 emitters would eventually produce smooth beam through the filter.
For night time, I plan to use a SF F04. I have one without the matte “glass” and I will put there a 5000K -> 3500K LEE filter and a black diaphragm with a modest hole, so that it lets out just a narrow warm beam.
We have stock of the Optisolis in 5000k if there is interest
one of the best spectrums I have seen
Also have 757 in 2000k and limited 100 pcs of NVSW219C in 5000k 90 CRI
Cheers
Mark
New product and we are building a new web site so lots of things not on the old will be on the new site, im guessing this board
yes we can do this in optisolis
Cheers
Mark
Looks like a meteor m43 would be a great host for an optisolis based mule. I think that one could easily put 30+ LEDs on this mcpcb as there is a lot of space where the paint could be scratched revealing suitably spaced copper plates.
The M43 provides 10+ amp to the LEDs in Turbo mode, where 30 Optisolis would be driven at 300ma max each with 3000lm output. My 3x optisolis aleph outputs 750 lm @3 amps, which makes a nice room sweeper. I can only imagine 4x of that. Will definitely do this mod as soon as optisolis are available again.
a tint shift snob/evangelist is here.
