Led4power.com : LD-4 CC linear drivers, ILC-0/1 illuminated tailcaps, optics, MOSX, copper DTP MCPCBs...

Thanks kiriba-ru. I often miss those conversations because I don’t mod much. Plus, my memory isn’t very reliable. So it’s probably true that it has been “discussed hundred times” and I missed it or forgot it.

Example of fully populated 3XV-C mosX board:

Parts needed(for LD-B4):

First two "components" on the left are two small 0.6mm FR4 boards which are actual "LED" wire pads.

mosX boards are very thermally conductive everywhere which makes soldering wires directly to board very challenging, it's possible (but not easy) on 20mm size boards, but these boards are >30mm - soldering wires directly would be a nightmare. So these thin FR4 plates serve as very good thermal insulator - soldering wires should be easier than on standard copper DTP boards because those usually have 0.2mm FR4 thick laminate on surface.

For same reason there is small JST smd connector on board for signal wires (gate,NTC,moon) - in theory I could make 3rd FR4 plate for those, but I think connectors are simpler to use for end user, especially if wires with male connector are already soldered on LD-B4, in that case build is as easy as with any other normal 2-wire driver.

All parts placed:

Soldered, everything fits nicely:

Just tried it outside of flashlight with bench power supply, tint is absolutely fantastic (still can't believe it's only 70CRI,looks better than some 90CRI LEDs) complete uniformity across the whole beam, and brightness looks already impressive at just 9Amps. I think I will put this board in Kiriba-ru C8 head and drive it at 15-18Amps.

Seems very clever! Will the FR4 boards tolerate 25-35A? I want to use three parallel Luxeon MZs in direct drive (for short perdiods of time). I will also use the Kiriba-ru C8 head.

Wow! that looks very nice and professional!

There are many vias so electrical resistance is very low, that's really last part of whole flashlight circuit you should be worried about, springs will be the bottleneck for sure,even bypassed ones if not done properly.

Ok, good! I’m worried about everything! :wink:
Currently “working” on a suitable FET-switch with loneoceans, fady and lexel. Will I be able to route leads suitable for 30A through the hole in the middle of your new pcbs?

Springs are an important point. I want high spring pressure and low resistance. Maybe a multiple leads will be used for the bypass and the top of the spring might be covered with a copper or brass disc (gold plated).

LD-B4 driver has holes for up to 18AWG wires max., this is good enough for 30+Amps, there is plenty of space for wires in spacers. You don’t have to worry about switch :wink:

Soldered all variations(Luxeon V 4000K, LH351D 4000K 90CRI) just to check how everything fits, also tested them with bench power supply.

6x variants definitely put a lot of light at just ~10A, beam looks good with all optics, clear optics have some tint separation like usual, frosted cute versions bled light completely.

now we just need those spacers. :innocent: I plan to mix one of the frosted Cute4’s with 4000K and 5000K 90CRI LEDs if they blend well like you mention.

Demo build1 -C8 6xLH351D 4000K 90CRI (still not 100% finished):

This is how spacer will look like,a little unconventional and more complicated, but it has several advantages - build is easier once spacer is fixed and can't move, spacer is compressed with high pressure, and all wires can be soldered on driver first which is usually much easier. Hole in C8 should be ~8-8.5mm.

Driver LD-B4 18Amp for this build:

Those FR4 0.6mm thermal insulating boards are working as expected, soldering thick wires on them is very easy.

Moon mode:

I used common 1288 Omten switch and 30Q cell,so I didn't test light under full potential yet, but it's damn bright for 90CRI and it's size. Brightest 90CRI C8? Hm maybe even brightest single cell 90CRI light?

Of course big die dome led and relatively small TIR size (but much bigger than carlco optics) makes this light flooder, but it still can burn paper:)

And relatively big ALU spacer definitely works well as thermal storage because light didn't get crazy hot very fast, but it also took more time to cool down. If you are going to use this kind of light in <1min pulses with >1min pause,it is beneficial.

Only aluminium? Will it be also available in copper?

You don’t want that chunk in copper

And it will not help a lot for the heat path either.

Great work! You’re really churning stuff out lately! I’d like to see what the maximum potential for this build is. djozz? :smiling_imp:

6x Luxeon V or 5700K 90CRI MZ will be more potent,351D is not best LED for max. output.

I wonder if this spacer will fit on the manta ray c8.2? Or in the future do you plan on making one if doesn’t?

PS. Btw, I forgot to add great looking spacer.

I got manta ray and it doesn't fit, it's too wide, don't know about height.

The driver have to be 22mm, isn’t it?
Is only the driver hole too wide or what is too wide?
Is the spacer too wide for the head?

Please explain the switch part ;). I don’t want to needlessy spend money on this if you are already developing a more suitable solution compared to what I have in mind.

I just want a FET-clicky switch with a coin cell battery or a capacitor.

I do not care about the weight. My point is that copper can be soldered. Then the led board does not move and it’s perfectly centered. And that’s always very important for me when I make multi-led.

Below is an example of what I mean:

I would think the switch could be a simple momentary FET Driver with only one mode to place in every Flashlight instead of the clicky-board.

Perhaps it has LEDs for iluminated tailcap and a bleeder resistor on the main Driver is needed. And this switch can be used as upgrade in each clicky Flashlight.

Perhaps short klick for mode Change, as reverse clickies behaves and Long click for power off.

Thats what I would think.