Prototype off-road/boating spotlight - King COB!

I would check that, doesn’t the Ledil reflectors sit atop the Ideal clamp?

Finally got my 900W DC booster. I ran two 3s2p 5200mAh packs in series, and was able to take the Vero29 to 37.85V and 3.28A… for an estimated 17500lm at about 140lm/W. I ran the rig for about 10 minutes, and the voltage on the fresh set of batteries dropped to 3.95V. So I’m thinking at 17500lm I can potentially have about 40-50 minutes of run time?

So far I’m pretty happy with this boost driver, being CV and CC and fully adjustable. and programmable. I kind of like having the convenience of a pot to adjust output, but the down/up buttons work fine.

Next step - design a connecting plate to hold the CoB, lens, and housing on one side and heatsink/fan on the other, with a housing for the electronics and batteries.

I replaced the voltage regulator and the driver seems to work but it shuts off the output at 1.3A and just shy of 35v.

I haven’t received a response from StratusLEDs.

UPDATE: Have the adapter plate made, and BS’ed together with two threaded rod I had laying around. The CoB sits in a pocket. Gaskets were provided for the TIR and for the interface between housing and base. I will make a new base with a deeper pocket for the COB, so I can eliminate the gasket, and use thermal paste at the juncture to allow the housing to aid in cooling.

I got a bit brave, running without a fan, and got my test rig to 39.9V and 4.2A. So I was able to reach max current at slightly under the forward voltage on the chart in Bridgelux’s datasheets. Their charts also show that at 4.2A the lumens should be around 170% of test current, which is 2.1A. So theoretically I should have over 20000 lumens off the CoB. No way to test this but it’s so bright, I had to wear shades while adjusting current and voltage.

Here’s my test area. Unfortunately because of rain, I could not go outside to test this. I was under the canopy by the door.

And the area lit up. The beam is definitely directional, with a very gradual tapering of brightness going to the perimeter of the beam. Can’t see it here because I’m holding the hot light with one hand and my cellphone with the other.

And a pic of the beam spot from 1 foot away. The spot was so bright I couldn’t look at it directly. The picture does not convey that unfortunately.

Looks nice. :slight_smile:

Well, I have no idea how you were getting 42v out of your StratusLED driver since the only “official” documentation claims it is 34v out MAX which, I think is why mine is shutting down at 35v. I suspect they have it programmed for output over voltage protection at 35v. If I could get a copy of the datasheet for the AP80SY chip, then I could probably swap some resistors to enable higher output voltages and probably also fix their 6A/3.5A mixup. Sadly, I haven’t heard a thing from ShiWei Semiconductor and google translate isn’t up to the task.

I may try to reach out to Amazon to see if I can get a refund since mine blew up. Annoyingly, Amazon isn’t allowing me to post a negative review of the StratusLED driver even though I have a photo of faulty component.

I have no clue too… But tje other two I have is behaving similarly to yours. I’m at a loss.

Can you look carefully at yours to see if there are any differences? I have a 4-40x microscope and a macro adapter for my camera so it’s easy to work on and photograph stuff like this. I really should get the 0.7x barlow lens for my microscope since it seems I’m going to be doing more surface mount work…

I have to take the other one out of the enclosure. Thought maybe the pots were somehow bad but they both read close to 10kOhm. The resistor mod does zip. I was able to run 2 Vero29s in parallel daisy chained with the pot, but I added 4MOhm of resistors at the pot LOLOL… I’m thinking adding a resistor in parallel with the LED maybe?

Update: Just tested with my multimeter in series with the LED, with the Vero29, I was able to turn the dial on the pot up to 900mA. When I checked the voltage across the terminals I got 43.2V?! Wha?

Thinking back, that driver board was in a batch of 3 I got back when I started this thread. Wondering if I just got “lucky” with one board, or possibly it has something to do with the CREE versus the Bridgelux CoB.

Speaking of which, I thought about what you mentioned earlier and picked up the 68V version. 18,500lm @ 1.7A. I think I should be able to get it to 2.4A with my 6s battery setup.

Redid the adapter plate, now fit to the IceLED heatsink…

Added brass inserts for the LED screw mounting holes:

With the LED installed:

The Lenina reflectors seem like a really good option for a compact Vero29 setup.

Why brass inserts? Why not just thread the aluminum directly?

LOL… picked up the wrong drill bit for M3 tap…

I was thinking about the Battle Lantern more this morning and I definitely think these Lenina reflectors and a Vero29 are the solution. Since LEDiL makes lenses for these that snap onto the front of the reflector, it allows me to make a much more compact stackup and the smaller OD gives me more face area for surface area increasing features to improve thermals. The shorter depth also gives me a shorter thermal path to the front.

So, I guess I need to order a Vero29. I’m also thinking that since this light will be thermally constrained, I might as well go for a high CRI part as there really isn’t a huge difference in lumens when under-driving them.

For the lumen bazooka, I’m still looking into designing my own driver and I’ve found a few chips that I think are up to the task of pushing a 72v Vero29 to full output.
.
.

LEDiL has a different base for Lena/Lenina reflectors… C15083 which is for Vero29 AND Vero29se! It is also much more compact than the giant C13867 which only works with Vero29.

If you design one I’d be willing to buy a few from you LOLOL.

I set up my Vero29 with my 6s2p battery config, and was able to reach 69.9V @ 2.4A. Way brighter than the previous setup. Glad I got the IceLED cooler because this thing does run a bit hotter.

Yeah, trying to keep something dumping ~200w cool will not be all that easy. I was playing around in CAD with using 3 cheap heatpipe tower coolers in a triangle with a single fan pulling air in the center since I know that once airflow goes above, ~5mph, there isn’t much additional cooling to be had. So, you’re better off adding more surface area and heatpipes if you have room.

Yes… That’s why i increased the depth of the pocket, so i can mount the housing directly to the adapter plate I made. Preliminary tests say it’s enough but that’s just by touch lolol. That said I’m not sure I believe IceLEDs claims of this suppirting up to 400W LEDs. Becausr if it can I might have to use it on my CPU lolol.

They’re probably assuming a steady state junction temp of 85C where the heatsink is up around 60C…
.
.
One of my goals is to keep the COB…COLD.

Likely… but since I have it running at 168W, I think it should be fine. Having a 1/4” aluminum plate and the housing (which has to weigh about 3 pounds) should help a lot. Maybe once I get my rotary axis up and running I’ll make a new heatsink housing. But for now, this is fast and effective.

I do have another solution but saving that for my CFT-90 project…

Well, no new CAD models from LEDiL despite e-mailing them this morning. :frowning: I suppose I could just measure some key dimensions on the Leninas I have and make some progress with that…
.
Allegedly, my L6 will be delivered tomorrow so I guess I might waste some time disassembling that and figuring out what sort of stupid I can do with it. :smiley: