I plan to add a low voltage cutoff to turn it off before the car is unstartable. I could add PWM dimming based on temperature, but it has enough heat dissipation to operate at 100% even in the worst case scenario (100F) weather, so I probably won’t bother. Maybe the next version.
Most definitely. It seems to run much cooler than my prototype tests on the kitchen floor. It was on for a couple of hours in ~90F when I took those night shots and I was surprised that it barely got warm to the touch. I’m going to borrow a large battery charger and test it on the car inside the garage so there’s no air movement and up the current as much as heat will allow. I’ll rerun the power with at least 10ga too.
So far so good. I saw Gore vents mentioned on somebody’s website, read Gore’s paper on the root cause of condensation, and bought one on eBay for the heck of it. Maybe it works, maybe it’s a placebo.
I picked the XP-G3 for the most lumens per watt, with the desired beam pattern with a 20mm Carclo lens. I spent a lot of time on Carclo’s website comparing beam patterns and LEDs. The XM-L2 is too fat. The XP-L HI is too expensive.
No, it draws about 25A, so probably no more than the rear window defroster.
You can definitely hear the idle change when switching it on an off, but that’s it.
The whole thing is 1143mm wide, or 45.0” - the exact width of the Pathfinder roof. The channel was made from 1.25”x2.5” rectangular tubing, .125” wall. After cutting off one side it is 57mm deep.
This is the heat sink: http://www.heatsinkusa.com/3-945/
Excellent job and results. I share your views on off the shelf units, and had wondered if anybody here had ventured into this realm for DIY. I’ve been researching these, as I can recess a 32” light bar in my Colorado’s front bumper, but everything seems sub par on output lumens/pattern vs price vs quality complaints.
The faceplate is made of acrylic, which is the generic name for Plexiglas. Everybody else uses polycarbonate (popular brand Lexan) which is stronger, but polycarbonate turns yellow in a couple of years. Acrylic does not yellow.
i was told by a big light manufacturer that PMMA (acrylic) could not take the heat.
and they said they treat polycarbonate so it doesn’t yellow.
are they blowing smoke?
why are most LED lenses made of PMMA if they can’t take the heat?
i’m about to have lenses made and i need to know which to use.
thanks
I’m not an expert, but Wikipedia says PMMA (acrylic) has a glass transition temperature of 105C and polycarbonate at 147C. So if your light is going to be hotter than boiling water, it could be an issue.
Auto headlights are made from polycarbonate and they have a coating to block the UV and keep them from yellowing. You’ve probably noticed how well that holds up.
So I finally got to perform a real-world heat test in the desert last month. It was dusk, 80 degrees F, no wind, and I remembered to bring my thermometer.
Ambient temp, 79.9F
I told my friends why I was running up to my car every two minutes and taking it’s temperature, but they weren’t really interested. They just sort of expect me to be doing weird things.
Starting temps normalized to 100 F/ 37.8 C, because my criteria was it would operate in 100 degree weather, parked, with no wind. The real world operating temperature was significantly lower than the prototype on the kitchen floor. A slight breeze came up around the 20 minute mark so I discount the data after that, but overall, I think I can take the advice of BLF members and increase the current from 2A to 3A.
I’m not going to change this light bar, however. I’m going to build a new one with Carclo 26.5mm optics instead of 20mm and use XP-G2 instead of XP-G3. I’ll have to increase the height of the bar from 1.25” to 1.5”, and reduce the number of LEDs from 45 to 36 to accommodate the larger lenses. I hate making the bar taller, but the vertical distribution is just too floody. I should complete a 6 LED proof of concept in January, and if the pattern meets expectations, start on El Jefe 2.0
4500K is the border between 4A/4D (below BBL, rosier) and between 4B/4C (above BBL, greener), so the common-as-dirt 4C would be perfect.
In my S2+ with a TIR lens, which mixes the light better and avoids the fried-egginess of reflectors, the 4C is beautiful. White with just a touch of warmth.
’Though, even my 3000K lights at night just make everything “pop”. Takes a bit getting used to, when everything and its grandmother is “cool white”, but once you do, you don’t want to go back. Now, even my 5000K lights look “cool”.
Ok, I don’t have any pointers. I will however wholeheartedly argue that you are doing something wrong!
I have on several occasions in talking with coworkers and acquaintances referred them to this very thread when they start talking light bars. This project and your dedication to excellence is inspiring and a beautiful thing to behold. It’s things like this that make BLF what it is. Well…there is more of course but to watch a DIY project go from start to finish with such outstanding results is very fun for a geek like me. :sunglasses:
For a proper outdoor light experience I’m right there with you on tint preference. This is of course highly subjective but in my opinion (which I completely agree with), a hair under 5000K just brings the world to life. :+1:
I have a few lights from my buddy Simon in 4C and I love that tint. They’re older XM-L2s but my tinted opinion still applies.
Thanks for the kind words. I do tend to obsess over details. The blacked-out grill on my car, for example. I painted it five times before I was happy wth it. My fingers were raw from sanding it down between paint jobs, but now it’s perfect. Anyway, I’m sure El Jefe Version 2 will go smoothly wth no hiccups…
Not to be ‘that guy’ here, but since you mentioned the main reason you plan on going with XP-G2s in V2.0 is for a tighter beam. Considering the timing, I might drop the suggestion of experimenting with the new XD16s. They should reduce the FWHM another full step from the G2 and have similar lm/LED. Additionally, the tint consistency is supposedly dramatically improved along with efficiency through secondary optics. Also, the Vf will stay inline with the G3. That may or may not be a benefit for your driver efficiency.
If being “that guy” means offering insightful advice to address my issues - then thank you for being that guy!
I missed the threads on the XD16, so I just did a bunch of catchup reading. I think you’re right, it would be a very good fit for this particular application, addressing the problem of too much (many?) lux on the ground in the first 15 meters in front of the car, while sticking with 20mm optics, which would be much preferred to dedoming a bunch of XP-G2s, or a complete redesign with 26.5mm optics.
I will now hold my breath until the XD16 starts shipping.
