This is very interesting: https://betterautomotivelighting.com...ght-bulbs-get/

I have some DDM Tuning H7s in my stock 996 headlight housings. I have a digital thermometer probe inside the housing. After 30 minutes of driving, the temp inside the housing (behind the actual reflective enclosure, but inside the plastic housing) was 56.7C or 134F when ambient air temps were around 75F. I stopped after 30 minutes, but the temps were still rising at that point. I wonder what I should allow as an upper limit on the temps? According to the article above, the halogens get pretty darned hot, and that's what these enclosures were designed for. Thoughts?

I modified the back door of the housing (the removable snap-close door) to add two air circulation holes, and a small 12V PC fan. The fenders which surround the headlight assembly were warm to the touch, but not actually hot. I didnt measure temp, but most people feel pain somewhere around 120F in their fingertips, so I suspect this was much lower, probably 105F or so. The temps noted above were measured WITH the little 12V blower rig installed and operating. I think the air inside the fender, above the fenderwell liner, is static and doesnt circulate much, so the little blower is just circulating a static, closed airspace inside the fender.

I really *REALLY* like these LEDs over the halogens. I can spend a lot of $$ tinkering with this before I will get anywhere even close to the gazigabuck costs of the Litronics. Plus its fun and I get to exercise my thermodynamics gene.

I'm looking at the PC liquid cooling rigs as the next step. The radiators are available in many different sizes, and I think I can get one to fit into the enclosure behind the LED. May have to modify the enclosure, dunno, but I have access to a 3D digital scanner and printer, so maybe I can make my own enclosure door.. But these water cooling rigs might do the trick to get the temps down for our wonderful Appalachian mountain roads at night.

 

Your radiators are directly below the headlights..
The LED's do make some heat, but compared to halogens not near as much,,
the LED bulbs I use in my Honda fit wont keep the ice off the lenses in the winter while driving.

 

As the cited article notes, the halogens are very hot in front, where the filament is, compared to the LEDs. The LEDs tend to be hot where the driver is located, ie behind the bulb.

 

Just pointing out you have a 200 degree heat source directly below the light,,
what I wondered was how much light are you trying to get to? because anything
above about 55 watts can screw up oncoming drivers night vision long enough to make things dicey..

I read the article,, The type of measurements in the article don't tell the whole story.
IR thermometers can be too isolating on what your reading, its easy to get much hotter
pinpoint readings than are really transferring to the surroundings..
Folks have the same issue with thermal cameras, there is way more to it than just pointing and getting a reading.
A flir cam with the right setup would show the radiant energy being transferred to the housing. Give you
a better wide angle view of how the housing/assembly is doing. You can rent a pretty good one for a couple hundred bucks a week.
FWIW ,, I commission large data center buildings and take hundred of thousands of heat readings, with all kinds of
thermal measurement gear, and every time I find something new.. Different part numbers all kids of variation there is a lot of
variation in materials when taking readings.

Comparitively The LED's I'm using don't have heat syncs on the rear, they use an aluminum shaft holding the LED elements inside the housing,

In the article they are identifying a 4000Lm LED, which can be substantially higher light power output than a 55W Halogen.
And focus type makes huge difference on how much heat passes through the lens face,, That's one of the reasons a lot of
HID kits don't work in stock Halogen housings, the focal length is out of spec. Most reflectors are designed to place the center
focus beyond the lens of the headlight.. The type of emitter is also huge in how much heat is generated.

No knocking the direction your going, I added heat syncs to some rally lights to help keep the housings from melting..

 

A worthwhile point. I may put the halogens back in and measure with them to compare.

Well....watts are a measure of power consumption, not light output. Granted, it generally mapped to light output in incandescent, but I dont believe it does in LEDs. Lumens is a far better measure of light output than watts. A 55 watt incandescent will be poorer than a 55 watt halogen, and for HIDs, watts makes almost no sense at all. The LEDs I'm running now are 50 watts per the mfg site, but more importantly they are about 4000 lumens. And both the LEDs and the HIDs drew less current than the halogens. Watts is power, Watts = voltage X amps, and the LEDs draw far lower amps than the incandescent, yet produce much more light. 55watt halogens were drawing close to 10 amps each, while the LEDs at 50watts are drawing about 7. The HIDs would draw about the same as the LEDs, but they seemed to have a very high current startup (I dont have gear to measure that), and also seemed to draw more as they age.

Agreed, the devil is always in the details. But the article is at least somewhat unique in dealing with the issue of LED temps in auto headlights.

On another vehicle of mine, I have tried halogens and (so far) two different types of LEDs. I have run into exactly what you are pointing out...the various brands of LEDs do not necessarily have the emitter located correctly. In fact one of the brands I tried had the emitter rotated slightly which caused horrible scattering of the cutoff. It was not adjustable. EVERYONE was flashing me. Different brand, no other changes, the cutoff is very distinct and very few people flashing me.

 

Some of the bulbs you can rotate the core separate from the base, Thats teh type I'm using, makes it so you can get the best compromise in light pattern and cutoff height..