traxtar928

Agreed... great work here. I'm also looking for a vent to use with my design. I'm interested in seeing how it interfaces with the headlight and fits inside the headlight cover. Space is at a premium, and I predict that you'll have to rethink the method you've used to secure the ballast.

Ed Scherer

Yeah, we'll see. Should know by this evening. The wing for mounting the driver isn't really critical. If it won't fit, I'll rework that detail as appropriate.

Glen McCartney

Ed, agree very nice solution. I'd be interested if you are planning on offering these for sale.

Ed Scherer

Well, it almost worked. There were a few minor fitting problems. One I was pretty much expecting: the vent socket hits the handles on the H4 lamp retaining clip. For symmetry reasons (don't want to have different left/right units), I don't want to move the vent.

It's also a little too tight on the top and bottom. Also, the headlight cover almost fit back on, but not quite.

I've done a redesign, fixing all the problems I noted. I'll fire up the 'Bot shortly to create another prototype. This next prototype is a little sexier (and shrunk down a bit) than the first prototype: more curves. Doesn't look quite so much like a disk drive.

Ed Scherer

Thanks for the positive feedback, guys.

Yeah, assuming they work out OK. I also need to see if ABS will work out. I may need to switch to another kind of plastic (Delrin, maybe) if ABS can't take the heat.


Yeah... The one I'm working with now is supposedly a 4000 lumen (for either low beam, high beam, I guess). My previous one is 2000 (low beam) / 2600 (high beam), and it seemed to throw plenty of light.

Ed Scherer

BTW, my current design is for exactly the kind of LED bulb as I showed back in post #15. There must be a removable heat sink on the LED bulb for this design to work.

The barrel between the heat sink and the rest of the bulb is ⌀ 22 mm. The one I have has a slight chamfer toward the heat sink end; that's very helpful for getting it through the cap without dislodging the O-ring. The hole in the cap is ⌀ 24 mm; the O-ring (which sits in a slot) takes care of the gap. It probably wouldn't be that hard to produce different caps if there are other sizes that are needed, but I'd rather avoid that if possible.

Ed Scherer

Here's the second prototype. I'm going to have to wait until tomorrow to try it out. One of the slots on the back suffered some damage when I screwed up one of the toolpaths. Doesn't affect functionality, but looks crummy. Then again, so does the melted ABS around some of the screw holes. Had a little registration (alignment issue) from the front to the back; probably about 0.5 mm shift (it looks like a mold seam).

Ed Scherer

The prototype shown in the previous post is much better. Seals perfectly and can be oriented either "wing side left" or "wing side right", so the symmetry worked out as I hoped.

Since I now have a version that seals well, I have turned my attention to the details of the LED driver module mounting and wiring. I've played around with a few orientations and think I've got something that will work well and will be way more compact than my first attempt.

Prototype #3 should be very similar to #2, but will have some details changed on the wing.

traxtar928

What's the thread/pitch on those vents? Care to sell a few of them? Wouldn't a 1mm hole in the plate do the same thing? We don't need IP67 water tight capability when the sealing method is already compromised by the o-ring.

hacker-pschorr

This is so awesome, nice work Ed!!!!

Ed Scherer

From post #29:
There's more info about the specific one I'm using in post #26:



No; I'm planning on using all mine, but I think there are some of the above available right now on eBay. If you don't get them, I might.




No, IMHO (well, probably more than MHO). Read post #25 and watch the video.

If you just put a hole there (or don't seal the bulb), you'll get air pumped out when the headlight is on and heats up and air sucked back in when the headlight is turned off and cooled down. That air will contain contaminants. I watched my unsealed headlights (with HIDs at that time) turn cloudy over a few years as I noted in the opening post in this thread. We've all seen the same effect on all sorts of stuff around us: drinking glasses that sit inside cabinets become foggy over time, glass light fixtures become foggy, the interior surfaces of window panes become foggy, etc. Contaminants are just deposited on all surfaces. We just notice it on glass (and other smooth surfaces) more than other stuff.

That's the whole purpose of these vents: they pass air, but not the contaminants.

I care about it enough (as one who likes clean stuff, but not wasting time on cleaning!) that I'm willing to pay the $5-$6 each (hope I can find them cheaper eventually) to avoid the problem.

Note that the original caps have a membrane vent. The membrane is visible under the round, trifurcated metal retainer on the right side of the cap:






Thanks! It will be more awesome when I'm finally done.

jcorenman

Ed, that is outstanding! Clearance details aside, it looks perfect!

Huh? How does the o-ring compromise the sealing?

Personally I think the membrane vent is required-- the housing needs to breath. Around here in the Pacific Northwet no one slows down for rain and you find yourself driving underwater with a little air mixed in.

A question for anyone who has had these lamps fired up: Normally a H4 lights one filament for low beams, and the other for high beams. The only time both filaments are on is pulling the hi/low lever (i.e. blinking), or if modified with diodes and a relay. Do these work the same? Or do all of the LED chips light up for high beams? Can you run both hi and low beam LEDs without melting them?

We also live where it gets dark and Bambi likes to play at night, so more lumens is always better. Glare is not an issue with H4 housings, clean and properly aimed (we have motors on ours).

Ed Scherer

Thanks, Jim; not perfect yet, though, nor will it ever be. When I get to 99%, I'll stop, make a bunch, and notify you. I've learned to be careful about striving for that last 1%.

I knew the answer to that for my previous LED H4s at one time, but have since forgotten. As I'll be experimenting with both the old LEDs and newest ones (that I've been designing the caps for), I'll try to finally record my findings and report them here.

I'd also like to make sure that the H4 LEDs operate as they were designed. If they're not designed to have both high and low beam LEDs operating concurrently (except perhaps momentarily as you noted), I'd like to make sure they work that way in the 928. I don't want to overheat them since heat is the biggest enemy of LEDs.

Ed Scherer

Jim, your comment about adjusters came at a really good time... because it's causing me to think a lot about:

• optimal, safe placement of the LED driver module;
• how and where wires need to be free and where they need to be secured; and
• making sure that the sealing cap doesn't hit the headlight supporting arm (the "U" shaped bracket on which the headlight assembly pivots) under any circumstances.

Most of us (I think?) , including me, don't have remote adjustable headlights; I had actually forgotten about that feature. We typically adjust our headlights manually through a very small range. But I need to make sure that I don't design something that won't work (or worse yet, breaks something) when features like the remote adjusters are present.

I've been reading various threads (not just on Rennlist 928 Forum) about the adjusters, hoping to find out more about the range of motion they will produce. I have yet to find any quantitative details.

I've already taken a careful look at what moves and what doesn't and examined the potential problem areas. For example, the area under the sealing cap is a "high risk zone" for possible pinching due to headlight adjustment. I just need to understand it better. Everything above that (above the pivot point on the headlight supporting arm) is quite safe, as long as it fits under the headlight cover.

What I'd appreciate from you, Jim, or anybody else who has or knows about the electric or hydraulic adjusters is answers to the following questions:

1. Can you tell me what the range of motion is, especially with the headlight cover off?
2. Would you be able to get that information if you don't already know?

I might eventually ask for a volunteer to try out a set of prototype caps on a car with adjusters (and you'd need to have or get a set of H4 LEDs like the ones these caps are designed for) just to makes sure that everything is OK.


The area I'm especially concerned about is the lower corners of the cap; they're close to the headlight supporting arm, and that gap will change as the headlight is adjusted. Aiming the headlight higher might make the cap hit the supporting arm. (And also worth considering is that aiming the headlight lower could strain the headlight cable if slack is insufficient).

jcorenman

Ed, I believe that the adjusters (fluid or electric) were standard on all cars that had H4 headlights, which is to say all of the ROWs and none of the US/Canada cars. Ours was a retrofit per info here from Alan. The fixed support rod at the bottom gets replaced by an the adjuster motor with its rod.

The adjustment range is not a lot, maybe half an inch extension of the rod but I'll have to measure it to be sure. I also think the"neutral" position of the adjuster rod is shorter than the fixed rod, compensated for by headlight alignment.

One option is to remove the fixed rod so you can swing the reflector assembly up and down. That will tell you where the interference is, and if it only happens at a silly angle then not a problem. Or send a prototype or mock-up and I am happy to try it here.

Cheers,