after doing some measurements... the d2s capsule needs to be moved further into the h9 90mm projectors by about 2.8mm. at the same time, the length of the arc is about 0.6mm shorter than the h9 filament. that being said... i'm gonna try and come up with a mounting plate that locates the capsule further into the housing. gonna see if i can add some adjustability for it to range from 3.1mm to 2.5mm for focusing purposes (the range co-relates with the ends of the h9 filament).
by the way Herman, another question. do you happen to know which direction the h9 filament is offset toward once its mounted into the projector.
This will move bulb in about 2mm (or however thick your backplate is) for a more ideal filament position.
The offset inside the H9 bulb, if any, is upward, so you would have to move bulb down. More than 1.5mm would not makse sense.
er... i'm confused now. if the filament in the h9 bulb is offset upward by about 1.5mm when in the upright position (connector portion of the bulb facing vertically down). shouldn't i be offsetting the capsule 1.5mm upward within the projector?
I probably use a different reference point than u, Byron. The important thing is to move HID bulb down about 1mm. This will make more light below cutoff. The pics I have seen of H9 Hella 90mm beam pattern with bulb totally centered is that hot spot is 95% hidden behind shield. Mobing HID bulb down 1mm (maybe 1.5mm is neded) will likely make hotspot 85% hidden, which is alot better.
ok... i'm gonna try both ways and see what happens to the beam. by the way, i disassembled the projector from my headlamp and 90mm's front glass housing... hooked it up and test to see how much i'm loosing in intensity. it was noticable! the colour at the cutoff was much more vivid and sharper with the projector "naked".
ok.. just tested 2 prototype d2s holders that places it further into the housing. the intensity at the edges of the beam is brighter than the original setup i had. as you have predicted herman, place the capsule 1.5mm down yielded a better beam intensity than having it 1.5mm up. the difference between the 2 is huge.
i have it further in than the base plate. its about 2.5mm further in from the outside surface of the base plate. and the width and intensity is definitely better than when i had it right on the base plate.
edit - the base plate is about 2mm thick.. so that means i have it about .5mm further in. not sure if .5mm would yield a noticable difference.
ok herman, here's some shots. do pardon me for the sloppy outdoor pic of the beam, it was windy and snowing... lol my whole engine bay was covered with about 2mm of snow during the pic shots (about 10mins).
some pix of the holder.
assembled holder on hella 90
2 piece d2s holder to fit directly onto hella 90
initial setup - i'm using my spare d2r capsule in this pic
initial setup - projector outside of headlamp and front glass
new d2s holder - projecto outside of headlamp and front glass
you will notice that the newer version has less foreground lighting. also notice on the newer version, the right side of the beam extends further than the initial (which as that weird step down, like as if it was fuzzy vertical cutoff).
actually herman.. its all on 1 side only. the headlamps were not on the car. the projectors were naked. i just mounted 1 projector at a time and took pix individually. the bottom pic with, what seems like 2 peaks, is from my shield. with the new setup, my 2mm bent shield is too much.
oh herman, just noticed something about my d2s capsule. on the glass, there seem to be spots etched onto the surface. its not noticable stare perpendicular to the glass. but if you look at the glass at a tangent, the spots becomes apparent. i can't get it off with alcohol nor windex. will these spots affect the output? d'oh!!!
Hard to tell without seing it. It sounds to me like density of your spots are minimal, so its probably nothing to worry about. Try to carefully remove it with a razor blade. The glass quartz, so its actually really difficult making scratches if you are used to lexan surfaces.
weather's been nice tonight. had some chance to tweak the installation. here's the end result with the projectors inside the headlamp, both installed and switched on.
not only did the cutoff sharpen'd dramatically but my colour band looks really nice, with strong blue and purple.
before i re-adjusted the driver side projector... it was crooked. this is with the projectors in my headlamp housing.
herman, if you like to use this for the faq, please feel free to do so. i'm quite pleased with the results. you recall how fuzzy my cutoff was when we last met up. my beam doesn't even cast much light against that wall, while you's and anthony (i hope i remember his name correctly) were nice and clear.
It really makes me wonder how my newly aquired ECE Bosch single Xenon will perform in comparison to yours. I am really happy to finally have an easy way to remove bulb without screwing out 3 bolts. Regardless, we have to meet up after I get my retro in...
I am the one that put the H9 halogen 90mm units on my Camaro (it's on the retrofit list). I have had them for a couple of years now, but I could not resist upgrading to HID!
I had the bulbs and ballasts sitting on my desk for a couple of months and then a couple of weeks ago started fitting the bulbs to 2 new H9 units that I bought. I used a grade 8 1" washer for the bulb retainer and it fits perfect. I tried out the lights and even experimented with tilting the bulb a hair, so that the arc would be ~1mm lower in the housing, but that seemed to be worse when judging by the pattern on a wall, so I left them dead center resting against the back plate (the bulb fits nicely within the ridge around the hole). The funny thing is that I searched through thsi forum a couple of weeks back for info about placing the bulb and could not find anything... and I remember someone doing a whole study on bulb placement and taking pictures, but I was not able to find that post anymore?) Anyways, I put the lights on my car on friday and drove it during the weekend and I was a little dissappointed because it seems there is less down the road illumination than with the H9's... Looking into the bulb specs right before reading this post I realized that the bulb needs to go further into the housing. Playing with my Maglite confirmed this. From your pics, I can see that mounting the bulb deeper will definitely improve the beam a lot!
Now, my question to Byron is: Did you end up making the bulb holder that Herman sketched or are your two rings replacing the original backplate? I don't quite understand your holder I guess... If I understand your final bulb position right, it is 0.5mm further in and 1.5mm down?
I was thinking of grinding a little of the little "posts" that the screws holding the backplate screw into (they define the axial position of the backplate). That way the backplate will sit a little deeper (I will also need to grind a little off the edge of the housing so that the seal does not get compressed too much). Did any of you try that? Is there a reason why that wouldn't work.
if you don't mind waiting till this wkend. i will be pulling my passenger side headlamp out to bend the shield a bit more. Just to even out the colour band between the both sides. Then I can take a pic of how the holder is attached to the projector.
The original backplate is no longer used, my holder replaces it. On mine, the backplate screw posts were positioned just fine. I didnt need to grind them. Just used my holder and screwed it on directly onto those posts. Measured the position of the holder to the front lense with a venier calliper, and it already resulted in a 2.5mm shift inward. I left it as at that. I also made 2 versions of the holder, 1 is 1.5mm down, and the other is 1.5mm up (since the h9 filament is offset from axis 1.5mm up). Tried both, and the 1.5mm down version came out being the best. However I didnt try in between the 1.5mm down and 1.5mm up (ie. 1mm down).
On a side note, if you would like. You can try 2.8mm and 3.1mm inward shift and see what happens to the beam. The filament on the h9 is 4.8mm while the arc length of the d2s is 4.2mm. thats about 0.6mm short. That yields a range of 2.5 ~ 3.1mm to play with. I would not recommend grinding the posts, as it becomes very difficult to maintain the perpendicular plane to the axis of the projector (3 posts forming a plane). I would rather offset the capsule base further in by design of the holder instead. Much easier to maintain accuracy.
(specs for the h9 and d2s bulbs thanks to herman!)
Aha! So you completely ditched the original backplate! That's what I thought! My current bulb holder looks a lot like your original one. However, I couldn't really see in the pictures if your bulb was resting directly against the backplate or if you had ground down the ridge. Just curious...
I realized it might be difficult to maintain alignment when you grind down the posts, but I thought I would mark it off before I start grinding and maybe I can pull it off... But you're probably right and I should just do it with the holder design. However, I am gonna have to completely redesign it. I also have to think about weatherproofing as the headlight area on the Camaro's is completely open. Right now I have gaskets in between the rings and at the bottom of a 1-1/2" PVC cap over the HV connector. The only part that's not completely waterproof is where the HV wire comes out of the bottom of the cap. Did you have to waterproof your lights or are they in a weatherproof enclosure?
I am starting to think of a bulb holder similar to yours that bolts directly to the posts... How thick is your "base plate" that bolts to the posts?
Sorry for all the questions, I am trying to take some of the trial and error out of this project... I know how it is, because over the past two years I have been BOMBARDED with emails asking for info from other Camaro owners that want to put 90mm units on their cars!! This is of course because the stock lighting on the 93-97 Camaro's is downright worthless!
the aluminum plate i used is 1.7mm thick. the plastic portion is about 1/4" thick. as for weatherproof. the projectors themselves are already weatherproof. with my new holder, i've sealed the up the holder against the projector. it will be splash proof, but not immersion proof, of course. lol!
improving my headlamp splash proof is next on my agenda. i have similar problems as your's. the rear portion of the projector is fully exposed to element within the headlamp bucket. so far, i don't have any problems yet. just condensation problems when my headlamp gets caked up with wet snow. also trying to find something to help seal up the hv plug. i dont have much room to work with at all
no problem in helpping out man. if its not for you gyus creating this site... i would still be pulling my hair out trying to tweak my hid kit in my stock headlamp. lol!
After driving around with the tweaked projectors. I can see why josch (iirc) modified the shield to eliminate the step up. Im thinking of lowering the upper cutoff. Its too high for hid use. I end up having the beam lower than it should be. Gonna try to make up a plate to lower the upper step in the cutoff. Fortunately I have a pair of spare shields to play with.
Let me know what you find out! I am curious what you end up with in terms of the shield. I think I may make the cutoff straight too at some point, or at least reduce the right hand flare. And if I go though the trouble of moving the bulb with a new holder, I don't want to have to do it again because the shield is now altered.
ok sam... just shorten the flare by about 0.5mm. i've added a piece of aluminum (0.3mm thick) to the upper flare area and extended it past the original position by 0.5mm. adjusted the headlamp up as well. i kept the flare but just minimized its height. the original flare step difference is 1mm. will see how this works out.
pic of the beam on wall with mod'd shield.
herman... i really would like to compare my 90's against your retro. the area i'm most interested in, is the width of the beam and intensity at the left and right side. can't wait till winter is over man! how soon will you have your retro done?
Looks really good, Byron...
At least picture shows that beam has gotten wider with less foreground light. Just what you wanted. After seeing how great they actually can get, I almost regret buying the ECE single Xenon Bosch projectors.
My project will take a while... I have to work more overtime and we are looking into moving to a bigger place. Both very time consuming, making my HID project going into fall. I have to think of something... I still don't even have all parts I need.
herman - whoh.. so you sold the entire hid system and headlamp from your sho? i can understand the other priorities. but i hope somehow things works out sooner. i still haven't had a chance to see how wide the beam cast on the highway... what with the wet conditions we have lately. i'm still not 100% satisfied till i can see if its truely as wide as an OE hid projector.
just received my new headlamps. holy crap herman... i thought i will only see a little improvement. but compare the new to my sand/buff/clearcoat ones... my s/b/c ones looks like crap! can't wait to swap front lenses and see what happens.
Byron, did you get new headlamps? What do you mean? Did you get new headlamps woth clear lenses to mount your projectors in? What car do you have BTW?
I have been experimenting with a new bulb holder and I am a little discouraged... Obviously, when you make a completely new backplate, you have to determine where the centerline of the lamp is. This is not easy! I have measured it several different ways, with slightly different results. I am using 16gauge steel sheetmetal which is 1.58mm thick, which (according to my calculations and measurements) should result in a 2.8mm move of the bulb inwards as compared to the original backplate. How did you determine the centerline precisely?
With the original backplate, the centerline is well defined but the bulb is too far out. I am seriously starting to think about grinding down the posts to accomplish moving the bulb inwards. I think it would be easier to measure that precisely than redetermining the centeline precisely.
I moved the bulb around slightly while it was on and that showed that it is VERY sensitive to position (which we already knew), but the rule of thumb that says get it within 1mm is not true when you're trying to perfect the light distribution. It really comes down to fractions of a mm.
I am doing the best I can with a caliper with dial indicator (0.01" graduations = 0.254mm), so I should be able to measure down to almost 0.1mm precision. But I guess it's the precision of the holes in the plate that is not precise enough. How did you tackle this? It looks like you were able to position your bulb much more precisely.
Im just swapping out the front lense. The front portion of my 90s are mounted in my OE headlamp. But my front lense is a little cloudy which I sand/buf/clearcoated. I knew it would not be as good as a new front lense, but I didnt realize there was THAT much difference between new and sand/buff/clearcoat. So Ill be swapping out the front lense and re-doing the mounting of my projectors (trying to seal it up better so I dont get fogging when the lense is coated with wet snow).
Stupid question, are you using a drill press? On another note, I made an ACAD drawing of the template. Printed it on a 1:1 scale, cut out the template and stick it to the piece to be cut. Do you have ACAD2000? I can email you the file when I get home.
as for my car, its a domestic "grocery getter"... lol!
i'm the green one, this was at darknights 2004. we were in the show field.
I would use a drill press if I had one. Maybe I will just get one, I would like to have one anyways. I have been using the original backplate as a template and clamped it in a vise, so I would not think that it would be that far off. I am still convinced that most of the placement error comes from my determination of the centerline. An ACAD file with a template that has the exact placement of the lamp centerline would be very useful! Could you send that to me? I could print it at work. My email is: mullersa at bellsouth dot net (Didn't put my actual address to prevent spam, but you can make it up )
Its not a bad idea to use the original backplate as the template actually. Depending on how it is clamped to your sheet, and the angle at which the drill is held. It can be off quite a bit. Thats why I would strongly suggest a drill press. It does not have to a floor standing version. A small bench top version would be sufficient. Less you are planning to do some large scale work that needs the height capacity of a floor standing version. Just make sure you pick up one that will suit most of what you are going to make.
On another note, I also use the drill press as a el-cheapo milling center. Youll need a very sharp flute on the drill bit or go with a milling bit. I used a small milling bit I borrowed from work to mill the center hole, using the stuck-on template as a guide as I mill out the hole.
Ill send you the ACAD file when I get home tonight. I assume you can read ACAD2000 version? Do let me know, if not I can export to different formats like dxf and such.
I did buy a hole saw that works pretty decent, you still have to use a Dremel to break the sharp edges though and it's not very precise. But I made the big hole first (on a big sheet) so that I can then drill the smaller holes more precisely around the big hole using it as a reference. It turned out my mounted backplate had the big hole almost exactly on the centerline of the lamp (as I measured it). I am just starting to wonder about my determination of the centerline...
I can read ACAD200 at work yes... No problem there...
ok.. file sent. let me know how it turns out. definitely pix too! since you have it deeper in than i did, by 0.3mm. really curious if this works out better. the file i sent you has the d2s hole offset by 1mm down. you can use acad to move the circle around (it also has a key in it, but i didn't p-line/p-edit it, so make sure you select it all). i'm gonna try the 1mm offset probably next wk if not this wkend.