My FX-R Projector Retrofit Thread

[e30trooper]

good job man , cant wait to see the set up! What FX-R's did you get? 1stgen,2nd gen or the 1.1's ? I got lucky on a pair of 1st gens, the 2nds had a few issues with the end of the cut off slooping off a lil. Anyways get'er done! Make vids. Im my retro thread i never updated pix when i re did my brackets and such.

[ethrtyiS]

so far excellent work. love to see the final output on these.

[bmwman91]

I work at one ;)

http://www.iscomach.com/

Sweet. Once I get a bit further along with this, I'll hit you up for a quotation. It sounds like a lot of people may be interested, so perhaps we can get a good number made. To seal these, it may still need to be a machined part, but I am trying like hell to constrain myself to a 2D cutting operation to keep the cost down.

good job man , cant wait to see the set up! What FX-R's did you get? 1stgen,2nd gen or the 1.1's ? I got lucky on a pair of 1st gens, the 2nds had a few issues with the end of the cut off slooping off a lil. Anyways get'er done! Make vids. Im my retro thread i never updated pix when i re did my brackets and such.

I got 1.1's. What are the differences between the different ones?

[e30trooper]

Sweet. Once I get a bit further along with this, I'll hit you up for a quotation. It sounds like a lot of people may be interested, so perhaps we can get a good number made. To seal these, it may still need to be a machined part, but I am trying like hell to constrain myself to a 2D cutting operation to keep the cost down.


I got 1.1's. What are the differences between the different ones?

well 1st gen are the first design, 2.0 are newer but had some flaws. the 1.1's are 1st gens thats are a lil different

fxr 1.0
.

fxr 2.0

[bmwman91]

Yikes, the 2.0 looks terrible. Maybe the 1.1's are the 1.0 housings that they put the (upgraded) clear lenses into?

[bmwman91]

Alrighty, I made some progress on this today. The housings have been trimmed down and I'll be figuring out exactly how I want to build an adapter plate.

First, I trimmed out a bunch of the crap inside with my trusty 1" end mill. The cut-off wheel I used next had a rather large arbor and I needed to get this crap out of the way first.


I did most of the trimming with a cut-off wheel in the mill. The housing was mounted face-down on the rotary table and manually spun.


I made a few passes to get things close to the final dimensions. The cut-off wheel is pretty nifty and didn't cause any chipping.


The last 0.015" was knocked off with a 6-flute 1" end mill at 1200RPM.


So, here they are. With this done, I can now nail down the adapter design.

[ethrtyiS]

looks good so far.

[Neil1138]

I did an FX-R retrofit in my e46 a short while back. A lot of dremmeling and such but the final result was worth it: FX-R low beams + custom shield/clear lens P46 projector fogs.



I've considered doing another retrofit in the e30 but haven't yet. To be honest though, FX-Rs by themselves are lacking. They have a pretty wide throw so the intensity of light is pretty low. But combined with some (proper) projector fog lights, the total light output is stunning. I imagine it would be the same in the e30. If I were you, I would keep the stock projectors and try and retrofit those into your fog lights!

I might make some custom curved shields for the e30 projectors and replace the lenses with some clear ones. Could be worth it.

Your approach is hella legit. Can't wait to see the final result!

[bmwman91]

Alrighty, I have given this some more thought and I now know how I plan to tackle this.

There will not be a cut-out for the city light bulb. I have been using the same ones for close to 7 years now; the replacement interval is super long. I would much rather have a hermetic seal that I am confident in than easy access to a part that almost never needs replacing.

The plate will be secured to the housing via two screw blocks and Permatex Ultra Black RTV. The screw blocks are in there because I just don't trust adhesives that see water/salt water, heat cycling & have something as important as headlights depending on them. Also, using a softer sealant like RTV means that, should you need to replace the city light bulb, these plates will just need some unscrewing and a little pulling, and then $2 worth of RTV to reassemble.







The big blue boot thing will be a project all of its own. I cannot find any off the shelf parts that can fit over a 5" perimeter, so I will make the necessary molding plugs/cavities on a wood lathe, seal them with some Shellac & then cast the parts from some sort of urethane-rubber or silicone compound that can take heat & exposure to car fluids.

Do it right, or don't bother, I say!

[george graves]

subscribed

[bmwman91]

Ok boys & girls, time for another update. I mocked-up a prototype adapter plate (it is usable, but there are some design changes I want to make now that I have the part in my hands).

My employer was throwing away some old manufacturing fixtures a couple of years ago, and I offered to take them off of their hands instead. These things were made with 120lbs (each!) of machined aluminum, in various bar & plate forms. So, needless to say, I have had close to 100lbs of aluminum plate laying around for years, waiting for me to do something with it. Anyway, I found a suitable piece & needed to trim it down for the adapter plate. I started with a bandsaw, but it was way too slow.


So I stuck it in a vise and milled it into shape!






Next, I needed to get the rotary table centered. This little dial gadget is super handy! Put it in a 1/4" collet, set the spindle to ~100RPM and adjust the X/Y position until the needle stops wobbling.


I got down to business & made all of the back-side cuts. The 3 larger holes in the row were used to set the center position of the work piece so that all of the various rounds could be cut.


Here it is, half-way done with the top-side operations. It didn't take much work since I had cut the inner profile perimeter already. Once I knocked ~.6" off of the inside, the middle bit fell right out!


The crappiest part of this was cutting the outer perimeter. It involved a delicate dance of milling, stopping to remove one bolt, milling, stopping to put it back, etc x 4. By the end, it was not holding the part securely. So, I cut my losses (no pun) and chucked it up on the lathe to clean it up.


All cleaned up, although it got taken down to 4.99" instead of the intended 5". Whatever, it is a prototype.


So, here is one. Whoever says you NEED a CNC mill is telling lies. With some effort, you can design most parts to be made manually. Hell, the digital position readout unit on the mill crapped out last year, and I ran this part without any electronic positioning at all. As long as you know what "backlash" is, you'll be fine. All you need is the ability to add & subtract in your head, and a decent dimensioned drawing. Oh, and calipers are a good idea, too!


So, how did it fit?


Well, it fit almost perfectly. Snug, with zero slop (much to my surprise). It does rotate ~0.5 degree thanks to an "oops" in the cutout for the city light, but radially there is no movement at all.






I will be taking the lessons learned here & making two more that should come out better. I did most cuts in a single pass, which is always a recipe for rough milled surfaces.


I will probably also mill a counterbore for the two mounting screws to get them flush (the heads would hit the FX-R projector assembly otherwise). I didn't account for the fact that a countersinking bit wouldn't fit here! I also plan to thin the outer ring by ~3mm (off of the outside) to get better clearance with the headlight bracket (it fits, but the rubber boot I plan to make would be really tight in there with this OD).


I also managed to get off by a couple of handle-turns down in the cut-out for the bi-xenon solenoid, and I forgot that the surface I was accidentally cutting through was a sealing surface.




Up by the city light, I started out being off by half of a cutter diameter in the radius. It's a common rookie mistake (and I'm a rookie at machining).


Lastly, there's a nibble in the outer perimeter from when I was cutting it out & the clamping wasn't quite adequate.


So, the next ones will be a lot cleaner. I'll probably do all cuts in 3 passes (-0.020", -0.005", 0.000"), spend more time with some of the centering setup, counterbore the 2 mounting holes, rough-cut the perimeter & finish it up on the lathe, and reduce the outer diameter by ~1/4". Yeah, it is probably overkill nit-picking this stuff. But, when you have the tools for it.....

[AndrewBird]

Very impressive work on the mill! That takes a lot of patience. Do you know the difference between conventional and climb milling? If not, look it up. Has a lot to do with surface finish, especially if your mill isn't the most rigid. Cutter speed is important as well. Aluminum has a cutting speed of about 300 sfm. Using that you can calculate RPM's based off cutter diameter using the following formula:

Cutting speed(300) x 4 divided by cutter diameter.

Tap Plastics has some RTV silicone you could probably use to make your boots with:

http://www.tapplastics.com/shop/category.php?bid=32&

It is designed to be used as the mold, but I would think it would do fine as the part as well. Not sure about chemical resistance though.



And, and it's called a slitting saw, not a cut-off wheel. ;)

[bmwman91]

Very impressive work on the mill! That takes a lot of patience. Do you know the difference between conventional and climb milling? If not, look it up. Has a lot to do with surface finish, especially if your mill isn't the most rigid. Cutter speed is important as well. Aluminum has a cutting speed of about 300 sfm. Using that you can calculate RPM's based off cutter diameter using the following formula:

Cutting speed(300) x 4 divided by cutter diameter.

Tap Plastics has some RTV silicone you could probably use to make your boots with:

http://www.tapplastics.com/shop/category.php?bid=32&

It is designed to be used as the mold, but I would think it would do fine as the part as well. Not sure about chemical resistance though.



And, and it's called a slitting saw, not a cut-off wheel. ;)

Thanks, I was hoping that someone with real machining experience could chime in!

"Cut-off wheel" lol, I blame years of using a Dremel for almost everything! Everything looks like a "cut off wheel" to me now.

I do know the difference between conventional & climb milling. I guess I have avoided climb milling because all they ever taught us back in school (engineering) was that, "climb milling can give a better finish, but wears out the cutting tool faster." I think that undergrad MechE students probably only had to spend 10 hours on a manual mill, which was/is totally inadequate. Hell, by this point I bet I only have 50 hours on a mill!

Anyway, by your recommended calculation, I should be using 2400 RPM with my 1/2" end mill & 4800 RPM with the 1/4" one. So, some of the lousy finish is probably due to that. The other culprit was using a single heavy pass for all cuts since I figured I would be using this as a proof-of-concept part. What are the units of "sfm"? If it is inches per minute, I think I may have a hard time hitting 300ipm by-hand!

For the actual parts, my plan is to do cuts in 2 or 3 steps, with the final pass removing 0.005". Would climb-cutting that last pass yield a better finish? I have also read that using Kerosene will give a better finish than cutting oil; any thoughts there? I think that it is time to get out my 1000 page Manufacturing Processes text/bible & re-read the section on milling.

Thanks for the TAP link. I will check their stuff out!

[siazul]

Wow nice job, I have some FX-Rs in my closet next to a set of ellipsoids that I plan on retrofitting as well

[burkey001]

Awesome work.
If this ends up being a product I will "add to cart"