heinzboehmer's 2002 Topaz 6MT Coupe

[thetypicalm3guy]

When I replaced my mirrors, I bought a genuine BMW part for the driver's side and an S-Tec branded one for the passenger's side. FCP said S-Tec was OE for the passenger's side, but not for the driver's side. I thought that was weird, but the price difference wasn't enough, so I just trusted them.

The passenger's side mirror fit with no issues, but the driver's side mirror was very wavy. So wavy that I drove the car for about five minutes before driving back home and swapping the old, corroded glass back in. Thought it was funny that the genuine unit was worse than the OE one, but whatever. Returned the BMW one and bought an S-Tec one to use instead, since it had worked perfectly on the driver's side. And of course, turns out S-Tec is actually OE, since the BMW logo is clearly ground off:



Unfortunately the S-Tec part was also wavy. Not surprising, since it turned out to be the same part, but at least I got the same thing for $30 less.

Also, I noticed something that I didn't notice last time. The glass was perfectly flat when not installed, but became wavy once clipped in. Seems like the clipping (and holding) mechanism is distorting the glass. Great.

I went and pressed on the eight "preload" tabs (pointed out below) with my fingers until they weren't sticking out as much. This seems to have reduced the force on the glass enough to not distort it when installed. Nice, fixed.



Hopefully the tabs don't spring back and re-distort the glass. If they do, I'll probably cut them off and some foam/double sided foam tape for preload instead. I'm slightly annoyed at BMW for switching to the newer glass mechanism. The old one works great and is held in place with a radial locking mechanism, instead of using a bunch of tabs to clip in. I see no need for the switch, but I guess these new ones are probably cheaper to produce.

Anyway, after much more effort than needed for a simple mirror glass swap, I finally have glass that doesn't look terrible

Dude this EXACT same thing happened to me i kept returning OE bmw mirrors due to waviness. Currently have waviness and the glass is wobbling too, super frustrating

[heinzboehmer]

Dude this EXACT same thing happened to me i kept returning OE bmw mirrors due to waviness. Currently have waviness and the glass is wobbling too, super frustrating

So annoying. Worst part about all of this is that the passenger side one was perfect.

Hopefully the "bending the tabs" thing is a permanent fix. If not, I plan on cutting all of the tabs off and attaching the glass with some of that double sided foam tape. Might be worth adding some to yours to stop the wobble.

[Casa de Mesa]

So annoying. Worst part about all of this is that the passenger side one was perfect.

Hopefully the "bending the tabs" thing is a permanent fix. If not, I plan on cutting all of the tabs off and attaching the glass with some of that double sided foam tape. Might be worth adding some to yours to stop the wobble.

Love it when you update this thread, it's always informative!

On another vehicle, I once used foam tape to secure the mirror glass. Similarly, I still used the factory retaining clips but supplemented them with foam tape. Net result: It was *extremely* shaky at highway speeds, super annoying. So if you go down this route, make sure you have enough of the original mount to keep the glass from wobbling. That is, don't go full on postal on those tabs, find a happy medium. It might be a different enough design on the M3 mirrors where my experience won't translate to the same results here, but figured it was worth mentioning.

[heinzboehmer]

Love it when you update this thread, it's always informative!

On another vehicle, I once used foam tape to secure the mirror glass. Similarly, I still used the factory retaining clips but supplemented them with foam tape. Net result: It was *extremely* shaky at highway speeds, super annoying. So if you go down this route, make sure you have enough of the original mount to keep the glass from wobbling. That is, don't go full on postal on those tabs, find a happy medium. It might be a different enough design on the M3 mirrors where my experience won't translate to the same results here, but figured it was worth mentioning.

Appreciate the info! So far, glass has been good though. The bendy fix seems to have worked well.

[heinzboehmer]

Discovered an oil leak recently coming from the front of my engine. Originally thought it was coming from the crush washer on the bolt that holds the passenger side timing chain guide, but after cleaning things up and going for another drive, it became clear that it was actually coming from the lower passenger side VANOS bolt.

Last time I changed the gasket (~10k mi ago), I used some Toyota FIPG to seal that area, so I was surprised it was leaking. However, I recently swapped out the BimmerWorld VANOS line I had on the car for one of the BMW updated ones (didn't like the fast that there's a service interval on the stainless ones) and the top passenger side VANOS bolt felt weird when torquing.

I decided to replace the gasket and redo the RTV. Since I was pretty sure I would need to do a thread repair, I borrowed a Time-Sert kit (thanks again eacmen !) and got ready to do that too.

And of course, the threads were pretty messed up:



I'm thinking this is what caused the leak to start, but just a guess. Oh well, I have a good amount of practice with Time-Serts now, so the repair went by super smoothly. I ended up doing everything by hand, as I was unable to get a drill in there and reaally didn't feel like draining the coolant (mostly this, I hate the mess it makes) and removing the front of the car.

Drilled out:



New threads for the insert:



Aaand insert in place:




Because that hole also houses a dowel, I was unable to get the counterboring tool in there, which means that the end of the insert is further forwards than where the step for the dowel would usually be. Decided to take the easy route and shorten the dowel, instead of trying to counter bore inside of the hole with limited access.

I also filed down all those marks on the head —and the corresponding ones on the VANOS— to ensure that the mating surfaces were as flat as possible. Wonder what idiot past mechanic made those. Definitely wasn't me when I had just bought the car and was tearing into the engine to replace a failed timing chain guide bolt with limited experience working on cars.

Then I bought a new dowel (mangled the old one getting it out) and measured the difference in depth between the undamaged hole and the newly repaired one. Came out to 1.75mm, so I set off to remove that from the dowel. Ended up removing 1.8mm (decided to bevel it after I had already cut it), but that still left plenty of dowel to center the VANOS onto the head:



(yes, I know that's a terrible way to use a caliper to measure this, just set it up like that for the photo)

And here's the difference between the two sides of the VANOS with both dowels fully seated. I must have checked a million times that the correct dowel was going into the correct hole lol:




Pretty happy with how accurate I managed to get this. Was all thanks to my amazing lathe (i.e. shitty 10" drill press and a dremel).

And here's the engine back together (along with a sneak peek of the next project):



Oh! Almost forgot to mention that I found the VANOS gasket installed the other way around when I took the unit off. I'm sure that didn't contribute to the leak at all. Really need to find that previous mechanic and talk to him about why you shouldn't rush those types of jobs in order to make it to a dyno session

[heinzboehmer]

So, about that next project. I bought all the necessary bits to do an MK60 swap a few months ago, but hadn't started the project because I knew it was going to be a long one (and yes, I still underestimated the effort). Decided that I needed to get it done, so I booked a track day on Oct 31 to motivate myself.

Have actually been working (very slowly) on this for about a month now. Visits from friends and family, waiting on parts and a couple screw up have all made this take much longer than I anticipated. Oh, and also the fact that the amount of work involved is A LOT. This past weekend, I spent 24(!) hours total just bending brake lines. Admittedly, I do work very slowly when it comes to cars, but my pace is probably only 2-3x slower than a seasoned mechanic. 8 hours to rerun brake lines sounds about right for someone who knows what they're doing and that's just one part of the project.

Anyway, onto the progress updates. Plan on doing a full writeup on this, so I'll add more DIYish details there and I'll stick to updates and tips (i.e. explaining what I had to redo) here.

Took one of those "here's all you need for this project" type pictures before starting. Not pictured are some more bottles of brake fluid and the five million meters of rear brake lines hanging off of the workbench:



Started by taking apart and cleaning the used parts I bought. They looked pretty good, but the paint on the motor housing and on the bracket were failing, presumably from sitting in brake fluid for a while. Stripped those down and repainted with POR-15:







When cleaning the parts up, I also replaced all seals and hardware with new ones because I had excellent access to everything and I absolutely despise brake fluid. Wasn't too happy about buying those $180 each brake pressure sensors though.

Next came ripping out all the MK20 stuff. I decided that I wanted to have all of this look as if it came from the factory like this when I was done, so I ripped everything out (including all the hard brake lines) and replaced with brand new parts. Thought it was a great opportunity to refresh the 20 year old parts in the brake system as well.

Started by getting all the fluid out. Pressure bleeder with just pressurized air connected to the brake fluid reservoir did a great job of draining the system. Then I moved on to removing the hardware and lines. I tried to keep any lines that I could potentially use as templates intact. Everything else was cut/bent/ripped out because I was lazy. Example:



Some pictures right before I ripped out all the mess of brake lines that went behind the engine. Stoked to remove this ridiculously overly-complex system:




Disgusting

[heinzboehmer]

Next came test fitting the new module. Gave everything a quick wipe (full clean coming later) and quickly realized it wasn't a straight replacement. Seems like MK20 cars have an extra tab welded to the shock tower for the accumulator. Tried bending it out of the way first:



But that didn't really leave enough clearance for the module plug, so I just cut it off with a dremel and added some POR-15 (not pictured) to the exposed metal:



Module fit well now! Fortunately the one threaded mounting spot that the MK60 uses was already present on my car (I think it was also used by the MK20 stuff, but I forget), so no need to fabricate that:



Next came making the new lines. I removed the gas tank, but left the subframe in place. Snaking the line above the subframe wasn't terrible, but trying to snake the lines above the gas tank would have been basically impossible, so highly recommend dropping the tank if you attempt this (although what I actually recommend is not taking on this project).

For the sections that were the same between the two systems, I used the old lines as a guide. Those white plastic clips were extremely useful for holding everything together and measuring:





And here's the newly bent line next to the original. Pretty happy with the result:



For the rest of the stuff that I had no reference for, I used a clothes hanger and LOTS of comparing against the pictures in Casa de Mesa 's journal (and some more he sent me after I asked for help). The clothes hanger was much easier to play with to try and get all the angles right before going to bend the actual line. Example of making an end of a line that connects to the module:



And an example of a finished line (driver front and not the same as what I was using the hanger for):



Things were looking good, but then came the rear lines. I used the ones removed from the car as templates, but I wasn't able to follow them 100% (those lines are looong and hard to handle). Was probably only off by 2-3mm per bend, but after 25ish bends, those tolerances add up quick. I also made the mistake of starting at the calipers for those, which made getting them into the engine bay quite hard. The end result was two lines that looked terrible and fit even worse. Spent about an hour trying to rebend them by hand so that they would not foul against the gas tank or the body and eventually gave up. Went upstairs and immediately ordered two replacements. Fortunately they're decently cheap.

When the replacements came in, I started at the module side and then worked my way back. Used a combination of the wire hanger and my hands to get the correct bends in the area gas tank above the gas tank and the result was significantly better than before. I'm super glad I redid those, as I would have forever not trusted my brakes fully, fearing that the lines would rub through at some point. No pics because I was exhausted by the time I finished, but I promise they look good. Would have redone them again if they didn't

Here's the finished lines at the module (things are crooked because nothing was tightened down yet):



And a picture that brings me a lot of joy. Look at that reduction in complexity!



24 work hours later (+ 6ish wasted hours from the original run of rear lines), I have brake lines again! Tightened everything up and ran a "leakdown" test with the pressure bleeder (no fluid) pressurized to 20 psi. I know that's nowhere near the pressures that the system will see, but the bleeder pressurizes the reservoir too and that's not really made to be under pressure, so I didn't feel comfortable pushing past 20 psi. I plan on running another leak test once I fill the system up so that I can test with more realistic pressures too.

Here's the initial pressure and temp:



And here are the same things ~22 hrs later:



I think a drop of ~0.4 psi (temp corrected) over 22 hrs is pretty good! Especially when you consider this is with just air in the system, should be harder for fluid to leak out. Also, I saw most of the pressure drop happen over the course of the first couple hours, it stayed pretty much unchanged for the rest of the time. I'm assuming it was mostly just the temp inside the reservoir coming down (after it went up when I pumped it up) and the gauge settling. Regardless, I'm happy but will run more tests to make sure this is all 100% leak free.

[heinzboehmer]

So the lines were a lot of work, but they turned out well. I'm happy that they kinda look like the car came like this from the factory and not like the copper line that runs to the ice maker on the fridge that someone installed in 5 min and didn't bother to even try to make look remotely presentable.

Besides the lines, I still need to do a good amount of things. Namely the wiring.

When I started this project, I somehow got into my head that the MK20 module wiring ran to it across the engine bay from the DME area. I was very wrong about that. Turns out the MK20 parts of the body harness split out by the glovebox and come out into the drug bin. On MK60 cars, the wiring runs behind the dash/carpet and comes out into the DME area. This meant that I would need to fish the connected into the interior and then solder/extend 40-something wires, run then behind the dash and have them come out into the DME area. I decided that I really didn't want to have 40 splices living under my carpet for both troubleshooting and fire reasons, so I went another direction.

Plan is to build an adapter harness that just plugs into the existing MK20 connector and terminates in a stock-like MK60 harness. Initially, I'll just throw it across the engine bay, so that I can debug any issues that come up, but the plan is to eventually run it through the interior behind the dash. I want to do this so that I don't have a big bundle of wires just hanging over the engine and also so that the install looks as stock as possible.

First thing I did was jump on eBay and picked up a non-M MK20 for cheap(ish). I then cracked it open, desoldered the board and cut away the connector from the housing:




Took some measurements and created a super simple breakout PCB with everything labeled for ease of translation between the MK20 and MK60 wiring. Pic of the test fit stage, still don't have the boards in my hand:



The pins with Xs are those for the connectors that I'll wire directly to the MK60 plug. These are basically anything that is already in the DME area (brake fluid level sensor, pressure sensors, etc.). I decided to still include them in the PCB because I had the space and didn't want to have to order another version if I changed my mind.

I also was decently sketched out about the two power wires, since each of those run to a 30 A fuse and I really wanted to avoid burning up that board. Even though I do technically have a piece of paper that says I should be good with electronics, I have never really done much higher power PCB design. I read up as much as I could about it and then reached out to people who actually know what they're doing for input (thanks again Bry5on !). The traces I made should be capable of flowing 32.5 A at 40°C temp rise and 20 A at 13°C temp rise. I'm fairly confident I won't cause a fire, but the order minimum for the boards was 5, so I plan on sacrificing one and running 100-200 A through it just to see what happens. I'll also take some measurements of the actual temps that the board gets to at 30 A before trying to make it catch on fire.

For the rest of the connectors, I decided that I didn't want to spend hours digging through TE catalogs to try and find them, so I just bought this kit: https://tulayswirewerks.com/product/...connector-kit/ I'm actually super glad I did because the owner of that shop, Kevin, turned out to be an awesome person and let me bounce ideas off of him for a couple days. He seems to be very experienced, so I felt like I was in good hands. He also provided some current draw data from an MK60 track car. Plots showed the MK60 current draw stayed very close to 0 most of the time. It was only under hard braking that it shot up to 20-25 A per wire. Eased my worries about the traces as well.

Anyway, I'll get to building the harness once I receive the PCBs. Should be super straightforward now that the hard parts (figuring out MK20->MK60 mapping and stressing over the traces) are done.

[heinzboehmer]

Also wanted to quickly give my thoughts about this whole project now that I'm done with a good chunk of it.

Is it worth it for the increase in performance? Debatable

Is worth it financially? Absolutely not

Is it worth it for the learning experiences? Absolutely yes

Would I do this again? Absolutely not


So, all in all, this is something that I probably would not recommend others do. It is a TON of work and the cost is high. Yes, I could have done it cheaper (skip on all the "while you're in there"s and don't go on an adapter harness side project). And yes, there are ways to cheat and make it easier (especially with the brake lines), but I knew that I wouldn't be satisfied with anything that wasn't basically factory. So to do this to that standard, the amount of work and cost are both going to have to be pretty high.

The reason why this made sense to me was that the car already has a ton of stuff done to it that I would have had to redo if I sold it and bought an MK60 car. The amount of work required to redo all of that was higher than the amount of work needed for this swap, so I decided to go ahead with it. Oh and I also might have a slight problem with planning out projects revolving around things that I haven't done before

But if you don't already have a fully sorted car or don't find yourself craving a BIG project, it makes way more sense to just sell the car and buy one that came from the factory with an MK60. Or maybe even pay the likely large (but completely justified) amount of money to have someone like George Hill do this for you.

[heinzboehmer]

PCBs just arrived today! So I took another of those "lay out all the parts on a table at right angles to each other" pictures:



Now just have to transform all that into one harness.

Here's a close up of the board. Fits great and looks awesome too:



Man, I really like how those black PCBs look. The quality of the finish is surprisingly good too. Slight deviation from car content, but I've used this company (JLCPCB) before to make aluminum faceplates for guitar pedals and the quality is awesome. Here's a pic of when I was trying to figure out what color combinations to go with for this particular pedal:

[Bry5on]

You’re an animal. Also, JLC is great, I have a couple boards that they did in my EV conversion and I’m going on a couple years with no issues.

[Tbonem3]

Props dude

[Casa de Mesa]

Wow man. I'm totally in awe of this. This is pretty epic level stuff. Way to send it!

So glad the photos were helpful. I can't wait to see your reaction/feelings/write up when you finally put something other than pressurized air in those lines.

This build continues to be one of my favorites - you simply do not disappoint!

[heinzboehmer]

You’re an animal. Also, JLC is great, I have a couple boards that they did in my EV conversion and I’m going on a couple years with no issues.

Props dude

Wow man. I'm totally in awe of this. This is pretty epic level stuff. Way to send it!

So glad the photos were helpful. I can't wait to see your reaction/feelings/write up when you finally put something other than pressurized air in those lines.

This build continues to be one of my favorites - you simply do not disappoint!

Appreciate it! Also very excited to get the car back on the road (and mostly back on the track).

[nuc1]

Also wanted to quickly give my thoughts about this whole project now that I'm done with a good chunk of it.

Is it worth it for the increase in performance? Debatable

Is worth it financially? Absolutely not

Is it worth it for the learning experiences? Absolutely yes

Would I do this again? Absolutely not

Sums up just about every car project I've ever done.

Awesome work.

Mike