E82 128i: when your projected "track car" turns into your daily

I know there's a constant pattern with this thread, but I'm again beginning to lose motivation on continuing to build the 128i. This time not because of the effort involved- the hardest jobs have already been completed. I'm beginning to feel a bit overwhelmed by the amount of money I would need to spend to actually make the car competitive. 

The Bilstein B12 cup kit currently on the car is great on the street- very OEM-like. I mean, the Eibach springs it comes with are similar to OEM Sport package spring rates. The problem is, on track, there's a lot of body roll, taking away quite a bit of confidence at 10/10ths. 

So, I've been looking into coilovers... I should preface by saying that if I'm going to get further into time attack, I can run a 17x9 wheel (currently on 17x8.5s) under my stock fenders. But there's a catch. These cars have narrow bodies and most of the coilovers on the market actually reduce inner clearance. Only the more motorsports oriented coilovers offer additional clearance- namely MCS, JRZ, GC and TCKline. 

There have allegedly been some folks online that have customized Ohlins to work (custom made spacers and custom, shorter springs), but there's no one that sells a set that will just bolt up. And even if you do build them, by the time you're done buying camber plates to go with them, you're looking at close to $4000 from what I can tell. 

TC Klines are by far the cheapest option, with their single adjustables coming in at around ~$2500. However, they're build on decades old twin-tube Konis that are limited in the rates you can run on the street and it seems like going with a higher rate will severely diminish the car's ride quality. They recommend running dual adjustable rears, but that bumps the price up to ~$3000. They recommended 300/600 rates for a dual duty car (considerably lower than most other motorsports oriented coilovers) and 400/700 for a more aggressive setup. I have my concerns that 300/600 will still be a bit too low for track duty and 400/700 would be harsh on the street. 

The "correct" answer is probably a set of MCS coilovers. Unfortunately, the cheapest ones I could find are from Vorshlag and by the time you're done, for a set of single adjustables, you're in for ~$4700. That's more than I paid for the car. I reached out to Bimmerworld, for a similar set, they quoted me over $6000. Everyone that has an MCS setup seems to swear by them. They sound amazing and supposedly allow you to run sky-high rates while still maintain reasonable ride quality... but damn. They literally cost more than I paid for the car. 

Don't get me wrong, you can run other coilovers for less money, there are plenty in the under $3000 range that will work great, but the vast majority of them limit your inner wheel and tire clearance. Running an OTS set of Ohlins, or Bilsteins, or KWs would limit me to my current 17x8.5" wheels, forcing me to give up time. 

A set of 17x9s and 255 wide tires would also likely run me another $2500 or so. 

And the car is still running an open rear diff, which makes it difficult to rotate on throttle on track. Picking up a 3.73 rear diff (relatively cheap and plentiful), plus a LSD will run me anywhere from $2000 to $4000, depending on which route I wanted to take. 

There was also a bit of squirm in the rear end under hard braking. Surely the OEM rear subframe bushings aren't helping, even though I have a set of cheap inserts in them. The right answer is a set of aluminum bushings ($300+), dropping the subframe and pressing them in. 

And let's not forget about a custom alignment and corner balancing ($600+). 

By the time I actually prepped the car to be competitive, I'm looking at another $8400 to $12,000+, on top of what I've already spent on the car. And that's not factoring in safety (at least a partial cage isn't a bad idea), upgraded brakes or track insurance (more expensive than the events themselves!). 

I'm starting to think: maybe I can't really afford this. I might not have deep enough pockets to play this game. 

You can afford it, but maybe you can't afford your standards?

I'm still confused about this "rotate on throttle" concept of which you speak 

In reply to Pete. (l33t FS) :

That's a fair assessment. 

The car is about ~6 seconds behind the season leader on a ~2 minute course configuration (to be fair, he's a few seconds in front of everyone else as well). That's a pretty big gap. I know the driver- we used to autocross and time attack together- he's fast and his car is well prepped. 

There's certainly some time to be found by tightening the rusty nut behind the wheel, but not that much time. I can't add anymore power. The suspension is soft, the car lacks confidence in high speed braking or transitions and attempting to apply throttle mid-corner results in additional understeer. 

I have little to no motivation to do HPDE or open lapping these days- I just see it as additional wear and tear on the car with no real goal in sight. But I do enjoy competition. The problem is, in order to be competitive, I need to spend more money. Quite a bit more money. 

Everyone else in the class is already running coilovers, a LSD and wider wheels/tires. I understand pretty well that if I would like to catch up, I need to also stiffen the suspension, pick up an LSD and it's probably a good idea to invest in wider wheels/tires. Otherwise, I'm essentially just running open laps. 

So while technically, I can afford to go to the events, if I actually want to be competitive with the top driver(s), I need to spend quite a bit of money. 

I had MCS on my E36 M3 and have them currently on my 996.  I wouldn't say that the ride quality of them on the 996 is all that great with the higher spring rates I have, the PSS9's they replaced were definitely far more civilized on the street.  The M3 was basically a track only car, but it seemed to ride well enough on the street that I don't remember it beating me up.  I had the TC Kline doubles on my E46 M3, and they were awesome.  Totally streetable and the car handled really nicely.

It's hard to not be competitive, but sometimes it's good to take a step back and ask yourself if you're having fun.  Will you have more fun if you spend the thousands of dollars to be competitive?

In reply to docwyte :

I usually have more fun if I'm competitive  this last event, some frustration surfaced running behind a couple caged E36s- nipping at their heels all day. Unfortunately I kept having to let off the throttle on the back straight, hurting my times, but it was unavoidable, as their cars were more planted at the following corner (turn 4/Bikers Berm). My 128i felt extremely floppy and disconcerting in that high speed braking zone and the transition into turn 4. Which sucks, as having to slow on the back straight repeatedly (to avoid rear ending them) was negating my power advantage. I felt like I was one suspension change away from smoking them both and jumping into 2nd place, which would ironically allow me to run faster on the back straight. 

I reached out to TC Kline- they recommended doubles in the rear, to which, he also informed me their doubles require me to switch to M3 lower control arms/spring perches (~$300 on eBay). At that point, we're close to $3300 all in on fancy Koni's. At a certain point, spending the extra $1400 for an MCS setup starts to make more sense- the "buy once, cry once" mantra. Assuming I continue going down this rabbit hole. 
 

If you can recall, what are/were the spring rates you are/were running, on the the MCS and TC Kline setup? I'm sure the motion ratios are a bit different, but I'm pretty sure all of these cars (E82, E36/46 & 996) are all strut based in the front, so I'm assuming they're comparable. 
 

Would you say the MCS' are worth the extra cash? 

In reply to roninsoldier83 :

It's been a hot minute since I've owned either of those cars.  If I recall I think I had 450/600's on the E46 M3 and something a bit higher than that on the E36 M3 with MCS'.

The MCS are definitely in a different level of performance compared to a koni or a bilstein, they're a race part vs a street part.  I'm not sure that the single adjustable MCS are worth the price compared to a single adjustable Koni.  Race parts also means they have a service interval, MCS says they need to be rebuilt every 30k miles, or 5 years.  My fronts will be due soon just because of time, not mileage. 

I really liked them on my E36 M3, which is why I grabbed the set for my 996 when they popped up used, locally.  I'm a little disappointed with ride quality of them on my 996, although the spring rates are fairly high.  I either need to buy lower rate springs that are the same length as I currently have, so I can hopefully avoid paying for a new corner balance, or install the springs I do have and then pay for a whole new alignment/corner balance.  Either way, I'll probably wait to do that until I send the front struts back to MCS for servicing.

I just finished putting together a breakdown of what it’s cost so far to fix and maintain my 128i, what I’ve spent to make it somewhat quick and reliable on track, and what it’ll likely cost to turn it into a legitimate contender in its class. If you're curious (or want a reality check before diving into a similar build), feel free to take a peek.

 

Was the timing chain tensioner an easy job?

edit: that's a pretty serious spend man. I read all of it and voluntarily went down a similar path - even paid a shop for some of mine. Far more that I hoped to spend! There is temptation to cut bait but I really like the way it drives. 

Adding up everything you've put into a project is a cardinal sin, lol!  It's always sobering to do that and then think about what you could've bought with that money instead.  For instance, when I did my last really stupid build, the LSx swapped 951, I easily could've bought a 996 GT3 instead.  Which was a much better car, that happened to have doubled in value if I'd bought one back then...

In reply to OHSCrifle :

It's been a while, but I remember the timing chain tensioner being a very easy job. If memory serves me correctly, I felt like it only took me 5-10 minutes. The fact that it doesn't stand out in my mind tells me that the pain was minimal lol. 
 

Yeah, it's a very big spend. If I just wanted to do the bare minimum in terms of maintenance and wasn't planning on tracking the car, I likely could have gotten away with half of what I spent. I went full over-kill, hoping to make the car reliable on track. 

In reply to docwyte :

You're not wrong lol. Although I keep a detailed list of everything I buy and what it cost me, which I find to also potentially be helpful when it comes time to sell the car. 
 

It's a sobering number to say the least. If this was a street car only, I wouldn't be able to justify it- as there are far too many cars out there for $20k that are better to drive (namely a nicely kept 135i lol). For something track focused, honestly, this is what any car would have cost me by the time I got done prepping it for track work:

-Bigger radiator

-Oil cooler

-Baffled oil pan

-All deferred maintenance completed to keep it reliable.

-Camber mods (plates & control arms)

-Wheels & tires.

-Suspension (even if I will likely replace it with something more track friendly). 
 

Realistically, no matter which car/platform I went with, this wasn't going to be cheap. I suppose I might have been able to save a bit of money if I bought a car that came with a factory oil cooler (even though I might have upgraded it anyway), a factory LSD and maybe an oil pan designed to take high lateral Gs without starving, but there aren't a ton of cars like that out there, and they usually cost a good chunk of change. 
 

I suppose my conclusion is that this hobby is very expensive and perhaps my salary level doesn't quite match the budget needed to be a top runner. 

So, if I want to continue down this path, I have options. 

I've consulted the hive: https://grassrootsmotorsports.com/forum/grm/17x85-vs-17x9-how-much-time-am-i-really-leaving-on-the-table/279129/page1/

Andy Hollis thinks running a 17x9 over a 17x8.5 would likely only result in a couple tenths of a second. Andy is a much smarter, more experienced man than I am. The Facebook E82 group seems to agree. As such, I'm now re-thinking a potential build strategy. 

The options: 

-TC Kline coilovers. Their single adjustable Koni based kit is only $2500, but after speaking with TC, he recommended their DA kit in the rear at the very least, bumping the price up to $3000. But the DA kit requires I buy M3 rear spring perches ($300), bringing the price up to $3300. Being that the car is still street driven and our tracks are a bit on the bumpy side, he recommends 300/600 rates (400/700 can be used, but I'm assuming they won't ride very well). This package includes camber plates, but that's still a lot for Konis. 
https://www.tcklineracing.com/webdocs/Items/Details978.cfm

-MSC TT1 single adjustable kit from Vorschlag. Due to a higher quality shock, they recommend running either 450/800 or 600/900 spring rates. They have options, either I run my stock non-M spring perches and they sell me an adapter for a grand total of $4779, or they sell me a set without adapters for $4499 and I have to buy M3 spring perches ($300), bringing the total up to ~$4800. This includes Vorschlag camber plates. 
https://vorshlag-store.com/collections/bmw-e82-92-suspension/products/mcs-tt1-single-adjustable-monotube-dampers-bmw-e9x-e8x-non-m?variant=33555608775

-Ohlins coilovers (more track focused kit). All Ohlins (to my knowledge) are dual flow valve. The more track focused kit from recommended vendors like 3DM run spring rates of 399/685 or 456/799, somewhat of a middle ground between the TCK and MCS setup, which is also reflected in the price, at $3700, which includes Vorschlag camber plates and everything needed to install. 
https://3dmmotorsport.com/collections/3dm-suspension-kits-bmw-1-series-e82/products/bmw-1-series-e82-3dm-trackday-suspension-kits

There's a downside to the Ohlins kit. The TCK and MCS kits use shorter springs, allowing me to run 9" wide wheels if I so desire. The Ohlins kits use a taller spring setup, which creates a bit of a clearance issue wider wheels/tires. Spacing them away from the suspension just results in clearance issues with the front fenders. 

I've been picking the brain of fe1rx on 1Addicts, who is running Ohlins, happens to be an engineer that has come up with some creative solutions. Here's what I've learned: 

-In order to gain additional tire clearance with Ohlins (for 9" wheels), I would have to swap out the front springs to shorter, 7" springs. Swift makes quite a few 7" springs that could potentially fit: 



-The E82 specific Ohlins (MI01 dampers) have a 103mm stroke before contact with the bump stop occurs. 
-It was recommended that I try to use a spring that has a usable stroke of 103mm or greater to avoid the spring becoming less effective (out of its usable range) or worst case scenario, coil binding.
-The 060 spring listed above has a usable range of 106mm (coil binding at 118mm). Due to its usable stroke, it would be a great fit with the MI01 dampers (although the rate is a bit lower than I would like). 
-fe1rx is currently using the 060 spring with Ohlins old MI00 kit (technically designed for the E9x, not the E8x) with its longer stroke (113mm), but by using minimal preload, somewhat effectively increasing the usable stroke. 
-If desired I could potentially use the stiffer 070 or 080 springs (substantially lower 93mm usable stroke, but don't coil bind until 119mm) by running a lower preload value. 
-If I was so inclined, I could disassemble the strut and use packers to decrease the stroke of the shock. The negative side effect of this would be a decrease in suspension travel. 
-While a 7" spring allows for more tire clearance, it also lowers the ride height substantially due to the lower overall spring length. Hence the use of the spacers fe1rx designed, which raise the ride height an additional 10mm, shown here: 




The shorter springs, combined with the use of spacers should allow me to maintain a streetable ride height, while being able to run 9" wide wheels. The issue is that the spacers cost $70 (dirt cheap) and the shorter springs cost an additional $200, bringing the grand total up to nearly $4000 for a set of customized Ohlins that can run 9" wide wheels. 

If I plan on sticking with my current 8.5" wheels, the Ohlins seem like a good option. Good, streetable ride quality and a major improvement in terms of handling on track (current Bilstein/Eibach rates are around 148/456). Without modifications, they'll only run me about $400 more than the set of Konis being recommended by TCK, admittedly the rears on the TCK kit would be DA, but they're still a pretty inferior shock vs Ohlins. 

If I ever wanted to go up to a 9" wheel, that's where things get interesting. Either I can modify the Ohlins, which by the time all is said and done, will cost me around $4000, or I can just jump straight into a set of MCS singles for $4800. 

Now, there are other kits out there for a good bit less money than the Ohlins that will run my current 8.5" wheels, but I'm hesitant to go down that path, as I'm not sure they offer enough savings to justify skipping the Ohlins and no one has seemingly engineered them to run a 9" wheel if I ever decide I want to go down that path. 

My questions: 

-If sticking with the 8.5" wheels I already have, are the MCS singles worth an extra $1100 over the Ohlins? I'm inclined to think they're not, but I don't have enough experience to comment one way or another. 

-If I ever decide I need those last couple of tenths (unlikely if we're only talking about a couple tenths), would it have been a better idea to have just bought the MCS dampers in the first place, as the price to modify the Ohlins brings them to within $800 of the MCS setup? 

-Should I look into some of the less expensive (~$2500) coilovers (KW, Bilstein, Redshift, etc) and just make a decision to stick with my 8.5" wheels long term? I suspect that by the time I added Vorschlag camber plates to any of these setups, I'll likely start getting pretty close to $3000. 

roninsoldier83 said:

In reply to OHSCrifle :

It's been a while, but I remember the timing chain tensioner being a very easy job. If memory serves me correctly, I felt like it only took me 5-10 minutes. The fact that it doesn't stand out in my mind tells me that the pain was minimal lol. 

Thanks

In reply to roninsoldier83 :

The only way around it is to buy a car that's already setup for track work.  I've managed to do that once, the '98 M3 I bought.  It was really the only car I've ever bought that was good to go right out of the box.  I really wish I'd been able to keep it....

In reply to docwyte :

Yeah, buying a fully built track car is certainly the easier route to go. Unfortunately, I'm so deep into the 128i that it really doesn't make sense to abandon the platform/car at this point. I would like to say that if I had it to do all over again, I would have searched for a fully built track car- it certainly would have been cheaper and less painful. The upside is the experience and knowledge I gained from undertaking this two year long project. I'm not sure I would ever really want to bring another car back from the near-dead like this again, but it did help in making me less intimidated to do large jobs in the future. 

Why did you sell your E36 M3? 

I have a 128i that has followed a really similar track to yours, except I added safety equipment, coilovers, and a LSD. I am thinking I might sell it in favor of a NC2 because I miss the simplcity of a Miata and even with extensive work, somehow the E82 doesn't feel "right".

Additionally, I am like you in that I am tired of the buy, build, break, fix, sell cycle. 

In reply to YoursTruly :

I currently have both my 128i and an ND2 Miata. I'm 100% convinced that there's nothing I could ever do to make it nearly as much fun as the ND, short of figuring out how to remove about ~900 lbs worth of weight. If you're looking for the feel of a lightweight roadster, nothing else is ever going to feel "right". I recently traded my NC2 in on my ND2. 

The NC2 is a great car. Tremendous amount of roadster bang for your buck. They also have several SCCA specific classes in which they're the "car to have" and a large emerging aftermarket. 

I've owned all 4 generations of Miata- I find the ND2+ to be the most fun, but I wouldn't kick any of them out of my garage. 

 

In reply to roninsoldier83 :

I wasn't using it enough.  There was no interior behind the front seats, which were race buckets and it had a roll bar in it.  Perfect track car, but it sat around the other 51 weeks of the year.  I debated pulling the roll bar and seats and making it more stock, but decided a 996 Turbo would be a more fun street/track car...

It really sounds like you're at a crossroads and you need to figure out what you really want from this car.  Are you trying to improve yourself or the vehicle?  If its yourself, I doubt that spending money on an Ohlins setup is going to help much with that.  If its the vehicle, you probably will want to ask yourself whether or not the 128i is the right platform to make a full on track monster.  You're always going to HP limited, and at some point, you're going to want to improve the safety of the car and then you won't want to street drive it anymore.  

Forgetting all that above, what about swapping springs around on the B12 kit?  You could probably cobble up some kind of coilover conversion for the front with some kind of sleeve kit like this: 

And you can easily put coilover springs and a height adjuster on the rear.  I've considered doing that to my wagon with springs that are similar in rate to stock because the rear is kinda saggy, its like ~$200 for some height adjusters and $100 a spring.  If you really decide to go bananas, I'll bet you could find someone to revalve your Bilsteins for less than the cost of a set of Ohlins.  

In reply to rothwem :

My goal with the car is a bit more objective at this point. I have a few autoX trophies on the shelf in my garage, but no TT trophies. I would like to win a regional TT championship. That would involve improving both the car and the driver. In the past, due to some unforeseen circumstances, I've never stuck with TT long enough to reach that goal. 
 

I would also like to not completely ruin the car as a car, if possible. Keep in mind, I have no truck and I have no trailer; translation: towing my vehicle(s) to the track isn't an option. Even if I had a trailer, I have no place to store it- my HOA would fine me if I tried to keep it on the property. Some of the tracks here are 3+ hours away (namely La Junta Raceway, PMP is also ~2 hours away and HPR is still over an hour). I've made that drive in a car with a super stiff suspension, no A/C and no cruise control. I'm not interested in doing that again if I don't have to. I would rather run in a class that allows me to keep a true dual purpose car. 
 

As far as the platform goes, it really depends on the class. The current class I'm running with is an SCCA class with a modified power to weight ratio rule set. My 128i slots perfectly into one of the classes there (Production B). There's no reason I can see to change platforms. Completely by accident, I've maxed out the power to weight ratio for the class- meaning I don't have to spend a penny on power or weight reduction mods and I can keep a full interior and A/C, while still maintaining a slight power advantage over pretty much every car in the field. 
 

For comparison, the field consists of 1st gen BRZ/FRS, E36 328i's, RX-8's and a mid-90's 2V V8 Mustang. The twins and RX-8's are definitely more agile, have a lower center of gravity and naturally better handling... but that's offset by the fact that my 128i puts down 221whp, which is more than either of those platforms could ever hope to put down without forced induction. Fully modded, there's no reason to believe I can't keep up with or beat them all. 
 

I could use my much lighter ND2 Miata, but that just bumps me into a faster class... same with getting another S2000... I love RX-8s but they're pretty gutless, a substantial reliability downgrade and if I money shift on track, the Renesis is far more expensive to replace vs the N52... same goes with the twins to a lesser degree. 
 

For the record, after giving it some thought, I'm no longer leaning towards Ohlins, I'm now leaning towards the MCS dampers. In the grand scheme of things, it might be worth delaying gratification for an extra few weeks to get the proper setup to begin with. 

So… I’ve been going back and forth on this for a while, but I think it’s finally time to say it out loud: I’m seriously considering selling the 128i. Again. Yeah, I know. 

When I bought the car two years ago, the goal was pretty clear — build a dedicated track car. And in a lot of ways, I kind of did. I sunk a ton of time and money into sorting out every major mechanical issue, and then followed that with reliability upgrades, a square wheel/tire setup, and plenty of prep aimed at eventual time trials use. The car’s in a great spot mechanically — reliable and consistent, both on and off track.

But here’s the thing: somewhere along the way, my goals changed.

I’ve realized I don’t really want to chase lap times anymore. I took the car to an autoX and couldn't get into it... and while time trials has the potential to be fun, between the cost of making the car truly competitive (expensive suspension + corner balancing, solid rear subframe bushings, LSD, safety gear, more tire/wheel setups, track insurance, etc.), and the reality of how much I’ve already spent, the idea of pushing deeper into time trials just doesn’t excite me like it once did. The sky-high costs honestly just add to my anxiety these days (as it turns out, I'm not a wealthy man!). When I added up all the cash I've spent on the 128i, I realized I also need to start being more diligent about buying track-day insurance, which essentially doubles the cost of an event. Can I afford this hobby? Well, kind of, but it would be at the expense of other, more important, areas of my life. 

At the same time, the recent upgrades — especially the stickier RT660s — made the car feel more capable but less fun on the street. I don't find 200tw tires terribly playful on the street. And when I compare it to the rest of my garage, that’s where the 128i starts to feel redundant:

• 2020 ND2 Miata Club (Brembo/BBS/Recaro) – honestly hard to beat in the fun department

• 2021 Mazda 3 Turbo – my semi-daily

• The 128i – capable, but currently without a clear purpose

That puts the 128i in a weird spot. It’s no longer fun enough for me to reach for it on backroad drives, and it’s no longer the track car I thought I was building. I’ve kicked around the idea of dialing it back with a more streetable tire and maybe adding a 3.73 diff (and LSD) to bring back some playful character. But I also wonder if I’m just trying to justify keeping a car I might not have much use for anymore. 

Anyway, I wanted to share where my head is at for those following the thread. This car has come a long way, and part of me still loves it. But I’m starting to think it might be time to let someone else enjoy it and use the money (and mental bandwidth) elsewhere... even if I also dislike the idea of pouring so much of my time, money and hard work down the drain. 

Appreciate all the support and input over the build — I’ll keep this thread updated as I decide.

Time to pump the brakes on my semi-annual "I'm gonna sell the 128i" post. 

A friend of mine just reminded me of why I started my blog ~7 months ago and why I started doing track reliability mods on the 128i in the first place. It wasn't supposed to be confining myself to a ruleset. It was supposed to be about adventure, traveling to race tracks across the country. I can't believe I lost sight of the very first post I made on the blog: 

https://www.investigatingtheapex.com/post/where-it-started-and-where-we-re-heading

After all the time, money, and sweat I’ve put into prepping my 128i for the track… it was finally time to ask: how does it actually feel in the real world?

This weekend, I took my lightly-modded E82 on a proper canyon run with the BMW CCA through Estes Park, Devils Gulch, Glen Haven, and down to Drake—and I may have just built the perfect canyon car by accident.

Is it better than my ND Miata in the mountains? Shockingly… yes.

Is Miata still the answer? Not this time.

Did my 128i finally redeem itself after a mediocre first track outing? Absolutely. 

This one actually surprised me just a bit. 

Read the full write-up here: 

 

Well, it's been a while. It's finally time to start a project on the 128i that I've been talking about for a couple of years now. 

It all started in a junkyard... which, seems to be a theme with this rolling pile of bolts. 

A junkyard about 40 minutes north of me recently took in an automatic 128i. For those that aren't familiar, the E8x/E9x cars come with a variation of final drive ratios: 

128i/325i/328i Manual RWD: 3.23
128i/325i/328i Automatic RWD: 3.73
325xi/328xi Manual X-Drive AWD: 3.38
325xi/328xi Automatic X-Drive AWD: 3.91

330i (2006 only) Manual: 3.15
330i (2006 only) Automatic: 3.64

All of these cars come with the "small" rear diff and are interchangable with each other (although AWD cars would obviously have to swap both front and rear to avoid drivetrain issues). Although I believe the 330i diffs have a slightly different input flange/pinion yoke, but I'm pretty confident you can swap over parts from your stock diff to make the 330i diff(s) work as well. 

The turbo cars (135i/335i) come with larger diffs that I don't believe bolt in without swapping driveshafts (not sure if the axles are compatible). Same with the M3/1M. I'm not sure about the diesel 335d, but given the sizable amount of extra torque load, I would assume it also comes with a beefier housing. 

After punching some numbers, I decided I wanted to run 3.73 gears, which should give me an extra ~15% torque multiplication, and let's be honest: as a long-time Honda and Miata guy, I'm just a sucker for short gearing. Here in Colorado, most of the BMWs on the road (and in junkyards) are AWD, so I had to keep my eyes open for a non-turbo RWD automatic car. Not an extremely rare car, but it does require a bit of looking. When this car popped up, I decided it was time to take a long-overdue trip to go rip out a diff: 




It's been a while since I've been to a junkyard or even touched a wrench. I clearly didn't have my head on straight today. I grabbed a set of everything I thought I would need, but didn't go overboard. Note to self: always go overboard. 

I also didn't bother watching a video or researching what tools I would need. I mean, it's a rear diff, how hard could it be? Oh how my arrogant idiocy rears its ugly head! 

I want to be clear, if you're working on a nice garage floor (rather than laying in gravel and broken glass) with a lifetime worth of proper tools, this shouldn't be a terrible job. Yeah. You know what's coming next... 

So, the rear of this fine junkyard machine was sitting on top of a couple of welded together steel wheels, as you would expect at such a facility. Why does this matter? The wheels were directly crushing the center-mounted exhaust hanger. Translation: since I couldn't get to the center exhaust hanger and didn't want to try and work in the area surrounding it (for fear of the car crushing down on me), I couldn't fully lower the exhaust. Which, wasn't the worst thing in the world, but moving the exhaust fully out of the way would have made this job a hell of a lot easier. 

I had to get creative to get some of the eTorx axle bolts off (due to the axle being in the way of the bottom bolt and the axles not spinning freely), but all of them broke loose without too much of a fight. 

I also managed to get the four eTorx bolts to the driveshaft off without much of a problem either. I thought everything was going relatively smoothly, minus the exhaust being in the way. 

There are only three bolts actually mounting the diff to the subframe. The rear two are 18mm hex bolts that zipped off no problem. The front bolt? It's a 21mm hex bolt with a 21mm nut on the other side of it... I brought a 21mm socket... exactly one of them. I also brought a set of ratcheting wrenches... that only went up to 19mm. After brainstorming and trying everything I could think of, I came to the conclusion there was no way I was getting that nut off without another 21mm tool. My house was 40 minutes away. Parts stores were much closer, but after realizing there were more tools I should have grabbed to make this job (and another job I was planning, more on that later) easier, I decided to make the long trek back to the house to get extra 21mm tools. 

Upon returning, I got the 21mm nut and bolt off almost instantly. The junkyard kindly allowed me to bring a transmission jack (they usually don't allow traditional jacks, but I assured them it was far too flimsy to lift a car). I went to lower the diff and discovered that the extremely rusty tail-end flange appeared to be rusted onto the driveshaft. Lots of profanity, pulling and wild kicking ensued, to no avail. I brought a mallet, but in retrospect, I wish I had come prepared with a legitimate BFH. I also brought a long flathead screwdriver to act as a pry tool, but was wishing I brought my chisel set. Yet again, when going to the junkyard, always go overboard. 

I found I actually had to lift the diff back up (good thing they let me bring my tranny jack in!) and eventually, after fighting me for ~20 minutes, the diff was free: 



I'll likely clean it up and do something about the rust on the tail end before I go any further. 

As a bonus, when we looked inside the car, we saw everything had been torn apart and three perfectly good vents were just sitting on the floor, waiting to be snatched up! I had actually put new vents on my list of parts to replace, as my stock vents have a broken slat and are extremely weather faded in comparison to the rest of the dash. By comparison, these were in much better shape! 

On the same lot, I also managed to come across a 328xi/X-drive. You might be wondering, why in the world would I be looking for such a car? Well, remember the last time I was on track and complained about the excessive body roll and understeer? I figured an easy way to help with both problems without spending a ton of cash on a set of stiffer coilovers is a thicker rear sway bar. 

Just like the rear diffs, these cars come with several variations of rear sway bar: 

Base 128i: does NOT come with a rear sway bar
128i w/Sport or M-Sport package: 12mm rear sway bar
E9x RWD: 13mm rear sway bar
E9x AWD: 15mm rear sway bar

For reference, the M cars (M3/1M) come with an absolutely massive (by comparison) 22.5mm rear sway bar.

I've heard of some variation of an E9x that comes with a 14mm rear bar (a sport package maybe?), but I didn't see the point in searching for such a unicorn. 

Most people that track these cars recommend you try running the smallest rear sway bar you can, controlling rotation and body roll with spring rate. For a dedicated track car, that's probably good advice. On the street, I actually like the way my Bilstein B12 kit rides and the way it controls the car on back roads for the most part. I'm not keen on parting with this setup anytime soon. 

Most folks will tell you that with the stock open diff, running a bigger rear sway bar on these cars will just increase the car's tendency to one-tire-fire coming out of corners. I believe it. But, along with the diff I just picked up, I'll be putting in an order for a Wavetrac LSD to swap into it. Translation: no more one-tire-fire means I can run a bigger rear bar. Which is probably why BMW gave the M-cars substantially upgraded rear sway bars: they come with LSDs stock. 

I wasn't sure that I really wanted to jump from a 12mm sway bar to a 22.5mm rear sway bar, as it's a massive jump that might have required a full suspension re-tuning. So I decided to skip the M3 bar. I considered an aftermarket setup, but since this is essentially a street car, I like the idea of OEM bars (hollow=lighter and they come with rubber bushings that never need to be greased). 

Doing a bit of not-terribly-precise math, I estimated that the 12mm rear sway bar on my 128i (I believe all the OEM bars are hollow) was likely good for about 50-82 lb/in of rear spring rate- about as much as a wet noodle. Jumping up to a 15mm rear bar should bump that up to around 120-200 lb-in in the rear; which should certainly be noticeable. 

In order to get the rear sway bar out, you do have to drop the rear subframe. If you're trying not the damage the brake lines and parking brake lines, it seems like it could be a bit of a tedious job. On the other hands, if you're in a junkyard and don't really give a damn? 10 minutes. No joke or exaggeration. I literally had that bar out and in my hands in roughly 10 minutes- with ease. 

Rear subframe dropped without an issue: 



Bar came right out: 




I was expecting the rear diff to be an easy job and the dropping of the rear subframe to be a pain in the neck. I received the exact opposite. 

How much did this entire treasure trove cost me?

-3.73 rear differential: $97.95
-15mm rear sway bar: $30.75
-All three interior vents: $16.00



For reference, the used OEM vents go for about $130 on eBay ($70 for the center and $30 for each side vent), the rear diff goes for about $300 on eBay and OEM 15mm rear sway bars go for about $90 shipped. So a few hours at the junkyard saved me roughly ~$375 vs ordering used parts off of eBay. Not bad at all. A much better deal if you come with all the tools you need the first time! 

I considered skipping the rear sway bar, but I plan on swapping over to solid aluminum rear subframe bushings, which requires dropping the rear subframe anyway... so, I figured, "while I'm in there!". 

And now, after watching a plethora of YouTube videos, I just have to order a Wavetrac LSD and try not to mess anything up while swapping it into this diff in my garage. Fingers crossed! 

That's a score on the rear diff!