Yesterday I was installing new brake lines into my 1954 Willys CJ-3B, and while it was quite a simple job, there is one section of the car that’s just a massive pain in the butt, and it got me thinking about all the engineering packaging decisions that have compromised serviceability and thus lost me hours of my time.
Specifically, where the brake master cylinder bolts to the frame is a bit of a shitshow on an old Willys Jeep. The master cylinder is sandwiched between a bracket and the frame via two bolts that thread into — often brake inside of (after rusting) — the frame. Since my Jeep is from California, I narrowly managed to thread those two bolts out, but once the master cylinder was in, routing new brake lines was rough.
The thing about new brake lines is that they’re rarely perfect, meaning you have to fine-tune some of the bends to get them to meet the fittings. My issue was that the brake line has to go through a small hole in my frame before immediately mating up to the master cylinder. Here you can see it from up top”
Here’s a side view:
And here’s a little closer look:
Trying to bend the line, then thread it through that small hole (which already has a hose going through it) took me forever, especially since the fitting was right there after the bend, and especially since I had to do this while on my back.
Another classic example of a packaging decision that has ruined a few of my afternoons is the oil filter housing location on post-~1992 Jeep Cherokee XJs. These things leak like a sieve, and to fix them one has to remove the oil filter adapter from the engine block; it’s held on by a single torx bolt, whose head is so ridiculously close to the unibody rail that you have to either buy a special tool (shown below) or custom-make your own tool to remove it.
I myself had to buy a torx socket, cut it down with an angle grinder, then use a box wrench to just barely fit between that rail and the oil filter adapter.
Don’t even get me started on the upper shock bolts on the Jeep Cherokee XJ, which tend to break in the weld nuts captured in the unibody. The result? You have to use an air chisel to break the weld nut and then you have to fish the nut and broken bolt out from behind. Then you have to fish a new bolt in through the top and hold it in place while you thread a nut on — it’s rough, as YouTuber D&E In The Garage shows below:
And then there was my multi-day nightmare with my Chevy HHR control arms!
As with many wrenching endeavors, if you have the right tools is key, and once you’ve figured out how to do it the first time subsequent fixes are usually much easier.
Still, this is a thread for us to complain about tight packaging making repair jobs a bear, so have at it in the comments!
1. Removing/replacing the EGR on a Prius (stud from hell).
2. R/R or checking fluid level on the power steering pump, 1998 Olds 88
3. Impossibly small access hole to R/R rear bulbs on Toyota FJ Cruiser
4. Heater core, early Fox Mustang (already mentioned, but worth a 2nd vote)
5. R/R fuel pump on a Dodge Stealth
Honorable mention (not for difficulty of access, but placement) power steering pump mounted above alternator on Chevy trucks w/250 6cyl motor.
Best design ever: headlights on Chevy GMT800 trucks; unclip snd slide a bolt, whole assembly comes out without sharp edges or cramp-inducing hand twisting.
This is why I tried to stop buying newer American cars. My dad had a trailblazer everything was a fight with that thing. You just looked at it and went why would you do that. I’ve worked on other gms of the same era and unless it was the gmt800 trucks it was nope. It’s funny now because a lot of claimed the 4.2l is the American 2jz it’s an Isuzu engine and that is the best part of that SUV. Alot of Ford’s of that era I remember fighting over stupid things going who did this. I’ve had to pull a cab on a ram and while it wasn’t exactly fun it wasn’t the end of the world. The Japanese and Korean stuff is just so easy to work on except Nissan that’s a whole bunch of no.
This is why I prefer to hobby on cars built before 1970… and the older the better. Replaced a head on an ‘87 Dodge Aries 2.2 and it took days of ham-fisted mucking around with accessories and the timing belt. (And that was a “simple” car.) Changed a head gasket on my son’s 1929 Model A and it took three hours because we were slow. Favorite trick with the Model A was shoving coiled rope down on top of the pistons through the spark plug holes, then cranking the engine to pop the head loose. Yes, this was recommended in the repair manual.
The 7mm allen head bolts securing rear VW/Audi calipers that require you to grind down a socket to access definitely get an honorable mention. Was annoying in 2004, still annoying in 2019.
Hmmm. Clutch master cylinder job in my old E28 was not enjoyable; getting the cylinder mounted on the pedal bracket was hard enough. Getting the hard line attached without cross threading was harder. Doing the job 3 times before discovering the crack in the pedal bracket that was causing failure was a learning experience accompanied by much swearing and gnashing of teeth.
More recently, the rear shocks in my ’06 Sequoia were a complete no go for me. There’s about 3/4″ between the bottom of the body and the frame rail to get a wrench on the top shock nut. I took one look at that rust covered piece of shit and said nope. I’m assuming the shop I sent it to cut them off to install the Ranchos I got.
I’ve worked for BMW, Audi, VW and Stellantis
How much time you got?
Jokes aside, the new inline 6 from stellantis is such a shitshow of terrible layout choices.
You know how on their old v8s and v6s they just put the thermostat at the front top of the motor in the open? maybe takes a half hour tops to replace it and be on the road.
So for the inline six they put in at the bottom of the motor, bolted to the back of the water pump and for extra “fuck you mechanics” covered access to the bolts with hard coolant tubes. The book is 3 hours I think, and can take longer than that. Double bonus; they’re unreliable and fail stuck closed!
Captive brake rotors could only have been designed by the devil himself.
Having to press the bearings out of the knuckle to replace a rotor is nothing but spiteful meanness.
Not sure all vehicles they were on, I’ve encountered them on the Ford festiva that was a Mazda Kia mashup mentioned in today’s morning dump and also older Hondas.
Not that hard to fix, but a stupid way to break, repeatedly, was the 90-degree fan belt routing on Corvair engines (with its vertical crankshaft pulley and horizontal cooling fan pulley). https://www.youtube.com/shorts/CMD5RQJngT8
Recency bias!
The cam phaser leak just happened on my 2016 Sonata. Repairing it requires supporting the engine , pulling the engine mount, and replacing the parts.
The upper intake manifold needs to come off to replace the plugs on my wife’s ’13 Sienna. They’re almost due.
And heaven help those who need a head gasket job on the Sienna. Full engine drop needed. 23.5 hours book rate.
If you are still having nightmares about the control are bushings on that HHR i raise you a PT crusier. Those are traped in the frame rail of the unibody and are held together with a clip that rusts and breaks. The only way to get to the nut is cutting the frame rail open and claping the nut with vise grips. Also the hell that is a water pump and timing belt job.
I had a first gen Audi 5000 (circa 1979) and the parking brake cable went diagonally through the gas tank
I mean, at least 70% of every repair I’ve ever done on a 21st century VW product was in some way hampered by an irritating engineering decision.
Soooo… my 1963 Chrysler 300J.
Let’s just make the engine, SOLID LIFTER and long ram manifold. Take the entire intake system off (8-12 hours) to set the valves. Granted, back in the day ( and lots today) definitely performance focused, but sheesh!!