To whoever the asshole was that did the last timing belt replacement on this car: I hope you rot in hell

There's a certain amount of, for lack of a better word, forensics that goes on whenever we work on this car. It's over twenty years old and it's not exactly low maintenance. It has had four or five previous owners, at least some of whom did their own work, and it's not an easy car to work on by any stretch. So every time we open the hood, we play a game well known to software engineers called "Where are the bodies buried?"

Yesterday there were a lot of them.

We decided to replace the timing belt after we tried to do compression test and discovered that not only was the timing horrifically misadjusted, we couldn't correct it. We would get close, and then cogs would slip because there was so much slack in the belt. We never bothered to put the engine back together after the first compression test so we figured it wouldn't be too hard to continue taking pieces off, get to the timing belt, and put it all back together.

Well, I didn't think it would be that hard. If the BF had other thoughts, he kept them to himself.

1991 Mitsubishi 3000GT VR-4 Timing Belt Change Statistics

Start time	2:00 PM
Stop time	4:00 AM
Belts replaced	Timing, Power Steering, Alternator
Weight Shed	.5 lb - part of the air conditioning pulley
Fluids Replaced	Coolant
Fluids spilled all over the engine, the engine jack, the drop cloth, the floor, the people doing the work	Coolant
New spider webs built from car to ground during work, total	1
Mosquitoes squished, total	10, approx.
List of atrocities committed by previous owners and/or mechanics	See list, below
	engine bracket bolts not tightened
	engine mount bolts not tightened
	existing timing tensioner not adjusted
	existing timing belt not tight
Front Bank PSI before timing belt change, average	181
Rear Bank PSI before timing belt change, average	90
Front Bank PSI after timing belt change, average	165
Rear Bank PSI after timing belt change, average	163
PSI range across all cylinders before timing belt change	90
PSI range across all cylinders after timing belt change	10

We spent fourteen hours yesterday to get to this point. That's not a typo. The car was already partially apart for previous compression tests and we still have four to six hours of work left to get it back together. I took a bath in coolant. For a while it was looking like I'd have to Gojo my hair. 

This is what I looked like after about six hours:

If you are guests of ours and you always wondered why we send you to the bathroom at the end of the hall rather than the close one right next to the front door, this is why. We use the close one to clean up messes like the one in this photo.

But this is the only photo that matters:

Compression test results: EXCELLENT. Conclusion: ENGINE NOT AT ALL RUINED.

P.S. That bottle of Pink Soap in the first photo is for cleaning oil painting brushes. I use it to clean myself up after car work because it's not full of grit and it's a lot gentler and more effective than Gojo. And it doesn't smell like oranges. Yesterday I showered with it. I was that dirty.

More timing

First, the scope of the work. We have a spreadsheet to keep track of what we need to do. It looks like this:

As you can see, we have some work to do. 

When we last left our heroes, they (we) had just performed a compression test with less than stellar results. Then we had a bunch of weddings to go to and we didn't get time to work on the car for a while. But we finally got a few free hours to try to work out the timing and perform a new compression test.

The problem with compression tests is that they're worthless if the engine timing is off. You're not going to get much compression if valves are open when you don't expect them to be. So step one was fixing the timing.

This involved shimmying a piece of cardboard a cam shaft gear and the timing belt, sliding the gear into correct timing, and removing the cardboard. This had to be done for 3 gears. It took a while.

And then it didn't work.

We've suspected for a while that something is wrong with the timing belt tensioner. The tensioning changes when we turn the engine over a few times (slack moves to various different parts of the belt) and the belt jumps teeth at the lightest touch. When we adjusted one cam shaft gear, others slipped. We couldn't keep one gear in time long enough to fix the others, and even when we got close, we still didn't really know if we were at all in time with the cam shaft that drives the pistons.

So now we've spent a week getting endless boxes of parts, and before we do anything else we're going to replace the timing belt and all that goes with it -- tensioner, water pump, and anything else we can think of. 

Then, maybe we'll get a worthwhile compression test.

The latest and greatest in 20th century blogging technology

I've finally added subscription links for RSS feeds. They're on the sidebar.

Commence the biting of nails.

At this point, a vastly oversimplified explanation of engines is in order. In an engine, there are hollow cylinders (6 in our case). They are arranged in banks (2 in our case, and yours too if you’ve ever seen ‘DOHC’ on your car anywhere). These cylinders have valves that open and close to let in fuel and air and discharge exhaust. Each cylinder has a piston that moves back and forth inside it and its function is to, among other things, work in concert with the valves to control the density of gasses in the cylinder. The orchestration of this mechanical dance is called timing and it controlled by a timing belt mounted on timing cogs which control cam shafts running though the cylinder banks. 

A compression test is a commonly performed diagnostic procedure which provides a lot of information about the state of the engine and how well it’s doing its job of producing compression. 

So we did one.

As this is a blog about our experiences learning to work on cars and not a how-to, I’m skipping over all the irrelevant stuff about how a compression test is done (per cylinder: remove the spark plug, insert a special compression testing tool in its place, disable ignition, and turn the engine over for about 5 seconds) and the equally tedious recounting of all the stuff we had to disconnect and/or remove in order to get access to the spark plug banks. The tutorial on how to do this ran 16 pages* and it still didn’t account for the effects California’s special emission standards on engine design. We had some extra mystery hoses that we couldn’t identify for a while. 

Normal compression for this car is spec’d between 115 and 150 psi per cylinder. On a new car, we’d expect compression across all the cylinders to be at the high end of that range, with maybe a 10% psi variance from cylinder to cylinder, if that. If one cylinder were to drop more than about 10 psi from the group, we’d know something was wrong with that cylinder and we’d start investigating the spark plug more closely and worry about malfunctioning values. 

Front bank compression (psi): 180 178 182

Rear bank compression (psi):   92   93   88

What we can read from this is that all the cylinders within a bank are extremely consistent with each other but the banks themselves are wildly different. So, looking at these results, it’s very possible that there’s nothing wrong with the engine itself and we just need to adjust the timing.

Except for one thing.

This engine is classified as an interference engine, meaning that during normal operation, the pistons and the valves occupy the same space at different times. It is only by the grace of highly advanced mechanical engineering that they do not attempt to occupy the same space at the same time. 

Oh, and timing. 

The timing belt is the only thing preventing a piston from shoving itself as hard as it can straight into a valve. So if the timing belt were to break or become ill-adjusted somehow, the engine might bend all its piston rods or shatter all its valves or both. The failure mode is, to put it lightly, catastrophic.

So in that context, these compression results were terrifying. On the one hand, the internals of the valves may well be fine. On the other hand, the timing is way off (and steadily getting worse, because the timing belt seemed to slip a tooth every time we accidentally touched it). So in our quest to determine the state of the engine, we might have accidentally made things a whole lot worse.

*Just to get to the spark plugs. Not even to do the compression test.

State of Play

The last post was all intro, and this post will explain why this car deserves a blog.

We bought a project car. We knew that. It's almost 25 years old, it has almost 100k miles on it, and even a mint condition specimen could stand to use some work if you want to put it on the track, which we do.

This car is not mint. Not even close.

We bought it on a Saturday, in Santa Cruz. Since it was my dream car and the BF is a really nice guy, I got to drive it back home, up Hwy 1, across 84, up Skyline, and over the 92 bridge. Since the A/C didn't work, I had the windows open the whole time. I was in the GT, and the BF was in his 300+ hp sports car following me. It was like a Top Gear special. It was glorious.

I spent dinner alternating between gazing off into space and explaining to the BF how I would completely understand if he decided the car would be too much work and we should sell it and find something else, but that he would have to be the one to make that decision because I had completely fallen in love with this thing and I was willing to do whatever it took to make it run like a dream. He laughed at me all through dinner.

Sunday, I took the motorcycle out on a test ride for an upcoming bachelor party event (which probably deserves its own blog post and I probably won't get to it, so ask me the next time you see me in person) and the BF followed in The Car. Once again, it was sunny and warm and the roads were amazing and twisty and there was no one on them and the scenery was incredible.

And then I pulled over to swap directions and when I looked back, the BF had the hood up.

"I have no oil pressure at idle."

Testing ensued and the conclusion was reached that the car wasn't leaking oil, so he contented himself with theorizing that the previous owner had used the wrong weight oil and mildly revving the engine at stop lights on the way home. Then we put the car up on jack stands and took the wheels off.

That's when we discovered that the car needs two new front tie rods and a new CV joint. Basically, a new front steering rack.

I changed the oil while that was going on (to the correct weight)*.

We pulled the back wheels off and mercifully found nothing wrong there.

Last weekend we did a compression test, discovered the timing belt was incorrectly tensioned (a Very Bad Thing on an interference engine) and found a 90 psi different between the front cylinder bank and the back cylinder bank. This is concerning.

Today, we got results from the company that analyzed our oil sample:

"We're not ready to call this one a lemon yet, but there are some areas of concern. Metals are high, especially copper, lead, and tin (bronze), since they could show bearing wear.  If there's a bearing issue, that could be related to the low oil pressure.  Iron (from steel parts, like shafts) is also high, as is nickel (probably a coating/alloy).  Averages are based on ~2,100 miles of oil use for the 6G72 engine. The viscosity was a bit low for 10W/30, but we don't know if it sheared down or if it was 5W/20 to begin with. The TBN was okay at 3.6.  Check back in 2,000 miles to monitor."

We are desperately hoping this will all be addressed with correctly weighted engine oil.

Tonight, we attempt to manually rectify the timing for long enough that we can do another compression test.

*One of the things I find fascinating about writing is what authors choose to reveal and choose to hide. I can, for instance, just say that I changed the oil on the car like it's no big deal. I do not, just as an example, have to go into all the details about how the oil pan nut is on the side of the pan and how I didn't really appreciate the fluid dynamics implications of that positioning. Nor must you suffer through tedious renditions of hot engine oil shooting out sideways, completely missing the catch pan and drenching my arms up to both elbows. Even the resulting inch-deep pool of dirty engine oil on the visclear we had mercifully laid down can be elided, as can the description of what it's like to clean that stuff up as best one can, generating a small mountain of dirty paper towels in the process, and then having to lie on an oil-soaked piece of plastic for the next hour. Isn't it neat how I didn't have to say any of that?

Harbinger!

I'm back!

I've got a new project and I need a blog for it and I saw this thing kicking around in the corner. So, without further ado, please say hello to the newest member of the household.

Yes, this is now officially a car blog. And that is a 1991 Mitsubishi 3000GT (a.k.a. a Dodge Stealth). The VR-4 edition, of course. The one with 300+ horsepower and all wheel drive and all wheel steering. Its name is Harbinger and the Boyfriend won't tell me why. Apparently I haven't watched enough Battlestar Galactica.

...and no air conditioning and almost 100,000 miles and very cracked leather seats and a 45 lb weight in the trunk masquerading as a subwoofer. 

I have always wanted one of these cars, and always for the most questionable of reasons. It started when I was about 16 or 17. I remember the exact moment. I hadn't been driving long. I was cruising down the left lane of 202 South back home, returning from the KoP Mall (I told you I remembered the exact moment). I was in my Taurus (also a '91, as it happens) and because I was a) new at driving and b) in the left lane and c) in a car so slow and underpowered I would play heavy metal to get it up hills (it seemed to help. I don't know why) I was checking my rearview mirror a lot. A LOT. And there was no one behind me, which was just fine. 

And then, the next mirror check half a second later revealed an inky shadow made out of solid black metal. It was low and dangerous and it had come out of nowhere. So I pulled over to let it past and on the back it said "Stealth" and I decided there and then that it was the most fitting name for a car I'd ever seen. And I really really wanted one.

Fast-forward to a couple years ago when I developed a Gran Turismo habit. For the unfamiliar, it's a video game. You race cars around tracks. With the money you win racing, you can buy better cars. And it turns out that in that game, the 3000GT is a really good buy. It's cheap, it's powerful, and the AWD means it corners like nothing else out there. I spent a lot of fake money and very real time upgrading everything about my virtual 3000GT - suspension, flywheel, tires, turbo, etc. And I won a lot of races.

The thing is, I never really got into real cars. I like them. I like driving them. The Boyfriend has a very nice sports car and I love taking it out. But I haven't owned a car since I moved to San Francisco 7 years ago. And even when I did own them, I never worked on them. I never had the time. But I've always wanted to be someone who is into cars and has owned a bunch and does her own work on them and so on. 

In a household with barely a two car garage and no shop space, these are dangerous thoughts so I kept them to myself. But then the BF said something about replacing his car. I asked him what he wanted, and this happened:

BF: "Well, I've always wanted to be someone who knows about cars. I want to know why I like my car and I want to know what it's like to have owned a bunch of cars and work on them and maybe track them. And the only way to do that is to start buying old cars and working on them."

Those of you who know me will recall that I am an opportunist in the extreme, and I got opportunity served up on a silver platter with that comment. 

So, long story short, we spent a couple weeks on Craigslist and found this thing. We brought it home about 10 days ago and thus began our odyssey into Working On Cars.