Ubuntu + Hildon UI = in-Car PC UI

Awhile back, Ubuntu announced a mobile and embedded edition of it’s popular Linux distribution. The buzz was around the possibility of Ubuntu Mobile showing up on future UMPCs. The news caught my eye, but didn’t really get my attention until the plans for Ubuntu 7.10 (Gutsy Gibbon) were announced:

“Ubuntu 7.10 will be the first Ubuntu release to offer a complete mobile and embedded edition built with the Hildon user interface components” (developed by Nokia for the Maemo platform.)

Now that’s interesting. Could it be that we’ll see Ubuntu Mobile booting on Nokia N800’s? It’s certainly a possibility — and one that could bring a larger breadth of software to Nokia’s mobile Linux tablets.

However, as interesting as it may be if Nokia adopts Ubuntu, the possibilities for wider Hildon support didn’t hit me until my drive home today. It was one of those obvious moments. I had been using my Nokia N800 while walking to my car, so the touch- and small-screen friendly UI was fresh in my mind. Then I started thinking about my Car PC. It uses a 7″ touch screen and runs Ubuntu (a full distribution, with a UI designed for full-size monitors.) Running Gnome on my cheap, in-car 7″ monitor makes for a pretty lousy experience. Text is hard to read, and everything is too small to click on. However, if this news is right, Ubuntu 7.10 will change all of that. I’ll be able to run Hildon on my Car PC! That’s killer. Imagine having Canola running in-car, sitting on 100GB of multimedia…

Texas heat and a coolant leak

Somehow I always end up working on the car during my holidays, and this past week was no exception. It’s been hovering somewhere around a million degrees in Texas for much of July, and with heat-soaked concrete and traffic jams, the temperatures can be a little harsh on our combustion-powered friends. Upon returning home on Sunday, my Subaru promptly expelled it’s coolant into a stinky mist of boiling fluid dumped onto the exhaust manifold. (Fortunately, the car had returned home unharmed before launching this little performance, since the carburetor gods have been less friendly to a few other folks I know. Over the past week, two have had their radiators blow, and one killed an engine [overheating so far as to melt some plastic bits in the engine bay and ruin the heads.])

Expecting the worst, I waited for the car to cool down, then popped the hood to have a look around. Puddles of coolant stood in every nook of the front-right corner of the engine bay, though the source of the leak eluded me. I was at least expecting to see a blown hose, but unfortunately, everything looked just fine. Since I knew that more probing was needed, I pushed the car into the garage to let it completely cool overnight.

The next morning it was time to call in reinforcements. In this case, they came in the form of a Snap-on Cooling System Tester. It’s not uncommon for a coolant leak to only happen under pressure, so a cooling system tester can be used to pressurize the system. The device is pretty simple — it’s just a hand pump, but it’s crucial for testing hoses and fittings. (Thanks for letting me borrow the tester Rob!)

The pictures below show the process:

Step 1: Pop the radiator cap from a completely cooled engine:

Coolant test: Step 1

Step 2: Attach the right adapter…

Coolant test: Step 2

Step 3: Attach the pump, and pump it. Watch the pressure gauge and don’t exceed the pressure your system is likely to run. In my case, I didn’t want to exceed 15 psi (~1.1 bar.)

Coolant test: Step 3

Step 4: Watch the coolant spray out the pin-hole leak like a fountain!

Coolant test: Step 4

As you can see in the pictures, the leak was pretty easy to find with the right tools, but would have been extremely difficult without them. Fixing it was even easier, and it gave me a nice excuse to flush the system and replace the other hoses as well (I went with higher-strength hoses for the replacements.) Also of note, you may have noticed in the pictures that I’m running a high-pressure radiator cap. (The STi cap is rated 1.3 bar.) This may have caused the blown hose, so do be aware of this when swapping caps.

If you’d like to replicate this at home, remember that automotive coolant is considered toxic once it’s been run in an engine. Always wear protective gloves when working with automotive fluids, and recycle them appropriately. (See the EPA page on Antifreeze for more on this.)

Since this repair, the radiator lasted another year before also developing a crack and blowing coolant all over the engine bay. The stock WRX radiator features a plastic top, which was the failure point for mine. Since all the coolant hoses were replaced with high-strength lines, that plastic top may have simply been the next weakest link in an over-pressurized coolant system. Now that the car has an all-aluminum radiator, I’m curious where the next weakest link might turn up (hoping that it’s not catastrophic!)

Increasing your mileage

I saw this on lifehacker yesterday and it’s been bugging me since. In the post, “Increase Your Mileage. Save Gas“, they suggest four ways to reduce your vehicle’s fuel consumption: (1) Slow down, (2) Skip the AC, (3) Inflate your tires properly, and (4) Accelerate Slowly. Being a car geek, these over-simplifications bug me, so I dove into the original article, “Tips to Increase Your Car’s Miles Per Gallon” to see if it did a better job of explaining “why”. It doesn’t, so I’ll give it a shot.

First a little background (which has to be a bit simplified or it would be a book. Sorry.) There are exceptions, but in general, modern fuel injection systems are load (typically measured by MAF or MAP) and RPM based, and the amount of fuel injected typically increases with both of those metrics. Fundamentally, it has nothing to do with vehicle speed, the AC being on, or your tires. The fuel injectors spray however much fuel is needed to hit a target AFR (air-fuel ratio) for a given engine load and RPM.

Knowing that fuel injection rates are tied to engine load (and a few more details that I’ll try to throw in), let’s go over this list and explain why:

(1) Don’t accelerate quickly.

Accelerating hard wastes uses a lot of fuel for a few reasons. First, you’ve increased engine load; Second, the car is going to richen the mixture to reduce the potential for detonation as heat and load builds; And finally (and this is where it’s going to get messy), you might have switched the fuel injection system from Closed Loop to Open Loop.

In closed loop fueling, the car’s computer (ECU) is watching the Oxygen sensor and tuning the fuel trims automagically to target a stoichiometric air-fuel ratio of approximately 14.7 parts air to fuel. This ratio is generally agreed to be the most efficient at burning all the available fuel, and it’s also the ratio for which catalytic converters are designed to operate most efficiently. However, the car doesn’t always try to be a non-polluter. In my Subaru, for example, the factory ECU switches from closed-loop fueling to open-loop fueling at approximately 60% throttle (TPS). If you press the gas pedal past this cut-over point, the car no longer targets a 14.7 AFR, but instead switches fueling for maximum power (which in a Turbo car is anywhere from 12.5 AFR to 10.5 AFR depending on the quality of fuel — and yes, the better the fuel, the less you need.)

(2) Let your car brake itself.

It’s arguable whether or not engine braking is a good idea mechanically, but it’s generally true that the car needs less fuel under this condition. The reason, once again, is because of reduced engine load (ie., the engine is no longer working to push the vehicle, thus it’s no longer sucking in a large volume of air, thus it no longer needs to spray a lot of fuel to keep that large volume of air at a particular AFR.) The engine may even use less fuel under engine braking then it does when coasting in neutral. That said, the brakes are designed to slow car, not the engine, and the brakes are cheaper to replace, so this isn’t always a good idea.

(3) Drive at the speed limit on highways and freeways.

Speed has nothing to do with fuel consumption. It’s load vs. RPM. The simplified reason your car uses less fuel at 55 MPH then it does at 60 MPH is because if you stay in the same gear, RPM’s increase with speed (as does aerodynamic drag, which increases load.)

Of course, this rule doesn’t always hold true. If you can draft another vehicle, you’ll reduce engine load. Furthermore, gear selection plays a role. 55 MPH in 4th gear will run at a higher RPM and should use more fuel then 60 MPH in 5th; But 55 MPH in 5th gear might not produce enough torque to climb a hill, thus increasing load when you try to maintain that speed without downshifting. You can also sometimes run the car leaner at high RPM’s after you’ve past peak torque output, so increases in RPM don’t always linearly add fuel.

(4) Use cruise control.

The benefit of cruise control is it’s ability to be smooth. If you can avoid sudden load, you’ll need less fuel.

(5) Don’t use the air conditioner.

First of all, don’t die of heat exhaustion just to save fuel. If it’s 120F outside, you should probably think of your immediate health first. That said, this tip is good at low speeds since running the AC compressor raises engine load. However, this tip doesn’t always hold true at highway speeds where rolling up the windows can reduce aerodynamic drag.

(6) Accelerate before hills.

Mmmm… inertia. Going uphill raises load (or climbing stairs wouldn’t be exercise.) Accelerating early may allow inertia to carry you up the hill instead of asking the engine to do all the work.

(7) Clean out your car.

The key point here is to lighten your car. It takes less power to move a lighter vehicle, and less power == less load == less fuel (or quicker acceleration on the same fuel.)

(8) Check your tire pressure.

Low pressure increases rolling resistance and increased resistance == load. Unless you’re driving in sand or rock-crawling where low pressure == traction (and you’re not going fast enough for it to matter.)

(9) Change your air filter.

If your air filter is clogged the engine will have to work harder to suck air through it. Working harder == increased load.

(10) Get a hybrid car.


(11) Do more in one trip.

Doesn’t save fuel per mile, but miles per day, which is also a good thing.

In summary, if you really want to burn less fuel… reverse-engineer your ECU, study the fuel maps for your car, and drive accordingly ;-)

Tailpipe Tally

TailpipeTally provides an online tool for comparing various vehicle fuel consumption and emissions ratings against the latest hybrid’s (Prius and Civic, for example.) Much to my surprise, my Subaru didn’t score all that bad. Yes, it pushes out more Carbon Dioxide (which while not a health threat, is a “greenhouse gas”), but it produces fewer harmful gasses then the hybrids:

emissions table

(via treehuger)

Broken downpipe bracket

It may be hard to see in this picture if you don’t know what you’re looking for, but this is the source of a random metallic clanking noise the impreza picked up awhile back.

The cause? A broken exhaust mount. This mount never really fit right (which seems to be the status-quo for after-market car parts.) It’s a solid metal mount that connects to the bottom of the downpipe. It didn’t line up once the pipe was installed, but with a little re-shaping it bolted in. The part lasted about 6-8 months before snapping.

The replacement, which I don’t have a picture of off-hand, was made by modifying a left-over after-market exhaust hanger from a 4Runner that didn’t fit the truck. The new hanger is still stiff enough to keep the turbo from moving, but has a small amount of flex to avoid this problem again.

I’m not sure what motivated me to post about this, but I have a long list of draft posts in my head that all relate to after-market product quality (or lack thereof) and the apparent in-ability for some of these performance shops to understand their market and launch new products. Many of these small manufacturers miss out on the little details, like: (1) a focus on quality, (2) providing proper customer support, (3) defining a clear product strategy, and (4) properly understanding who their customer is and what need they are trying to solve. They do manage to sell parts though — which I guess is what happens when you’re market is addicted to improving performance.

subaru update

In my post, Cold Weather Fuel Leak, I mentioned that my Subaru Impreza was hit with the infamous fuel leak problem that seems to be affecting many of these early models. I haven’t posted much of an update since then, but rest assured that no self-respecting car hacker would pay the dealer the $400 they wanted to fix this little problem. Instead, the engine bay was properly torn apart over the holidays and the entire fuel system, from the filter to the return line, was replaced with new rails, connections, and steel-braided lines. In doing so, the fuel flow has also been changed from a serial to parallel pattern, the injectors were upgraded with higher-flowing units off a Japanese Impreza, and a few extra engine sensors were wired-in for additional data-logging.

Everything came back together just fine (with a few extra parts left over for good measure), but I’m still re-tuning the fueling maps. The car runs properly while in closed-loop fueling (meaning the time where output from the oxygen sensor is used to determine fueling), but goes ridiculously rich once the boost comes on (and it moves to open loop fueling.) Since I haven’t had time to debug this, I dropped the boost limiter down around 2psi, and just drive it on light throttle.

I have a couple theories as to what’s happening, and hopefully this weekend I’ll have enough time to dive into it further. The first possibility is that the injector scaling capability of the UTEC might not be working properly. The second is that several other people with similar setups have reported that in creating a better flowing fuel system, the base fuel pressure has gone up, which is why the injectors are flowing more fuel then expected. The final theory is that the stock fuel pump is no longer able to keep up with the new high-flow system. If this were true, fuel pressure would actually drop as boost comes on, and with low pressure, the injectors would be unable to properly atomize the fuel, thus causing unburnt fuel to leave the exhaust ports and be seen by the O2 sensor as running rich. This one is particularly dangerous. If true, then there is no telling how lean the engine is actually running.

Fixing the first problem involves software. Proving the second requires a fuel pressure gauge and OBD scanner capable of accessing fuel trim values (neither of which I have.) Solving it probably means an adjustable fuel pressure regulator. The third one is messy. It will also require a fuel pressure gauge, but fixing it will require an upgraded fuel pump and an adjustable fuel pressure regulator. Fortunately a buddy of mine has all the diagnostic equipment we need.

With something like 10 hours of personal time doing the labor, and another 10 to retune the car, there’s no way I’m actually saved money (if you count time as money.) But where’s the fun in just taking a car to the shop? Besides, I can now tell you what every hose under the hood goes to ;-)

If you’re interested in taking a wrench to your own Subaru, I’d suggest starting with the following discussion board sites first:

WRX Hackers