Transmission Fluid Change, Part 1

My Freestyle's engine puts its power to the wheels via a CVT transmission.  CVT stands for "Continuously Variable Transmission" so the oft-said phrase " CVT transmission" is a bit like saying "RAM memory", or "you only YOLO once".

Thanks Mitsubishi motors

A CVT works by connecting two conical pulleys mounted on shafts, one being the power from the engine and there other being the output to the wheels, with a belt. The pulleys then change where the belt rides on the pulley to change the rotation speed of the input shaft compared to the output shaft. Or something like that.  Wikipedia contains an excellent description and history of this technology.

Air Bag Light Blinking: Christmas in July

Thanks
picgifs

Just as I was about to take this in for inspection: the air bag light started to blink and then stayed on steady.  Another light for my dashboard Christmas tree.  The car will pass PA inspection with the air bag light on (weird, but true!), but I think I'd like to be sure the air bags actually work.

When I saw the light blinking, it followed a pattern.  Thinking back to the days when you would count the short/long beeps the BIOS generated during boot-up to diagnose problems, I decided to turn the car on and count the flashes.  I counted four, then a pause, then three, then a longer pause, then the same pattern again before the light stayed on steady.

I'm going to guess the long pause was to mark separation between the message being repeated, this leaves me with the following possible meanings

More Obvious

These seem like easy ways to encode the information, we'll try this first

• 4 and 3
  I'm seeing two error codes, one being four and the other being three.
• 7
  Add the error codes together to get a single number.  Why not just flash that many times?  Maybe it's easier to count two smaller numbers.  Hey, I'm guessing here.
• 43
  Concatenate the blinks, that's the error code.

Obscure

This might be what's in the internal state of the machine for the "obvious" answers above, but I'm not sure if this is where I need start looking

• 2^4 + 2^3 (22) or 16 and 8 or 168
  The numbers are powers of two, get these and do the same bit of addition or concatenation.
• 00001111 00000111: 15 and 7 or 157
  Let's assume little endian for right now, as this yields a small number, I don't think we'd have a "big" number for to use for an error code.

Found an answer...

A little interweb scrounging found that it's the concatenation of the two integers (43) and my front impact sensor on the passenger's side was bad.  But how to confirm?  My standard OBD (On Board Diagnostics) tool didn't return any codes, so I would need something more specific to the vehicle.  Some searching on Google Play revealed an excellent app, FordSys, that will read the magic Ford registers over CANBus (the protocol used to deliver the diagnostic information over the OBD port) and that found the following codes:

• B2296
  Front impact sensor failure, does not say which sensor, but the blink code does.  The common "fix" for this is to remove the wire, check for ground faults, clean and re-attach the sensor.
• B1342
  The ECU has some crash data stored, this is likely the result of the sensor failure.  A replacement is north of $600 or you can have the one in your car cleared for $50 from various vendors. I'm searching for a tool that can do this over the existing CAN network, as this part is buried under the center console. 

Just the beginning

There's still lots to do here, it's unlikely that the sensor itself failed, but that's possible.  Maybe with my mucking under the hood, I've disturbed a wiring harness somewhere.  I can do some testing later to see if I can isolate the problem.  For now, I'll just live with the flashing and try to avoid any head-on or offset collisions.

Outer Tie Rod Ends (penultimate task before PA inspection)

Worn tire

On the way home with the Freestyle, we noticed an imprecise feeling when we were steering and some tire squealing when we were headed around corners (my second oldest son was piloting).  The tire wear indicated poor alignment and when I lifted the car to check the brakes, I found that the front wheel had about 2mm of play, all of which pointed to some worn tie rod ends, and that, in fact, was the problem.

Indifference Curve

Indifference curves
Thanks wikipedia!

Shopping about for this part was interesting, there's all sorts of questions. How do you really know what's a quality part?  Is the brand name really an indicator of quality?  What about the reviews or what others have purchased?  Does a higher price translate into higher quality?  Will one site stock a collection of better parts and another sell lower quality parts, even if the lower quality parts cost more?

Economists attempt to model this behavior with the "indifference curve" mapping out where, given the blue budget line, the trade-off between good X and Y that would leave the consumer "indifferent" between the choices. The problem is, my basket of goods is just one thing and I don't have a clear idea of the price trade-offs. Adding to the problem, I'm not spending enough money to research the decision adequately, so the decision making will be far from a well-researched consumer activity.

Shopping About

For a part like this, controlling the steering, integral to the safety of the car, I purchased the cheapest option -- I'll get to why in a bit.  Here's what I found while shopping (data collected 9-Jul-2016, prices/url might have changed since then):

• Amazon: $27.78/each
• RockAuto.com: $16.67/each
• CarParts.com: $11.25/each or $21.38/pair
• JC Whitney: $20.72/each or $33.02/pair

Replacement tie rod end

I found CarParts.com by searching (I use Bing more than Google, still waiting for the return of AltaVista and/or Lycos) for the part number found at JC Whitney.  The part was the same and about 2/3 the price, which brings us to the real problem: at which point would I be indifferent making a trade-off between these different choices? An economist would say consumers behave rationally, using all available information to arrive at a purchasing decision.  That's bunk for the following reasons:

1. People approximate rational
   Some more than others, emotion plays a big part in how we act, I would posit a bigger part for people to assert otherwise.  Plus, we're so bombarded with marketing/ads I doubt we could make an unclouded decision.
2. Information itself is imperfect
   Information tends to be an incomplete mess requiring some amount of assessment on the part of the person.  People will perform this assessment imperfectly, and according to #1, not in a rational way.
3. "All available" information eludes market participants
   Consumers supposedly use "all available" information to make buying choices.  Here's the rub: most of the time we have a sub-set of data to interpret because we get board and give-up during the information collection process.
4. Agency issues on behalf of information suppliers
   Getting information about your parts from a parts vendor?  Guess what outcome they'd like?  The information available has likely been skewed by the information provider who has ulterior motives. 

Those who schlepped through Econ 101, instead of taking some other course, frequently mis-apply efficient market theory to consumer purchasing decisions, while those who mistakenly signed-up for Econ 201 would use behavioral theory to explain consumer actions.  Actual economists readily admit, for most purchases, consumer preference comes from marketing and advertising.

Then, it dawned on me: there's no quality licensing process for car parts, not even a Underwriters Laboratory that does independent testing to certify parts meet some minimum quality level.  A quality constraint placed by a rating agency would put a floor on the price, because, in theory, a vendor could sell a paper mache part for $1.

Cheapest?  Why?

At a decade and 300K miles, let's be honest, I think the tie rod ends will outlive the car, even if they're not the highest quality. There's something else at play: does a high-priced part imply a better part?  As a consumer, I lack the information to determine what part is "better" independent of the price.  In the case of the same part being roughly half price at a different vendor, there's not a strict linkage between price and quality.

Let's get on with it: Removal of the old Part

Loosened the wheel lug nuts, lifted and secured the car with jackstands, removed the tire and got to work.

Size of bolt

First step: Loosen (just a little) the retaining bolt on the inner tie rod.  I didn't have a wrench the correct size, so I used an adjustable wrench to hold the tie rod.
Next: Detach the outer tie rod from the steering knuckle.  For this one needs both a deep 21mm socket, a 21mm box-end wrench and a 10mm socket.

This is where I found that I didn't have a 21mm deep socket or wrench, so I needed to get those before I made any additional progress.  Returned from the store with  both tools for under $20.  Seems like the $20 bill is the new $5 bill and the $100 bill is the new $20 bill in terms of purchasing power.  Times have changed.  Anyway...

Driver's Side: No Problem

The driver's side presented no problems.

First job, loosen the locking nut on the tie rod, move it back about 1/4 or even less of a turn.  Do this first, because when the tie rod is attached to the steering knuckle it's much easier to apply the necessary force to break the tension on this bolt, as the bolt will be held in place by some rust and road grime as well as torque.

Next, I broke the tension on the tie rod nut with my newly purchased deep 21mm socket and a breaker bar, this required about the same amount of force as a tight lug nut.  Once the tension was broke, the nut and bolt rotated freely, as the opposite end is a ball joint in the tie rod end, so I needed to apply force to both the nut and bolt to complete the removal.  With a 21mm wrench holding the nut, a 10mm did the job of twisting the bolt, which turned freely due to the ball-joint construction of the tie-rod end.  Note:  the nut needed to be turned clock-wise to loosen when holding the bolt and rotating the nut.

With the driver's side completed in less than 10 minutes, even taking pictures, this job looks to be quicker than an oil change.

Passenger's Side: Six points too many

Not the best idea, this didn't work all that well

The passenger's side was a little more of a challenge.  The breaker bar did the hard work but then the 12 point 10mm socket that worked fine on the driver's side stripped while removing the bolt. At this point, I should have noticed that these parts were different (obviously cheaper), but it didn't register yet. Doused the nut in PB Blaster and waited.  Used vise grips to get a better hold, but they were slipping as well.  Seeing that I had a bit more of the bolt not stripped, a deep socket 6 point 10mm socket worked and I was able to wind back the bolt.  The part where the 10mm top of the nut failed looked to be a replacement part versus the driver's side which appeared to be original.

Starting to like the color sockets

The passenger's side tie rod was in much worse shape than the drivers. This is when I noticed that this was also a different part as well.  Tie rod ends should be replaced in pairs.  If one has worn to need a replacement, part on the other side is probably ready replacement as well.  The other aspect is the performance of the car itself, a nicely working tie rod on one side and a dodgey one on the other will result in odd handling.

Looking the parts, it appears like the driver's side is OEM and the passenger's side a replacement part.  The picture shows how the part has a torn boot, what you can't see is that the bolt in the tie rod just flops about with no stability.  This would make the car's handling wayward and a correct alignment would be impossible because the tire wouldn't stay in a predictable position.

This might be a problem

One thing you can't see in the picture is the complete lack of resistance offered by the nut protruding from part, it just flopped about with little resistance.  When in the car, this fault would result in wobbly handling and vague steering -- although I can't say I noticed much during my pre-purchase test drive; however, the issues were readily noted on the way home.

Six vs Twelve points: Less is more?

OK, I shouldn't have been using the 12 point socket, because these generally result in more frequent stripping of the fastener, as what happened in this case.  I've found that it's easier to strip a fastener with a 12 point socket, but finding a good reason as to why that's so is easier said than done -- it's almost a vi versus emacs level "debate".  Common sense says the 12 point sockets apply force to a smaller area of the bolt, but that's not settled science at this point.

Installing Parts

Since I just moved the locking bolt back a little (less than 1/8 turn), I used that as a guide for the new part, as not to mis-align the vehicle too much.  Given the state of the alignment, I'm going to have this re-done anyway with the new front tires.  Side by side, the parts were the same length, so the position of the bolt would imply the position of the new tie-rod close enough.

I put a little anti-seize on the outer tie rod end so alignment would be easier, the tension from the locking bolt will keep this part from moving. The ends then need to be filled with some grease and I have a "small" grease gun that attaches to the grease fittings to do the job.  I put about 5 or 6 pumps of grease, this resulted in the excess grease leaking out, so I knew the fitting was filled.  This is where I encountered my first non-metric bolt, the grease fitting needed a 5/16th socket to wind into place.

Done, note the grease dripping from the 

Used a breaker bar to tighten the bolt, tire back on and all done!   The spec calls for 85 ft/lbs of torque on this part, I think I'm close enough if not over.  This was a quick job, no more than 30 minutes a side, even with the problems removing passenger's side part, taking my time to make sure I did it correctly and taking some pictures.

Parts Comparison

I've included a picture of the three tie rod ends in this job, the closest being the tie rod I removed from the passenger's side with the torn boot, the farthest being the newly purchased replacement part and middle being what looks like the original.  I think with some grease, the original part would still be serviceable.  A little measuring revealed the dimensions of the OEM part were about the same as the replacements, with one key difference: the diameter of the end where the ball join sits is about 2mm bigger on the original part.  The failed replacement also didn't have a grease fitting, so one assumes it was (or maybe wasn't) lubed at the factory and the absence of a grease fitting and no hole for the fitting were "design for cost" feature deletions. 

I'm not of the quality of the part I purchased, but I would be very surprised if it outlasted the vehicle. Maybe I'll be surprised.

Alignment

As can be seen by the tire wear, an alignment is necessary as well as two front tires.  For this, I'll visit my local tire shop. A quick survey of local-area shop shows the alignment will cost about $75 and I'm trying to economize on the tires as well.

Cleaning Throttle Body

While taking things apart to fix that torque bushing, I removed the throttle body,

Close-up of dirty throttle body

held on by four 9mm bolts to get access to the bolts holding the torque arm in place.  It looked dirty. In the extreme close-up, notice the tar deposits on the top and bottom of the flap. Figuring things work better when clean, although I really don't have proof of such a hypothesis, I decided to remove the gunk before re-installation.  Cleaner specifically for this task exists, so that's what I used.

The Cleaning Job

Clean!

Completely removing the throttle body from the car made cleaning both sides a snap.  I'm not sure what comprises throttle body cleaner, but it melted away the plaque deposits on contact and probably reduced my lifespan by three to six weeks.  Wiped the run-off away and that was it.

If the part wasn't off the car, cleaning would have been much more difficult, as I would have needed to hold the spring-loaded flap open during the process so I could clean both sides.  The part went back on the car when reassembling the rest of the air intake system.  The four bolts were tightened to "hand tight".

Clean too!

Can't really be sure if this made or will make a difference in performance or fuel economy. Maybe the throttle will turn a little easier as I've removed some of the gummy material from the pivot points, so that actuator won't have to work too hard to reposition the throttle.  Since a spring pushes the throttle closed, that won't have to work as hard when closing the throttle.

Rest of intake dirty...

What was still very dirty was the intake itself.  I tried to clean this a little with just a rag, but I think, like the throttle body itself, removal first would yield the best results. Just inside the intake was a sensor, I didn't want to contaminate that with cleaner that could cause damage.

 

 Two observations worthy of note

1. This is a "throttle by wire" car
   A set of electrical wires controls the throttle, there's a servo in the throttle housing to create the force to overcome the spring and open the flap to increase air flow. No mechanical connection between the driver and the device feeding air to the engine. I would bet the engine management software decides how much to open the throttle, the position of the accelerator pedal being just one of the factors in the decision-making algorithm.
   
2. The spring-action of the throttle will result in closure by default
   This is design for graceful failure, I would guess.  The mechanical spring will force the throttle closed by default, so if the actuator in the throttle fails, the lack of air will result in the engine stalling, which seems to be the safest of alternatives.  A neat, fail-safe, design.

I'm not sure how to assess a cost, as I already had the cleaning chemicals on hand and I'm not sure how to account for few paper towels.  For tracking purposes, let's say the cost approximated zero.  RepairPal says this would cost about $200, labor being the biggest factor in the price. This was an easy job, something to consider tackling on your own and saving those two Benjamins for something else.

The Forgotten Oil Change: Brake Fluid

Thanks makeameme.org

When replacing the brakes, many (let say most?) home mechanics overlook doing the same for the brake fluid.  Brake fluid works hard and needs to be replaced with fresh fluid in order to get the best stopping performance.  Over time, brake fluid absorbs water from the atmosphere and becomes less effective -- the contaminated fluid can be compressed more easily this diminishing the efficacy of the brakes.  The extra moisture in the system will also cause problems for the braking lines, causing them to rust from the inside out.  All not good!  I inspected the brake lines on my Freestyle, but if the deterioration is happening on the inside, there's not much to see from the outside until the leaking starts.

Brake fluid costs about $8 for a large bottle, so why not replace that well? I'll tell you why: bleeding brakes is a bit of a pain and messy and there's the fear that it will be done completely wrong causing massive brake failure at an unpredictable time resulting in the car careening off the road, hitting an embankment and erupting into a giant fireball.  Or maybe it's just messy?

Out with the old

Looks like it was never changed

Several specialized tools exist for extracting the used brake fluid from the master cylinder, the one shown in the picture happens to be a turkey baster from the dollar store.  Not the greatest tool, but cheap and disposable. The fluid is... groady.  New brake fluid looks as translucent as light veggie oil.  What I see here looks like over-worked oil I've drained from vats while working as a teen-ager in fast food. Ewwww.  The picture just doesn't convey the abhorrent condition of the liquid, it's both brown and green with what looks like algae floating about -- give it another three months and we might have a new life form on our hands.

In with the new

HF suction brake bleeder

The replacement procedure is easy.  You'll need a way to catch the old fluid, I happen to have a HF special vacuum pump operating by my small (but powerful!) compressor.  Even though the vacuum draws the fluid, I still speed the process by having somebody push the brakes while using suction.

Start from wheel the furthest from the brake fluid reservoir, remove the dust jacket from the bleeder, give the bleeder a turn with a socket or wrench (11mm in my case) and some of the liquid will just start dripping out, attach the bleeder hose and have a buddy pump the brakes. When the fluid's clean, tighten the bleeder; just to make sure that I don't let air seep back in, tighten the bleeder while the vacuum is running; you can get the same effect by tightening the bleeder when your buddy is pushing in on the brake peddle.

Catching brake fluid

For this car, when the rear caliper was detached from the car when doing the brakes, I took the liberty of letting the fluid drain from the brake hose into a spare oil bottle. I still had to do the bleeding process, as the re-attached caliper contained both air and some old fluid, but since most of the old fluid drained out of the lines, the new fluid was visible in just a push or two of the brakes.

All that's left to do replace the dust cover, top-up the brake fluid and move to the next wheel in the rear.  Repeat again on the front wheel farthest from the master cylinder and then the remaining wheel.  Removing the wheels and lifting the car is the hardest part of the job.  I have the habit of lifting just one wheel at a time when I can, because that is the safest approach, keeping as many wheels in contact with the ground as possible.

Broken Bleeder

Decapitated bleeder

Bleeder screw that broke

Consider using some penetrating oil on the bleeders so they turn without incident.  Sometimes, the bleeders can break off, as they're essentially a hollow bolt. This happened on the front driver's caliper.  Having some other chores to do, I left well enough alone, when I get the chance, I'll pull out this bleeder and replace it, but that's another project and post.

Disposal 

Brake fluid is not motor oil, so don't secretly dump it in the used oil vat of your local reclamation center.  Please bring it to your area's hazardous waste collection center for proper disposal.  We all live here, so do the right thing.