A bit more than a week ago I volunteered for an aid station along the route of the Wasatch 100 mile endurance run - which, as the name implies, is a 100 mile race, starting and ending some distance apart in Northern Utah. This year, I was asked to be near-ish the start of the race, about 20.9 miles (30.4 km) from the start at a location in the mountains, above the Salt Lake Valley - a place that required the use of a high-clearance and somewhat rugged vehicle - such as my 2017 Jeep Rubicon.
 |
| Figure 1: The blinking "Sway Bar" light - not something that you want to see when you have shifted out of four-wheel drive! Click on the image for a larger version. |
Loaded with several hundred pounds of "stuff" I went up there, bouncing over the rough roads and despite enduring several bouts of rain, hail, lightning and thunder, managed to do what needed to be done in support of the race and runners and headed down.
Because of the rather rough road, I decided to push the button marked "Sway Bar" that disconnects the front left and right front tires from each other, allowing more independent vertical travel of each wheel, making the ride smoother and somewhat improving handing over the rougher parts. Everything went fine until - on the return trip, near the bottom of the unimproved portion of the mountain road, I pushed the button again and... the light kept blinking, on for a second and off for a second - and a couple minutes later, it started blinking twice as fast, letting me know that it wasn't "happy".
"What's the problem with that?"
Pretty much all modern road vehicles have a sway bar - or something analogous to it - that couple the vertical travel of the wheels on the same axle together to reduce body roll, which improves handling as one makes a turn - particularly around corners. At low speeds, such roll isn't too consequential, but at high speeds excess roll can result in... well... "problems" - which is why I was a bit apprehensive as I re-entered the city streets.
Knowing that this type of vehicle is known for "issues" with the sway bar disconnect, I did the normal things: Pushed the button on and off while rocking the vehicle back and forth (while parked, of course!), stopped and restarted the engine - and even pulled the fuse for the sway bar and put it back in - all things suggested online, but nothing seemed to work.
Stopping at a parking lot and crawling under the front of the vehicle while someone else rocked it back and forth did verify one thing: Despite the indicator on the dashboard telling me that the sway bar wasn't fully engaged, I could see that it was, in fact, locked together as it should be as evidenced by the fact that the two halves of the bar seemed to move together with the vehicle's motion - so at least I wasn't going to have to drive gingerly back on the freeway.
Fixing the problem:
 |
| Figure 2: Sway bar and disconnect mechanism, removed from the vehicle with the lead screw/motor in the upper-right. Click on the image for a larger version. |
The reasons for the issue seem to be various: Being an electromechanical part that is outside the vehicle, it's subject to the harsh environment of the road. Particularly in the case of some die-hard Jeepers (of which I'm not particularly, although I've made very good use of its rough and off-road capabilities) reports online indicate that it is particularly prone to degradation/contamination if one frequently fords rivers and spends lots of time in the mud: Moisture and dirt can ingress the mechanism and cause all sorts of things to go wrong.
Fortunately, one can also find online a few web pages and videos about this mechanism, so it wasn't with too much trepidation that, a week after the event - when I was going to change the oil, filters and rotate the tires anyway - I put the front of the vehicle on jack stands and removed the sway bar assembly entirely. This task wasn't too hard, as it consisted of:
- Remove the air dam. My vehicle had easily removable plastic pins that partially popped apart with the persuasion of two screwdrivers - and there are only eight of these pins.
- Disconnect the wire. There's a catch that when pressed, allows a latch to swing over the connector, at which point one can rock it loose: I disconnected the wire loom from the bracket on the sway bar disconnect body and draped it over the steering bar.
- Disconnect the sway bar at each of the wheels. This was easy - just a bolt on either side.
- Undo the two clamps that hold the sway bar to the frame. No problem here - just two bolts on each side.
- Maneuver the sway bar assembly out from under the vehicle. The entire sway bar assembly weighs probably about 45 pounds (22kg) so it's somewhat awkward, but it isn't too bad to handle.
 |
| Figure 3: Inside the portion where the lead screw motor goes: Very clean - no contamination! Click on the image for a larger version. |
With the sway bar removed from the vehicle, I first removed the end with the motor and connector and was pleased to find that it was perfectly clean - no sign at all of moisture or dirt. Next, I removed the other half of the housing, containing the gears and found that this, too, was free of obvious signs of moisture or dirt: The only thing that I noticed was that the original, yellow grease was black in the immediate vicinity of the gears and the outside ring - but this was likely to due to the very slight wear of the metal pieces themselves.
The way that this mechanism works is that the motor drives a spring-loaded lead screw, pushing an "outside" gear (e.g. one with teeth on the inside) by way of a fork, away from two identical gears on the ends each of the sway bar shafts which decouples them - and when this happens, they can move separately from each other. The use of a strong spring prevents stalling of the motor, but it requires that there be a bit of vehicle motion to allow the outside gear, under compression of the spring, to slip off to decouple the two shafts as they try to move relative to each other.
 |
| Figure 4: The fork with the outside gear-cam thingie. To disengage the sway bar, the outer gear is pushed out further than shown, disconnecting it from the end of the sway bar seen in the picture above and allowing the two halves of the rod to move independently. Click on the image for a larger version. |
In order to detect when the sway bar shafts are coupled properly, there's a rod that touches the fork that moves the outer gear and this goes to a switch to detect the position of the fork - and in this way, it can determine if the sway bar is coupled or uncoupled. With everything disassembled, I plugged the motor unit back in and pushed the sway bar button and the lead screw dutifully moved back and forth - and pushing on the bar used to sense the position of the fork seemed to satisfy the computer and when pushed in, it happily showed that the sway bar was properly engaged.
What was wrong?
I was fortunate in that there seemed to be nothing obviously wrong mechanically or electrically (e.g. no corrosion or dirt) - so why was I having problems?
I manually moved the fork back and forth, noticing that it seemed to "stick" occasionally. Removing the fork and moving just the outer gear by itself, I could feel this sticking, indicating that it wasn't the fork that was hanging up. Using a magnifier, I looked at the teeth of the gears and noticed some small blobs in the grease - but poking them with a small screwdriver caused them to yield.
 |
| Figure 5: Embedded in the grease are blobs of pink rubber from the seal, seen in the background. Click on the image for a larger version. |
Digging a few of these out, I rubbed them with a paper towel and discovered that they were of the same pink rubber that comprised the seals: Apparently, when the unit was manufactured, either the seal was pushed in too far, or there was a bit of extra "flash" on the molded portion of the seals - and as things moved back and forth, quite a few of these small pieces of rubber were liberated, finding their way into the works, jamming the mechanism.
Using paper towels, small screwdrivers and cotton swabs, I carefully cleaned all of the gears (the two sets on the sway bar ends and the "outside" ring gear) of the rubber. A bit of inspection seemed to indicate that wherever these rubber bits had been coming from had already worn away and more were not likely to follow any time soon.
 |
| Figure 6: More pink blobs - this time on the gear on the other sway bar. Hopefully whatever "flash" from the seal had produced them has since worn down and no more will be produced! Click on the image for a larger version. |
Putting an appropriate of synthetic grease to replace that removed, I reassembled the unit and put it back on the car, pushed the button. Upon reassembly, I applied a light layer of grease on all of the moving surfaces involved with the shifting fork - some of which may have been sparsely lubricated upon installation. I also put a few drops of light, synthetic (PTFE) oil on the leadscrew and the shaft that operated the sensing switch as both seemed to be totally devoid of any lubrication.
Although there was no sign of corrosion, I applied an appropriate amount of silicone dielectric grease to the electrical connector and its seal - just to be safe.
Did it work?
With the engine off, but in "4-Low", I could hear the lead screw motor move back and forth, and upon rocking the car gently I could hear the fork snap back and forth as it sought its proper position. Meanwhile, on the dashboard, the "Sway Bar" light properly indicated the state of the mechanism: Problem solved!
All of this took about two hours to complete, but now that I know my way around it, I could probably do it in about half the time.
Random comments:
I'd never really tried it before, but I was unsure if the motor would operate if the engine was not running: It does - pressing the "Sway Bar" button alternately winds the lead screw in and out - but it's not really obvious as to its position if the cam doesn't lock into place and the light turns on solid or goes out. Of course, this thing doesn't operate unless one has shifted to four wheel drive, low range.
This page stolen from ka7oei.blogspot.com.
[End]