Random suspension wobble in '01 with 229k miles

Hello all!

My '01, with about 229,000 miles, has developed a bit of a wobble in the steering or suspension that I can't quite track down. It started happening last summer (2019) but has gotten more noticeable over time. I thought I'd post and see if this sounds familiar to anyone.

I've heard of the steering shaking that some Gen1s develop, and I think the diagnostic for that is to unplug the steering computer and drive it; if the shaking stops, that's what it is. I don't think I have that problem, but I'd be happy to be corrected on that notion.

The symptom is that when I'm driving at highway speeds - 55 mph (90 km/h) or more - the car feels like one end or the other is jumping to the left or right by a small amount. One way I've described it is that a daemon is running alongside the car and occasionally putting his hand on the fender, near the tire, and pushing the car sideways for just a second. If you've ever driven on a deeply-grooved concrete freeway, such that the road fights you a little when you try to change lanes - it feels kind of like that, but in discrete bursts, not continuous.
I don't feel like I'm going to skid or lose control, but it feels weird.

As far as I can tell, the steering wheel is not moving when this happens; I don't feel the motion through the wheel. I do feel it in the seat of my pants, if that makes sense. The disturbances are small enough and random enough that I'm not sure I could correct them with the steering wheel, even if I wanted to. The car still seems to track straight in the lane, and when I change lanes or go around a big sweeping curve, I don't feel like it's oversteering or understeering any differently than it used to. At first I felt like it was just the back end of the car doing it, but now I'm less sure of that.

As far as I can tell, it either doesn't happen or I can't feel it below about 40-45 mph (65-70 km/h). I only notice it when I get up to highway speeds.

I'm pretty sure it was happening before I got new tires; I got two (General Altimax RT43) in mid-September 2019 and two more at the end of December 2019. However, it seems to be more noticeable since I got the new tires. I'm not sure whether it's naturally wearing out more over time, or if the new tires are making it more obvious. The previous Michelin Harmony tires I had on the car had about 72,000 miles on them. One of them had the cords shift in the tire (which felt really weird when driving; I found it when I pulled off the highway because it felt so bad) and one developed an unpatchable sidewall leak. All four Michelins still had decent tread depth, so I'm not sure that the new tread on the Generals would cause that much of a difference.

This weekend, I jacked up the front end and did a little inspecting. The passenger side ball joint does have a tear in the rubber boot, and a little turd of grease had leaked out. The driver's side looks OK. I pried between the lower control arm and the steering knuckle to check for ball joint play and I don't think I found any, but I've also never done that test on both a new and a busted ball joint to know what it should feel like. I know I will probably end up replacing at least the passenger ball joint, due to the torn boot, but it doesn't feel like the "smoking gun" to me.

I tried to inspect the two rubber bushings where the control arms bolt to the subframe, but both of those are pretty well hidden by the subframe. There is still some rubber in there; I grabbed the control arms and tried to wiggle them and they didn't rattle around on the bushings.

The front struts don't appear to be leaking oil. I reached in over the tire, grabbed the spring coils, and tried to push and pull the entire strut, both inboard/outboard and front/back, and I didn't get any movement or funny noises.

I also looked at the toe control links that hold the rear subframe to the body. I can't see the ends that plug into the subframe, but the ends that bolt to the body don't have any obvious missing chunks. There are some dry-rot or stress cracks in the rubber on those ends.

I didn't jack up the back end yet, so I'm not sure about the rear struts.

The front suspension probably needs work for other reasons. Sometimes when I turn the steering wheel, even with the car at 0 mph in the parking lot, I get a rumbling noise from the front end. I am interpreting that as the bearings at the top of the struts going out, because it sounds like it's coming from the top of the strut towers, rather than down lower where the steering rack is. I don't get that noise all the time, though. Also, sometimes, if I stop hard, accelerate hard, or otherwise change the front suspension load quickly, I get a little "click" or "tock" noise up there, which I think might be one of the control arm bushings.

My next action is probably to jack up the back end and pull and tug on things back there, to see if I can find anything loose on that end. I have a jack, jack stands, and hand tools. I also have Techstream, if I lug a PC out to the car.

Thanks!

 

You might not believe this, but I didn't do that on purpose. On the other hand, perhaps I was inspired by the old Car Talk credit of Director of Cadillac Steering, Toulouse Toutrack.

Checked all four wheels after both sets of tires, but I will check again. After once having to use the jack as an extension bar on the lug wrench on the side of the road after a tire store idiot had been at my car, and breaking a SNAP-ON socket in my driveway after a tire store idiot had been at my car, I always loosen and re-torque the wheel nuts myself after the car has been in the shop. But it's simple to check.

It doesn't seem to take any more throttle than usual, and I haven't noticed the wheels being hot, but I also haven't put my hand near all four wheels after I drove. I'll try driving it for a bit and then informally checking the wheel hubs.

No, just the tire install. I'd have to dig in some papers to find the last time I had an alignment done, but it's probably at least a couple of years and 10k miles ago.

Thanks!

 

tl;dr: I can reproduce the "clunk" by grabbing the driver's outboard end of the front sway bar and pushing and pulling on it. Toyota says the front sway bar can affect the wheel alignment, and after thinking about the geometry of the situation, I agree. I'm not sure yet if I need a sway bar bushing, the clamp that holds the bushing in, a sway bar, or the link bar with the two tiny ball joints on it, but I think I'm on the right track.

I checked all sixteen wheel nuts today and they were all tight. I didn't have time to drive it enough to check the front brakes, but I did spin both rear wheels by hand when I had that end up in the air. The right one spun freely with no noises; the left one spun maybe a touch less freely, and made that brake-shoe-scraping noise at a few points in the wheel rotation. I took the left rear wheel and drum off and played with the adjuster a little bit until I got the shoes as close to the drum as I could get them without any scraping noises.

With the rear in the air, I also pushed and pulled on various parts of the rear suspension (wheels, struts, the beam itself) and couldn't get any odd noises. I got a better look at what rubber bushings I could back there, and none of them looked obviously scary.

I've had that feeling before when there was a big gust of wind... but now it's doing it even when the wind is calm, or blowing maybe 5 mph.

This turned out to be a good suggestion, thanks! When I looked at it the other day, I only really inspected the tiny ball joints at the top end of the "link" that connects the sway bar to the strut; the wheels were still on and the lower end was hard to see. I think I probably grabbed the sway bar ends and tried to move them, with the wheels off the ground, but I didn't get any movement.

Today, I looked at the lower tiny ball joints (boots not obviously torn), but I also grabbed the ends of the sway bar and tried to make it move. With the wheels on the ground, the passenger side didn't get me much action, but the driver's side gives me what I think is the same "click" or "tock" noise I get sometimes when driving it.

I have an inspection camera, and I couldn't see a lot of the bushings, but I could detect a difference between the left and the right. There's a horizontal seam in the rubber, on both the front (FRONT) and rear side of the bushing. The front of the driver's side bushing has a bigger gap in this seam than the front of the passenger side bushing.

I don't know if the factory bushings were one solid piece, like a donut, and slid on from the ends of the sway bar, or if they were slit through on one side, like the letter C, so they could be pushed over the bar in close to the right spot. The parts fiche show them as one solid piece, but I'm not sure if I believe that yet or not.

I also could have a crack in the metal hoop clamp that holds the bushing down to the subframe; I didn't have time to get far enough into it to see if that's happening. I guess it's possible that I have a broken sway bar; I looked at the piece where the sway bar dips under the exhaust pipe, right in the middle of the car left-to-right, and I think I saw it trying to move when I was pulling on the driver's end of the bar, but I'm not 100% sure. In the coming days, I should be able to get someone else to pull on the end while I put my fingers and eyeballs closer to the middle, to see what's going on.

I'm guessing that having the front end in the air - front suspension at full droop - loads the sway bar in such a way that whatever is loose wasn't obvious. I only noticed it when checking with the wheels on the ground. (The factory manual procedure for the front ball joints says to check them not at full droop, but with the front wheels sitting on an ~8" (~200 mm) high block; I wonder if that's for a similar reason.)

The factory manual procedure says to check the alignment after you R&R the front sway bar. I had to think through it, but here's what I have: the strut has to be free to rotate in yaw, so you can go around corners. The sway bar link attaches to a tab on the trailing edge of the strut. If the sway bar is providing fore-and-aft input *exactly* on the centerline of the strut, then in theory, nothing much should happen. However, if the force is not exactly on the strut centerline, it will want to turn the strut a little... which might explain what I'm feeling.

I'm pretty sure that I need to replace something on the sway bar. Once I figure out what, I'll do it, and then go get an alignment.

Thanks for the suggestions!

 

tl;dr: It's at least the links from the sway bar to the strut, on both sides. The lower tiny ball joint on both sides is very stiff. I drove the car at speed with one of the links removed and the wobble settled down a lot. I plan to replace both links, and if that feels better, then get an alignment to dial it in completely.

For the whole saga, read on.

Working on it

I worked on it some more today. I took the right front wheel off, and this time, I was able to reproduce the noise with the suspension at full droop. With the wheel off, I could get a better grip/angle on the sway bar; I could grab the part that runs fore-and-aft, forward of where it comes out of the bushing on the subframe, tug it around, and get the noise to happen.

I could see the driver's side bushing a little better and I'm pretty sure the metal clamp isn't broken. The bushing doesn't obviously look bad after closer inspection.

I decided to take the link off and check it. Toyota is right; the ball joint stud will want to spin with the 17 mm nut. They say you can use a 5 mm Allen wrench on the stud, but mine took either a 5.5 or a 6 mm, depending on which stud I was working on. (But, at least one link and possibly both have been replaced with aftermarket parts.) I busted the nut loose with a 17 mm socket and breaker bar, and then stuck the Allen wrench in the stud and used either an open-end or box-end 17 mm wrench. On the driver's side, I could hold the Allen wrench in my hand, but on the passenger side top, I had to slip the hole in the handle end of my breaker bar over the Allen wrench to get enough leverage. You could probably also do this job with a 1/4" hex drive bit with a millimeter Allen on the other end, but if you do that, try to use your good hex bit, not the one from the Hazard Fraught "20 bits for $5, LOL" set.

Once the link was off the car, I grabbed the free end of the sway bar and pushed and pulled on it again, and I couldn't get the noise to happen on the driver's side again.

With the link in my hand, there was a definite difference between the upper and lower tiny ball joints. On the upper one, I could hold the steel rod of the link, right next to the ball joint, with one hand, and move the ball joint stud around with two fingers on my other hand. On the bottom one, I couldn't do that - I had to either thread the nut back on and use an open-end wrench, along the axis of the stud, to move the stud, or I had to hold the stud in my fingers and grab the link further towards the opposite end, to have a bigger lever.

The factory manual gives a spec that involves flipping the stud back and forth 5 times, putting the nut back on, twirling the stud with a torque wrench, and checking the torque reading during the 5th twirl. (Page SA-42 in the '01 manual.) Spec is 0.4 to 8.7 pound-inches, or 0.05 to 1.0 Newton-meters. To really do it how they want, you either need a beam-type torque wrench (which they show in the diagram), or a digital-reading type. I have a click-type wrench that goes down to 25 pound-inches, but no lower. I tried the factory procedure with the wrench that I had, set to 25 pound-inches. The lower tiny ball joint would always make the wrench click when I tried to turn it, and the upper would not, so I am interpreting that as a failed test. I also tried increasing the setting on the wrench, and the lower tiny ball joint would stop making the wrench click somewhere around 30 to 40 pound-inches; it wasn't super repeatable, but it was in that range.

After all that, I decided to check the passenger side as well, and found the same thing - upper tiny ball joint easy to move, lower tiny ball joint very stiff. With both links off, I could also rotate the entire sway bar several degrees in pitch, and it felt OK. I verified that both ends of the bar move at the same time, so I don't think the bar is broken. I also tugged and pushed on it, and I couldn't get the bushings to make any noise.

I don't think you can drive the car with both links removed, unless you also remove the sway bar (which the factory manual says involves dropping the whole front suspension beam). The reason is that when the front/leading ends of the bar are pitched all the way down, either the extreme outboard ends will hit the control arms, or the bar will hit the boots on the end of the steering rack. With the ends pitched all the way up, the ends contact the brake hoses. You might be able to use some tie wire or zip ties to hold it out of the way, but I'm not exactly sure what non-moving part of the car would be good to tie off to.

I reinstalled the passenger link, which was a little entertaining. I could get the 17 mm nut most of the way down with a ratchet, but then it started wanting to spin the stud again. The torque spec is 55 pound-feet or 74 Newton-meters, but using a regular socket on the torque wrench precludes using an Allen wrench to hold the stud. Luckily, I just happened to have a set of metric crow-foot wrenches, which I last remember using sometime before 2008. I used a crow-foot on the torque wrench, the Allen key and my breaker bar to hold the stud, and everything was fine.

(The radius of my naked torque wrench is approximately 17.5 inches, and the crow-foot added about 0.5 inch, which means I was putting (18/17.5)=2.86% more torque on the nut than what the wrench was set to. So to get exactly 55 pound-feet at the nut, I would need to set the wrench to 53.4 pound-feet. I set it to 54 pound-feet and didn't worry about it. I *did* take care to have the crow-foot in line with the handle of the wrench, although I'm not 100% sure that matters.)

Testing

With the passenger link on and the driver's link off, I drove the car at 65-70 mph (105-115 kmh). The wobble was greatly reduced. It was still there a little bit, but I attribute that to the passenger side link still being connected. Also, while I was driving on the city streets to get to the highway, I didn't get the noises I have been getting from the front suspension.

At speed, the front suspension did feel different overall , but I expected that with the sway bar disconnected. The easiest way to feel the difference was when I changed lanes across some uneven-height pavement; the roll of the front suspension didn't settle down as fast as normal. That's almost exactly what a sway bar is meant to help, so I didn't think that was unusual.

Shopping for parts

Olathe Toyota sells the links for $94 each; same part left and right. If you want new nuts, those are $3 each.

O'Reilly has various options, from their house brand at $30 (on the shelf in the store), up to an "Import Direct" brand for $74 special order. All of them appear to come with new nuts, and most of them appear to have nylocs. Interestingly, they have a "Precision Chassis" brand, special order at $36, with a grease zerk on both of the ball joints; the factory one and the other aftermarket ones don't appear to have those. Also, the cross-reference at O'Reilly says the same part fits certain years of of Pontiac Vibe, Scion tC, Corolla, Matrix, and (I think) most of the Gen2 Prii, as well as the Gen1.

Autozone has their house brand at $30 (on the shelf in the store), up to a "Whiteline" brand for $124 special order. They all seem to come with nuts.

Autozone also have some in the "Nolathane" brand, one of which (REV010010, $56) looks like the right part, and two of which (REV010006, $30, and REV010008, $28) appear to be wrong for a stock Gen1. Those two are very much shorter than the factory part, like maybe 3 or 4 inches (75-100mm) between ball joint stud centers, as opposed to about 11.25" (286 mm) on the factory part. Also, both ball joint studs are on the same side, while the stock part has them on opposite sides. I don't really know why they list those parts for this application, unless there are people that lower Prii. The cross-reference for the wrong-for-Prius ones shows a Hyundai Tiburon, Mitsubishi Diamante and Galant, and Toyota Camry, Corolla, Rav4, and Gen1 and Gen2 Prius.

Advance Auto has an "International Suspension Group" brand for $20 special order, their house brand for $30 (on the shelf in the store), up to a Moog special order for $64. They all seem to come with nuts.

These prices are all for the midwest USA; they may be higher in other areas of the country. I don't work for any of the companies mentioned.

Conclusion

I'm going to decide on a part and buy them tomorrow, and put them on the car in the coming days. Thanks to @Josey for pointing me in the direction of the sway bar links!