What's Wrong With Dan's Mazda 3?

[dbeng]

Well, you may remember I have been complaining about lack of traction in the past few events. Specifically massive amounts of inside wheelspin while trying to dig out of corners. I first mentioned it back in the May 3rd event at Brooksville and have been fighting it ever since.

When it first happened, I was convinced something was broken but a visual inspection of all the suspension components didn’t show anything unusual. It all looked good.

Today I put some time aside to stiffen the rear sway bar – it’s adjustable with three settings and I’ve been running in the middle setting since I installed it, so decided to try the firmest setting to try and improve the corning grip. On the Mazda 3 it’s quite a pain to adjust as the end links are so short it’s impossible to pivot them enough to disengage the threads in position. The end links have to be removed which is another bugger of a job requiring a crows-foot wrench to be installed completely blind up inside the control arm to stop the upper ball joint of the end link from rotating while the top nut is removed.
I finally got the two end link nuts loose, grabbed one of the sway bar arms to pivot the bar down and this happened:

IMG_1766-B.jpg (184.62 KiB) Viewed 10952 times

It’s not very clear in this picture, but the arm I grabbed (far side in the picture) rotated down and the rest of the swaybar stayed where it was. The welds holding the arm to the bar has sheared clean through. Since the arm has a hole bored through it that slips over the cross-tube everything still looked normal, but the arm was just pivoting on the tube and transmitting exactly zero torque to the opposite side. These pictures show the problem:

IMG_1771-B.jpg (163.93 KiB) Viewed 10952 times

So – there’s the cause of my traction problems. I had effectively no rear swaybar. Luckily I had kept the stock Mazda swaybar so I bolted it back on until I figure out a solution for my upgraded bar. Some swaybar has to be better than nothing.
Now I don’t feel so bad about my results from the last couple of events….

Oh yes - and the orange gloves are 100% due to watching Edd on Wheeler Dealers

[Carracer]

Wow, that explains a lot, when I was watching you run that front tire would unload every single turn. Way worse than I have ever seen on your car. My question is who makes a sway bar that can't handle autocross?

[Loren]

That's not too different from the construction of my swaybar. I'm gonna have to keep an eye on it!

Looks like you didn't have a weld failure, it was the actual tube that failed.

Wouldn't be hard to clean up the surfaces with a grinder and have Joe weld it back together for you. But, odds are that it would just break again.

Given that it's such a simple design... I'd consider replacing the center tube. Figure out what your OD is, and your wall thickness. Replace it with something with more wall thickness. (you weren't using that one on full soft, anyway)

Take a good look at your end links and any brackets that are in play. Those things can bend, too. I bent the rear swaybar endlink brackets on my Saturn many years ago. Some Hondas are prone to ripping their end link mounts out of the subframe (or whatever it is that they attach to). A beefy rear bar on a FWD car gets stressed A LOT, and your car is not light.

[twistedwankel]

Holy Shit. M is for masochists.

[dbeng]

Wow, that explains a lot, when I was watching you run that front tire would unload every single turn. Way worse than I have ever seen on your car. My question is who makes a sway bar that can't handle autocross?

Yes, it sure does explain the smoky burnouts through the first left-hander at Dunnellon
The swaybar is from a reputable company, I won't name names until I give them a chance to have their say - it did come with a lifetime warranty so let's see, I bought it just under 3 years ago. I just contacted them by email so I'm hoping for the best.

[dbeng]

That's not too different from the construction of my swaybar. I'm gonna have to keep an eye on it!

Looks like you didn't have a weld failure, it was the actual tube that failed.

Wouldn't be hard to clean up the surfaces with a grinder and have Joe weld it back together for you. But, odds are that it would just break again.

Given that it's such a simple design... I'd consider replacing the center tube. Figure out what your OD is, and your wall thickness. Replace it with something with more wall thickness. (you weren't using that one on full soft, anyway)

Take a good look at your end links and any brackets that are in play. Those things can bend, too. I bent the rear swaybar endlink brackets on my Saturn many years ago. Some Hondas are prone to ripping their end link mounts out of the subframe (or whatever it is that they attach to). A beefy rear bar on a FWD car gets stressed A LOT, and your car is not light.

Loren, the end links and brackets are all in good shape. I upgraded both at the same time I installed the swaybar and so far they seem to be holding up well.
Before I go it alone with repairs I'm contacting the manufacturer about their lifetime warranty. Hopefully they will stand by it.

It's hard to see in the pictures, but it is the welds that failed. The arm is tig welded to the outside diameter of the tube on one side and the end of the tube is welded to the inside of the bore on the arm on the other side, then a circular cap is welded on to cover the end of the tube. That's why in the second picture it looks like there is a bar still welded in the arm - it's just the sheet metal disc welded in after the arm is welded on. It confused me at first because externally everything looks completely normal. The weld that broke is completely hidden between the bar and the mount bracket so it was impossible to see until I removed the swaybar.

[Loren]

Warranty is worth pursuing.

[dbeng]

I've no idea where I got May 3rd from in my first post. It was the March 11th event when I first had the handling problems. I'm losing it in my old age

[dbeng]

A quick update and it's good news!
I purchased the swaybar from James Barone Racing (JBR) back in July 2015 and as mentioned it carries a lifetime warranty for the original owner.

I sent an e-mail last night to JBR's customer service with a quick description of the problem and a picture of the failed swaybar. By 9:00 this morning I had a reply apologizing for the failure of the part and that they would send out a replacement immediately, I just needed to confirm my address. By 10:30 I had a tracking number and it's due for delivery by end of the day Thursday.

A fantastically quick and painless resolution. Top marks to JBR!

I might even have time to get it on the car before this weekend's event.

[Loren]

That's awesome!

I'd inquire as to whether they've seen this kind of failure before, and if so... did they improve the product so that it won't happen again?

[CaptainSquirts]

So funny thing, I noticed a co worker parked odd at work so I brought it up. He mentioned his rear sway bar went belly up also. He also drives a Mazda 3! I showed him the image of your broken sway bar and he said same exact thing happened to him. He doesn't autocross so I would expect this to happen again if they haven't done any upgrades to the existing sway bar

[Rpwolf]

Some Hondas are prone to ripping their end link mounts out of the subframe (or whatever it is that they attach to). A beefy rear bar on a FWD car gets stressed A LOT, and your car is not light.

I've seen that on some preludes, so at the time I had my beefy alum. Bracket spot welded to the chassis. Worth the piece of mind.

Nice to hear of companies offering superb service.

[Evil MS3]

Sadly, most of us from the Mazdaspeed groups know to stay away from James Barone Racing products. A lot of his products have been known to fail. Sucks that it happened to you and I'm actually surprised they replaced it, others haven't been so lucky in the past with his customer Service.

[dbeng]

A (not so quick) update to the swaybar story:
The replacement swaybar arrived on schedule on Thursday – Priority mail 2-Day USPS, here it is in all its shiny new glory!

IMG_1784-A.jpg (106.96 KiB) Viewed 10743 times

I'd inquire as to whether they've seen this kind of failure before, and if so... did they improve the product so that it won't happen again?

I would expect this to happen again if they haven't done any upgrades to the existing sway bar

So, when I received the email from JBR acknowledging the problem it included the comment "I give you my word that the one we are sending you will not suffer from the same failure" So from this I assumed they had made a modification to upgrade the construction and took a closer look when I received the replacement swaybar:

IMG_1788-B.jpg (129.68 KiB) Viewed 10743 times

The weld fillet on the new bar (on the right in this picture) is double the size of the weld on the original ; 1/4” vs 1/8” So there is a definite improvement in the strength of the welds. I ran the original swaybar for almost 3 years of autocross & street use before the weld gave up, I have to think this one will stand up to similar use for a much longer time!
And here it is reassembled with my end links and brackets ready to be reinstalled:

Sadly, most of us from the Mazdaspeed groups know to stay away from James Barone Racing products. A lot of his products have been known to fail. Sucks that it happened to you and I'm actually surprised they replaced it, others haven't been so lucky in the past with his customer Service.

That's strange - I can only comment on my experience - this is the only product I have purchased from JBR and the customer service was as good as it gets regarding warranty claims.

Continued in next post:

[dbeng]

When I reinstalled the swaybar on Friday evening I planned to install on its stiffest settings, but my non OEM endlinks prevented this. The OEM endlinks are physically smaller and are sealed. The Moog end links I am using have grease fittings and are physically bigger. They are sold as being more robust than the OEM parts and I’ve had no problem with them since first installing the swaybar. But the larger size & protruding grease fittings prevented me from using the inner holes (stiffest setting). Also they are a pain to install compared to the OEM endlinks. The OEM endlinks have hex sockets in the ends of the male threads so you can use an Allen wrench to stop the thread from turning while the nuts are tightened. The Moog endlinks have a square section under the thread that needs to be held with an open ended wrench while the nut is tightened. Since the upper end of the end link is buried up inside a formed steel control arm there is very little access to get a wrench in – requiring a crows foot wrench to be finagled in through the small opening in the bottom of the control arm.

This all works OK with the end link in the middle setting on the bar, but when the end link is pivoted forward to reach the inner holes on the swaybar the large diameter of the ball joint housing and the grease fitting hits the inside of the control arm opening and there’s no way to fully tighten everything. I fought with it for half an hour when it was already getting dark so I gave up and set it back to the middle setting.
So it felt fantastic at yesterday’s event to “have my car back again”! It was great to be able to get traction and I think this picture sums it up perfectly (Thanks Philip):

5-20-18.jpg (139.62 KiB) Viewed 10743 times

I may try picking up a set of OEM Mazdaspeed 3 endlinks – I checked and they are available for $27 each on-line. They should be small enough to fit on the stiffest setting and much easier to install with the socket end threads, it will be interesting to see if the stiffer setting helps improve the handling even more as I don’t feel much oversteer in the current setting, I think it could handle a little more bar without making the handling too twitchy. I will update this thread if & when I make any changes in the future.

[dbeng]

Time for an update. It's already a couple of weeks old, but finally had time to post something here.

My pair of OEM Mazdaspeed 3 endlinks showed up, so I set about installing them to set the swaybar on the stiffest setting.

Here is the original Moog endlink in position mounted to the middle hole on the swaybar as I have run it since first installed. It is physically too big to pivot forward to the inner monuting hole on the swaybar as either the link or the grease fitting hits the inside of the lower control arm.

IMG_1823c.jpg (181.48 KiB) Viewed 10459 times

Here you can see the difference between the greasable Moog and the sealed Mazda links:

IMG_1829c.jpg (179.9 KiB) Viewed 10459 times

As mentioned earlier, the Mazda link has a socket head in the end of the thread to hold with a hex-key while the Moog has the square profile below the thread to hold with a wrench. Interestingly you can see where the rubber boot has been damaged with trying to cram in a crows foot wrench that is much thicker than the width of the square while the nut was tightened. I guess if I use this type in future I will have to sacrifice a wrench and grind it thinner to match the width of the square boss and prevent damaging the rubber boot.
Also note the white plastic hood over the lower ball joint on the Mazda endlink. I assume this is to act as a 'cushion' to prevent metal to metal contact between the end link and the lower control arm to avoid any knocking or rattling from the OEM suspension.

[dbeng]

So I soon found out that these plastic hoods also prevented the new endlinks from pivoting over to the inner hole on the swaybar - quickly fixed with a vise and a dead-blow hammer:

Now I was finally able to tighten everything down in the stiffest setting

IMG_1835c.jpg (135.25 KiB) Viewed 10457 times

So how does it feel?
On the drive up to Dunnellon for last week's event, I really liked the feel of the steering, it definitely felt more responsive and "tighter". My first run of the day was on a fairly dusty surface in the first run group and braking for the first left hander on the main runway felt the rear of the car break away with some quick oversteer! That's not to say it was unexpected as I knew what stiffening a swaybar does to the handling, but nevertheless it still caught me off-guard as I am so used to the handling for the last couple of years. I had another small moment at the end of the run, preparing for the right hander off the main runway. So it is fair to say that I now have the rotation available that I've heard about so much for well set up FWD cars. Initial impressions are It feels good - until it doesn't, it will take a little time to adapt my driving to make the most of it, I'm looking forward to the next event at Brooksville as it's a more familiar location and surface to get a real feel for the handling changes.

In the meantime, here's a technical summary of the new handling characteristics toward the end of the first run at Dunnellon
https://youtu.be/WG26coNTNhw

That's it for now...
Dan

[aw614]

That was my first impression on that first left turn in that course