Sign up ×
Motor Vehicle Maintenance & Repair Stack Exchange is a question and answer site for mechanics and DIY enthusiast owners of cars, trucks, and motorcycles. It's 100% free, no registration required.

Aftermarket strut tower bars are generally considered to be decorative on many brand new cars. However, it seems to be a bit more logical that a chassis with 100K miles would have developed some flex and could use a cross-member to stiffen things back up.

Background: the specific car in question is a 2004 Subaru Impreza WRX wagon, daily driver with 110K miles, none of them off road or racing.

So, the specific questions:

  1. Does an aftermarket strut tower bar have any measurable effect during an enthusiastic daily commute?

  2. If so, how do you measure the effect?

I have to admit, there have been plenty of times when I have wished that there was a bar across the engine bar just to give me something to lean on (as opposed to the intercooler!). That doesn't seem like a sufficient motivator, though.

EDIT to follow-up on some of the points in the answers: the fundamentals of the suspension have already been sorted: Hotchkis sway bars with new bushings, KYB Excel struts (that are effectively OEM replacements) with Kartboy endlinks. This isn't an autocross car but the easy targets have already been addressed. Now we're literally talking about tightening up an aging chassis.

And the answer can be no, there won't be a difference.

share|improve this question
Has anyone here measured the effects on lateral force? Seems like a lot of talk about measurements but not a lot of hard data – user7960 Jan 3 at 17:15

7 Answers 7

up vote 3 down vote accepted

From my experience with a number of Subarus (my current oldest one being an MY2000 Impreza turbo with PPP and 130,000 miles on the clock) I can state that a strut brace makes an amazing amount of difference to both my usage (definitely enthusiastic) and my wife's (normal commute plus kids to school etc)

It stiffened the handling up significantly, makes cornering feel much more precise and makes her feel safer when driving. This is an old, well used car - and before adding the strut brace it did feel 'tired' - maybe the wrong word, but everything does get looser with age:-)

For me it just makes some manoevres on the track feel much less likely to upset the balance of the car - essential at 140mph plus!

share|improve this answer
thanks for the feedback. My second question is still open: can you measure the effect? Is there an effective chassis-stiffometer that we could attach to the car to show the before and after benefits? – Bob Cross Mar 8 '11 at 19:28
@BobCross - now that is something I don't have an answer for, sorry. I'll ask the guys at my Subaru garage though, and ping an answer back here. – Rory Alsop Mar 8 '11 at 21:25
thanks - that's the real motivation for the question. We all know that they're cool and we also know the limitations of the "butt dyno": if it was expensive, it must be awesome! – Bob Cross Mar 8 '11 at 21:33
@Bob - very late response: the garage don't have written figures, but they recommend it (not from a sale - I got mine almost free) but they showed me the flex with it disconnected and then bolted back on while lifting one wheel on a power jack. The twist in the chassis was easily visible to the naked eye without the brace, and about half as much (but still visible) with the brace fitted. – Rory Alsop Jul 19 '11 at 13:51
@JeffSwensen, thinking further about this, I think a slalom would be much more important in discriminating between a loose and tight chassis. In a steady-state skidpad run, the whole system will take a set and hold it. In a slalom, if the chassis is going to flex from one extreme to another. That will have an effect on the run. NOW, all that said, I don't have a skidpad or a slalom.... – Bob Cross May 16 '12 at 12:25

I would say it depends on the chassis. Since the GG chassis Impreza is relatively stiff to begin with, I'd say any effect would be near-placebo. Improvements would be in steering/cornering feel, rather than something that can be measured quantitatively, like better lap times. Also, they make a LOT of different braces for your particular car - not just the strut tower brace. Look into Cusco.

I've put strut braces on S13/S14 240s where the ride stiffened up noticeably and cornering seemed a tad more responsive. However on higher mileage cars, this effect was almost negated by the next weakest link - worn suspension bushings, worn shocks, inadequate anti-roll bars, worn/misaligned tires/suspension. I've put a strut tower brace on a brand new M3 (brace was an actual M Motorsports part) that wasn't noticeable at all given the quality of the chassis.

Also keep in mind that the design of the brace is important. There are cheap ebay braces that just attach at both strut towers that can flex by stepping on it with your foot. Quality braces will mount on several points - both strut towers and two attachment points on the firewall.

edit to be more aligned with your original question.. years ago we worked on an old datsun. New control arms. new steering bits. new racing coilover setup. new wheels and tires. all new bushings replaced. And when the car was pushed to the limit? The chassis flexed. You can almost hear it in some places.

The sure fire solution? Seam welding and/or a roll cage. Your car may be a daily driver, but nothing looks as awesome as a cage in a street car :) (eclipsed only by contusions suffered from smashing your un-helmeted head against the cage).

More anectdotal "information". I had spoken to a pretty established chassis shop that worked primarily on Evolutions. They had seen cars with all manner of aftermarket bracing - front/rear stbs, front "power" brace, braces in the trunk, braces that went under the fender.. braces underneath the car that tied the frame rails together. While these upgrades enhanced steering feel and overall "stiffness", in practice these cars weren't as fast as cars with well sorted out "dialed in" suspension.

share|improve this answer
good points on the basics - I added notes to the original question. – Bob Cross Mar 9 '11 at 0:41
an interesting suggestion on the roll cage. I've decided to stay married instead.... ;-) – Bob Cross May 20 '11 at 20:30

If so, how do you measure the effect?

tl;dr: by measuring the tension between the unbraced mount points.

Today, I installed a Whiteline rear strut bar on my 2004 WRX wagon. The car has 162K miles on the odometer and has always seemed in need of a little more stiffness in the area of the rear seats. There's a large volume back there without any sort of cross bracing (unlike a sedan) and the brace certainly seemed worth a try.

In the spirit of my original question, however, I wanted to see if I could measure the effect on the chassis in a static scenario. I tried two situations: flat on four wheels and with the driver's side rear wheel on four stacked 2x10 boards (which puts the wheel about six inches above the ground). I installed the circular fittings on the car and attempted to collect some reference imagery along with a little qualitative data.

NOTE: yes, in an optimal world, I would have used steel wire and a tension gauge. In the spirit of run what you brung, though....

First, leaving the wheel flat on the garage floor:

The inside of my car with some paracord, a reference line and some junk.

What you see there are the connecting holes on the braces with a double line of fairly hefty paracord running between them and tightened with a trucker's hitch. I added the sticky pads for some visual reference at the time. I later added the green line from the bottom of both connecting holes using If you've used a trucker's hitch, you know that you can get the line pretty taught: it made a solid "bwong" when I flicked it.

Here is the equivalent image when one wheel is pretty high up, balanced on some quite sketchy lumber:

I'm so glad that my wife didn't see what I was doing....

This picture is both disappointing and satisfying at the same time. The car clearly isn't made of rubber and tin foil (that's good) but it's also not showing an obvious flex that I could use to say "see? this was clearly money well spent and I am a genius!" Sadly, if you zoom way in, you can see that the front line looks largely the same as in the first image.

What is less obvious is what's going on in the rear line (the part where the knots are). That line is showing some droop (a bit but it's there). What you can't see at all is the change in tension in the line. Whereas before I had a nice "bwong," at this point it was definitely mushy.

At the end of all this, I think I can conclude that, yes, for this specific car, I can measure a definite opportunity for improvement in chassis rigidity that this part will address.

Later, when it's not the snowiest goddamn winter that the world has ever seen, I'll see if I can find a way to measure differences in handling that don't involve me being a huge donkey on public roads.

share|improve this answer
I got a kick out of reading this. Well thought out and pretty entertaining. – David Lively Sep 15 at 3:25
@DavidLively thanks! I aim to please! – Bob Cross Sep 15 at 9:27

I would say that anything that you can do on the street would be noticeably impacted by a strut tower brace, even during "enthusiastic commuting". The street is no place to be doing "nine tenths" or "ten tenths" driving, and even with spirited driving, on the street you are unlikely to have any noticeable impact.

If you aren't running R compound tires, you probably don't have the traction to make a tower brace do anything, even if you have fairly high performance "street tires".

If you think you're really pushing your car on the street, try going out for an autocross event. You can probably find an "solo" or "autocross" event within an hour or two of where you live pretty much any weekend in the summer. The "solo" events, as the name implies, are doing timed runs against the clock, not other cars, through a course set up with pylons. A properly set up course should have no opportunity for running into any obstacles, and will give you a feel for what sort of environment it takes to need a tower brace.

I've never heard of a car's chassis "getting tired", so putting a brace in would really make no difference.

share|improve this answer
My old Impreza's chassis was definitely tired (now 136,000 miles), and as I wrote in my comment below, it is definitely stiffer with the brace. As my wife drives it these days I am happy it makes things a little safer. – Rory Alsop Jul 19 '11 at 13:53
@RoryAlsop The tiredness isn't the chassis so much as the struts and bushing. Replacing those would go a long way to improving the feel of the car. Metal doesn't wear out, it will eventually stress crack but these is no loss of strength really prior to that. – draksia Apr 29 '14 at 19:38
draksia - yes, I regularly replace struts, bushings, entire shocks :-) – Rory Alsop Jan 3 at 18:34

Depends on the car I suppose.

I've heard a lot of people raving about how much it helped, but my experience is different.

I upgraded my stock front strut tower brace on the MR2 to the TRD model, installed the rear TRD brace, and also did an underbody brace at the same time. No difference detected at all. The car had around 140,000 miles at the time, with many of those miles at autocross, rallycross, and rough dirt/gravel road rallys. I did the upgrades as I was concerned about the car getting beat to death. Figured I should reinforce it. Turns out that I was worried about it for nothing. :-)

My Eclipse has a cage in it and originally had a front brace, but not a rear one. I added a rear one primarily to give me a place to fasten equipment to on long road rallys. It made no difference otherwise. I removed the front brace as it was always threatening to short out the positive battery terminal. No change noted. The car had about 90,000 miles, with many hard autocross, rallycross, road rally, and open track days on it. However, it has also had an 8 point cage since about 30,000 miles or so. Definitely not as well built a car as the MR2, the braces may have helped if it didn't have the cage, maybe.

share|improve this answer

Man....I have just the opposite experience as above...Three cars of mine, have strut tower braces...An 02 Toy Avalon, 04 BMW 325ci and a 95 M3...All 3 braces were "hingeless". All 3 cars are unibody contstruction....

All 3 braces made a difference but not all the same. The Toyota Avalon (Solara) brace made the biggest difference. The M3 brace made the least.

No quantitative testing was performed but qualitatively, the "turn in" response times are perceptable on all. I would expect a small increase in handling times as well...Ride Quality suffers but minimally.

These, when made specifically for the car, are very easy to install and add minimum weight.

Some vehicles are reinforced better than others. I would suspect the most dramatic effects from Front engine, FWD vehicles.

I hope this helps.


share|improve this answer
As mentioned before, the real challenge is that of quantifiable metrics. "Turn in" is a subjective measure that's heavily biased by "I spent money on this so I'm going to assume that things are better." – Bob Cross May 14 '14 at 18:51

It made a noticeable difference on my old Legacy, but I was racing that car. You might not notice it with normal driving. If you are noticing frame flex in your car though, you have problems that a stabilizer bar won't help. If this is bothering you, have the frame checked.

share|improve this answer
As mentioned before, the real challenge is that of quantifiable metrics. Is "noticeable difference" a reading on the butt dyno? Or is there a specific measure that you can cite? – Bob Cross May 15 '14 at 2:00

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.