Maximizing an ND-Chassis Miata for Street, Autocross and Track | Project Triple Threat Mazda MX-5 Miata

Sponsored by
Update by Andy Hollis to the Mazda ND MX-5 project car
Apr 11, 2021

Mazda designed their latest MX-5 to be equally capable on track and the street. Our question: Can it also be developed into a front-running autocrosser without ruining those traits?

Time to do a little work.


Building a Track Rat

We started with the top-performing 2016 model, a Club-spec car equipped with the optional BBS wheels and big Brembo brakes. And we broke in this one hard and fast: During our first week with it we ran the car four times at the track–with the first outings taking place at Harris Hill Raceway, a small club-style track with elevation changes and a fair number of bumps. 

One day was used to run in the engine, heat-cycle the tires, and bed the brakes. We pushed the car hard the second day to establish some baseline lap times.

As delivered, the MX-5 handled quite neutrally. It made solid enough power for a club-style track, and the brakes delivered impressive performance. 

The car felt a bit skittish, though, and it was tearing up the outside edge of the tires–both indicators that an alignment was in order.

ND Miatas are not aligned, per se, at the factory. As with previous generations of the car, caster, camber and rear toe adjustments are accomplished via eccentric connections located at the inboard ends of the lower control arms.

During vehicle assembly, all of these eccentric connections are simply pointed straight up, but the acceptable alignment specifications are so wide that the car is still considered to be in spec. A visit to our friends at Automotive Specialists in Austin, Texas, revealed that on our car, rear toe was skewed to one side and a there was a mix of positive and negative camber. 

We aligned it with traditional Miata settings of zero toe all around, with maximum negative camber at the front and maximum positive caster. We then enhanced stability by giving the rear a half-degree more negative camber than we had at the front. 

Once it was back on track, the car felt much more predictable. It was also easier on the tires. The final benefit: lap times had decreased by seven-tenths of a second. 

Later that same week we ran on the super-smooth, high-speed Circuit of the Americas. While the top-down car was aero-limited on the long straights, it heeled over onto its rear bump stops in the smooth sweepers. 

We started looking for a fix by examining the camber curves posted by Keith Tanner at Flyin’ Miata. He showed that the front of the new MX-5 gains much less camber than the rear as the body rolls. Combine that with the rear bump stops contacting before the fronts, and you get a car that is neutral in steady-state cornering, but limited by front camber.

A proper suspension alignment definitely helped our MX-5. The factory gets it close, but it could be closer. Koni dampers will also make a big improvement. 

The easiest fix to control body roll and the resultant loss of camber is to install a larger front anti-roll bar. There are several inexpensive OE-style bars on the market, but as we covered last issue, installing the front bar involves a bit of maneuvering up front.

However, an alternative exists: If you are willing to plunk down more money and cut out the original front bar, you can enjoy the immediate gratification of a half-hour install with a splined circle track-style bar. An added benefit is that these bars can offer much more stiffness. 

We chose the Karcepts circle track-style anti-roll bar that’s made specifically for the ND-chassis MX-5. Their bars feature five-way adjustable arms and a replaceable center tube that can be ordered with one of three different wall thicknesses: 0.095-, 0120- and 0.188-inch. 

It’s a stiff bar, too: Karcepts’s math says that the middleweight bar–the one that we ordered–offers settings that range from 3.25 times stiffer than stock to 5.27 times stiffer. Compare that to the popular Progress adjustable front bar, which offers roughly 3.00 to 3.77 times additional stiffness. The Karcepts bar costs more, though: $575 vs. $226.

Once we had the new front bar installed and set to the middle hole, our lap times at Harris Hill dropped a full second. Despite some added mid-corner push, the car was now much easier to trail-brake deep into a corner; it also put down the power better when coming out of the turns. Tire wear improved as well, with the rears now wearing evenly across the face and the fronts wearing further from the outside edges.

The next major improvement came after we finished off the OE Bridgestone Potenza S-01 tires. We chose their replacements with an eye to maximum performance in all conditions for extended sessions, which made the Michelin Pilot Super Sport the natural choice. While not quite as quick as the leading extreme performance tires for short time trials in dry conditions, the Pilot Super Sports offer well-known all-weather versatility. Result? An improvement of another full second in lap times, with no increase in road noise or harshness. Win-win.

A set of Michelin Pilot Super Sport tires helped us at both long and short tracks.

Developing a Cone Dodger

As the Tire Rack SCCA Solo Nationals approached, we began the third leg of our triple threat development effort. Step one was to get proper autocross tires and another set of wheels to mount them on. 

We selected the winner of our latest autocross test, the Bridgestone Potenza RE-71R in a 225/45R17 size. Not only is this tire quick and easy to drive at the limit, but it also works well when pinched onto narrow rims. This is critical for anyone running the ND Miata in the SCCA’s Street category, where the 225’s larger overall diameter allows for a couple of extra mph in top speed over the original equipment 205/45R17 tire. Thanks to their softer sidewalls, other 225mm tires tend to get mushy and vague when crammed onto a 7-inch-wide rim, but we are happy with the RE-71R’s feel. 

Lightweight aftermarket wheels that are legal for C Street can have issues fitting over the MX-5’s Brembo brakes, as clearance between the wheel and the innermost edge of the caliper can be a challenge. This is why Mazda’s Brembo option package also requires the BBS wheel package: It provides a narrow inner mounting pad that easily clears the brakes. We decided to replace our BBS wheels with the popular Enkei RPF1, which offers a great combination of weight, strength and price; its ET43 offset specification requires no more than a 5mm spacer to meet the SCCA Street-class regulations, since OE is ET45 and rules allow plus or minus 7mm. 

We created enough clearance for the Enkei wheels by taking advantage of some slop in the caliper mounting holes to push the calipers outward. The calipers still just ever so slightly scratch the paint off the backs of the wheels under hard cornering loads, but the wheels fit and are legal. With the spacers added, the stock studs allow for about seven turns of the lug nuts, which is within Tire Rack’s recommended minimum engagement, though we won’t run them that way on the track.

Next it was time to get some early baselines. We took the Miata to the SCCA Championship Tour in College Station, Texas, so we could run against some of the best C Street autocrossers on concrete. We ran on the original Bilstein shocks, and found that while the large front anti-roll bar reduced the chassis’ tendency to crash onto the rear bump stops in sweepers, it continued to do so in quick transitions, making the car a challenge to slalom consistently. The car also felt vague and lacked crispness on initial turn-in.

We solved the latter problem after the next day’s first run by adding a quarter-inch of front toe-out while the car sat on the grid. That change put us a half-second closer to the fast guys, but we were still a solid second off the pace for the day. We also noticed that our car sat higher than the rest of the field and had the least amount of front negative camber. Clearly, there was work to be done.

First, we looked at our alignment. Removing all of the adjustable caster gave us more static negative camber–and while some will argue that caster provides beneficial dynamic camber, in the case of the ND Miata, it comes at the expense of too little static camber.  

We also found some slop in the mounting holes for the upper control arms. We were able to use those loose tolerances to move in the upper control arms as much as possible.  

Those two changes gave us another half-degree of negative camber. Testing at the track revealed that this equated to another half-second improvement.

That’s when we heard that the first sets of production Koni Sport dampers for the ND Miata were coming into the country. (They list for about $200 each, but street prices are less.) We’ve long been fans of these for Street-category autocross cars, since their low-pressure gas design allows for additional lowering of the car. (OE and some aftermarket monotube dampers require significant gas pressure, enough to raise ride height.) 

That lowering typically translates to more static front negative camber on a double A-arm suspension, which for a camber-challenged car like the ND Miata is money in the bank. And while the twin-tube design does not offer either the range or precision of adjustment of a high-end monotube, its off-the-shelf valving is typically sufficient for OE springs.

Mounting the Konis is relatively easy: Once the front upper control arm is removed from its inner mounts and the anti-roll bar is unhooked, the OE coil-over assembly can be unbolted and removed. In the rear, you can create enough clearance by just pushing down on the suspension. Other than drilling out the OE mounting washer to fit the Koni’s larger shaft, swapping the spring assembly between dampers requires only a quality spring compressor. 

Our Konis felt great out of the box. Revalving the fronts and removing the gas charges all around made them even better. 

The change in ride quality on the street was immediate and positive. The OE Bilsteins provided an annoying “jiggly” feel over any sort of irregular pavement, thanks to the fact that there was quite a bit of high-piston-speed compression damping, especially in the front. Combined with the high gas pressure’s significant lifting force, this kept the car off the bump stops, but at the cost of ride quality. They worked great on track, but not so much on the local twisties.

The Konis have a more digressive valving that reduces the high-speed compression damping. Combined with lower gas pressure, it means the twin-tube design feels much more connected with the road. The only downside that we noticed was observed at higher speeds over significant rises and falls: We’d feel it in the pit of our stomach when the car came down on the bump stops and the rebound damping kept it there for a moment. Still, this is a minor annoyance compared to the OE Bilsteins’ constant nagging.

Next we headed back to our standard autocross test course in Mineral Wells, Texas, to work through a series of adjustment settings on both the front bar and the shocks. We found that we could turn the same times with the bar set at any of the three softest settings, but we had to drive the car differently to do so. 

At full soft, the car was very neutral around sweepers, but we had to be perfect and early on our slalom inputs; putting the power down off the corners was also touchy. With the bar set to the middle hole–the third of the five holes– the car could be hammered through transitions and would easily put the power down off the sweepers, yet it pushed mid-corner. Hole No. 2 offered a little of each effect.

The shock absorbers were mostly useful for tightening up the steering response in transitions and keeping the car off the bump stops. While the rear had plenty of adjustment range, we found that we were using everything that the fronts had to offer.

So we decided to add more “more” to those front Konis: We sent them back to their maker for a revalve. We added about 30 percent to both the front rebound and compression curves, and also de-gassed them at the same time. The latter removed the last bit of static lifting force of the twin-tube, an improvement that lowered our car another quarter-inch in the front and also added a touch more negative camber.

Once we hit the country roads for the three-hour drive to our testing site, we learned that ride quality was still quite good with the shocks set to full soft. On the test course itself, we found that the extra adjustable damping range up front tremendously helped responsiveness. We were now finally turning times that were right with two of the same national-level competitors who had been a full second ahead of us earlier in the season.


Showing Up for the Big Show

While everyone likes a happy ending, racing can be a cruel mistress. Once we hit the pavement for the Tire Rack SCCA Solo Nationals, we learned that too many track events and not enough autocrosses meant that our driver couldn’t just phone it in, especially with a record-setting 86 entrants in the class. 

Almost all of our autocross seat time in the ND had come during short test and tune runs, so when it came time to string together more than a minute of execution at Nationals, our brain let us down. Our data logger showed us that on the first day, we nailed every single element on the course right with the leaders–just not all on the same run. Combine that with Day 2’s drying conditions, and we just didn’t finish where the car could have. Still, we took home a trophy and beat 71 other drivers–not bad for a month’s worth of autocross development.


Reflections on a Solid Finish

Okay, we didn’t win that big show. The scene didn’t end with us showered in champagne. 

We did succeed in producing a car that is equally adept at autocross, track and high-performance street use. A simple front anti-roll bar, alignment and tires found us a 3.5 second improvement on the track. Adding just a low-cost set of dampers and autocross-specific tires gave us the tools to run at the front nationally in C Street autocross competition.

In our book, that’s a win. 

Our MX-5 felt spot-on at the SCCA Solo Nationals. After a slow start, we bounced back to finish 14th out of 86 entrants. 

Join Free Join our community to easily find more project updates.
View comments on the GRM forums
Sponsored by

GRM Ad Dept

Our Preferred Partners