Driving Impressions: Lower Stress Bar

[insite]

in the midst of searching for the latest greatest 'go-fast' parts, i found myself in conversation with Ernie and Stan from Mantis Sport in Canada. i ran into them at the Road Atlanta PCA race this year; they campaign the only Cayman race car around:

[insite]

since they have now over one year of experience racing the Cayman platform, i asked what they've found a need for and what they've installed on their car. in addition to a host of other stuff, Ernie mentioned a lower stress bar to me. the front suspension pick-ups are all mounted to a solid single subframe. in back, however, the subframe is split. the two halves are reinforced with a shear plate, which is great for shear or tensile forces, but poor for compressive forces:

[insite]

the lower stress bar was designed to jack the subframes apart and hold them that way during driving. the result is that camber is maintained during hard cornering. most people thing that addition of an upper strut tower brace on this car is helpful. the problem at the rear is not the top of the strut towers, it's the bottom of the subframe. when you turn, the outside control arm is pushed toward the center of the car. this causes chassis flex and camber loss. here's a picture of the solution (i call it the ernie bar):

[insite]

install was a snap. i removed the nuts that fasten the diagonals to the shear plate. i pounded out the studs with a hammer. i test fit the ernie bar to set its length, then installed it permanently. piece of cake; takes about 15 minutes once the car is jacked up:

[insite]

impressions: this thing is great! i wasn't sure how much i'd notice it. i went to the track last monday and put it to good use. from the driver's seat, the car seems to set much more quickly on corner entry (a perception reinforced by swapping drivers and boxsters). from a measurement perspective, i was pulling significantly higher lateral acceleration numbers than i ever have. in the mildly banked corners, i was at 1.3 g's. in the unbanked corners, i was between 1.15 and 1.2G's. this was on STREET tires. once set, the car also seemed less prone to unsettlement over bumps. the car just doesn't seem to flex as much in back. here's one lap of telemetry:

[insite]

bottom line: this thing was like $130 shipped. it really works! between my last track day and this one, i did the following:

996 front brakes w/ racing pads
mantis lower stress bar
bilstein sport front dampers

lap times went from 1:15:20 to 1:12:72. that's 2.5 seconds on a 1.41 mile course! i'll take it.

[chaudanova]

wow, thanks for posting this up, that's awesome to hear. Do you know what G's you were pulling prior to the addition of the bar? And also, do you notice any difference on street driving?

[insite]

Originally Posted by chaudanova

wow, thanks for posting this up, that's awesome to hear. Do you know what G's you were pulling prior to the addition of the bar? And also, do you notice any difference on street driving?

on this course, i averaged 1.06G with a max of 1.13G. this thing put me in the stratosphere here. it really holds the camber. as for street driving, it doesn't effect comfort levels at all. i don't really notice it under normal driving circumstances. it's very noticable when i push it.

[BobMarley1]

how do I order one?

[MNBoxster]

Hi,

How are you measuring your lateral acceleration? You're throwing around numbers here (both Baseline and Modified) which are greater than those repeatedly published for the Boxster S (0.91g).

While that may be so of a Modified Car, I doubt it's the case with a Base Boxster. I suspect your Data Logger is miscalibrated. How else do you explain it?

Happy Motoring!... Jim'99

[blinkwatt]

Does the company that you bought the stress bar from have a website?

*Never mind I found it;
http://www.mantissport.ca/
I can't seem to find the lower stress bar though.

[insite]

Originally Posted by MNBoxster

How are you measuring your lateral acceleration? You're throwing around numbers here (both Baseline and Modified) which are greater than those repeatedly published for the Boxster S (0.91g).

While that may be so of a Modified Car, I doubt it's the case with a Base Boxster. I suspect your Data Logger is miscalibrated. How else do you explain it?

logger is not miscalibrated. it's a racepak gps; the accelerometers and GPS calcualted lat G's are the same. it's easy to explain:

1. modified car
2. turns 1 and 2 are slightly banked
3. alignment of a stock boxster s sucks, not enough camber

[insite]

Originally Posted by BobMarley1

how do I order one?

their website is here: http://www.mantissport.ca/

you can't order online, though. call them at 905.844.6219 and ask for stan. btw, any prices listed on their site are canadian.

[insite]

for those interested, here is a graph of my fastest lap showing lateral acceleration derived two different ways:

1. on board accelerometers (red line)
2. calculated from GPS polling data (green line)

the RacePak polls gps several times a second. there is some latency and interpolation. this is why the green line lags the red line by a little bit. magnitude, though, is very comparable between the two. the data is accurate.

[racer_d]

Interesting discovery. Reminds me of the problems I had with my 914 experiencing trailing arm flex under high g loading (expecially with wide sticky tires, compared to the then stock, skinny, non grippy ones )

I'll have to keep this mod in mind. Wonder how SCCA/PCA allows it for, say, Autocrossing?

[insite]

Originally Posted by racer_d

I'll have to keep this mod in mind. Wonder how SCCA/PCA allows it for, say, Autocrossing?

good question. i bet it falls into the same category as a strut tower brace. probably would bump you from showroom stock to production, but not from production to improved. just a guess.

[John V]

Magazines measure maximum lateral acceleration around a flat skidpad. They drive the car at a known radius and then measure time to complete a lap (or multiple laps). Then they calculate the lateral g's generated.

The maximum g's that a car can develop steady-state around a 200' skidpad is different than what it can generate around a banked corner. Or a corner that requires less steering lock. Or at high speeds. There are too many variables, which is why magazine test numbers are for the most part useless metrics.

Another discrepancy can come as a result of the leaning of the car, which induces a false lateral acceleration component. Fortunately, the more sophisticated data acquisition systems account for this with multiple accelerometers.

In other words - I believe your numbers. We see similar numbers in our DL1 data - peaks of up to 1.4g lateral but more imporantly, around 1.2g steady state. This is a stock-class Boxster S with (really old) Kumho V710 competition tires.

[chaudanova]

Insite, any updates on your lower stress bar, in terms of how its been holding up, any changes/updates on driving impressions on both street/track?

Thanks bud...

[insite]

it's still awesome. unnoticable under normal driving conditions. very noticable when you push the car, especially over bumps. without the bar, the car will lose camber over bumps, causing oversteer conditions. with the bar, things are much more tidy. i really recommend this.

[Topless]

Nice work Insite,

You patiently seek out leading edge performance upgrades, act as test pilot, verify your impressions with instrumentation and lap times and share your findings with all of us. Bravo!

This is not bolt on fluff, but tested and verified with results. 2.5 sec on a track you already have wired is pretty good stuff. Just for kicks, if you had not made these improvements to your suspension and braking and focused only on the motor, how much HP increase would you need to get the same lap times? 50? 80? Interesting question don't you think? I suspect you are already outrunning most of the 987's in your area with a stock 2.5 motor.

Thank you again for sharing your findings.