Stark Performance Paddle Shifter Extensions: Hand Crafted in the USA

As product development and manufacturing gets outsourced to overseas manufacturing facilities, it has become increasingly difficult to find products that are still made in the USA.  In the race for the lowest possible pricing, quality is often compromised while chasing the least common denominator.  At Stark Performance, quality is of paramount importance and if that means we have to do things the old fashioned way, by hand, in order to provide quality with reasonable pricing, then so be it.  Doing things manually, however, comes at a price.  Since a person’s time has value, the more skilled a person is, the more valuable their time.  As such, labor intensive manufacturing methods by skilled artisans are by default, more expensive than parts made by machine, or by manual labor in countries that only pay $5/week.  The parts we manufacture are made with attention to detail and fueled by passion.  The following is a little insight into the past 18 months of development and refinement of our paddle shifter extension production process.

The first step was hand sculpted prototypes made from light cure acrylic putty; finished and painted to seal the surface as well as distinguish this master pattern from production castings.  These prototypes represent over 160 hours of skillful sculpting, shaping, sanding, polishing, painting and prepping.


As a proof of concept, the hand sculpted prototypes were good, but had their flaws.  They were then 3D scanned and manipulated in CAD to refine and even out the edges and contours.


Once the CAD work was completed, the new master patterns were 3D printed using Fused Deposition Modeling (FDM).  While FDM is a good process for prototyping and proof of concept, it is not accurate enough for the tolerances required and the finish left much to be desired.  As such, 3D printing is not a good manufacturing process for a part that will be handled on a regular basis.  Another limiting factor for production was the 8 hours required to print each 3D master pattern.  In addition, there were several misprints before achieving final, acceptable patterns.  Once the master prints were complete, the ridges left over from printing were hand sanded smooth and the contact patches were further refined for a perfect fit.  Creating a smooth surface that mated to the contoured surface of the factory paddles was not an easy feat and required multiple techniques to achieve.

Using these master patterns, silicone molds were made to cast parts.  Each mold requires 4 days to make and is able to produce 15 – 25 parts before the silicone breaks down and fails to produce quality parts.  There were multiple iterations using different mold designs, different durometer silicones and even different silicone chemistries before the molds produced parts with consistent fitment.  This was due in part to the stiffness of the mold and its coefficient of thermal expansion that would cause distortion during the exothermic polymerization reaction that occurs in both the initial mold production as well as the casting process itself.

Through research and development several key pieces of specialized equipment were constructed to produce bubble free castings.  Since polyurethane resin absorbs water from the atmosphere and the isocyanate in the resin reacts with water to form carbon dioxide gas during the polymerization process, bubbles will form very easily.  In addition, the reaction is temperature sensitive and the liquid mixture solidifies faster when over 70 degrees Fahrenheit.  If not poured within 30 seconds of mixing, the viscosity will increase to the point of failing to fill the mold and result in a miscast.  The casting area therefore requires dehumidification and temperature control to mitigate bubble formation and ensure sufficient mold filling.

Each cast requires thorough degassing of the resin using a vacuum chamber to remove all the moisture and any entrapped air.  A secondary vacuum resin trap is used to draw the resin into the mold cavities and once the mold is filled, it is placed into a pressure pot to compress any remaining bubbles before the resin has polymerized.  The resin is allowed to fully polymerize in the pressure pot prior to de-molding.

P1000286 P1000287 P1000288

Once the resin is polymerized, the parts are de-molded and the vents and sprues from casting are manually removed.  The parts are left to finish curing overnight so they reach full hardness (Shore 80D).  The result is a part that is as hard and stiff as ABS plastic (used to make most automotive interior parts) and has a temperature of deflection of 200 degrees Fahrenheit.

One of the advantages of having plastic parts that are touched regularly is that plastic has a relatively low coefficient of thermal conductivity.  The human body is incapable of detecting temperature, but can detect heat transfer.  When something feels hot, it is simply warmer than the part of the body that is in contact with it.  The temperature differential results in heat transfer through conduction.  Metals feel hotter or colder to the touch than plastic because metals transfer heat at a faster rate than most other materials.  This is one of the reasons why modern automobiles use non-metallic parts for surfaces that are touched frequently and why these extensions are plastic as well.

Before any additional finishing work is done, the parts are checked for fitment using a water soluble paste.  If the fitment is only slightly off, the part will be adjusted until it has good contact or discarded altogether.

P1000301 P1000302 P1000303

Since a mold release is used to slightly prolong the short service life of the silicone mold, the parts are put into a vibratory tumbler to remove the mold release, ensuring good paint adhesion later on.


While the part finish is fair, there is still a parting line and the remainder of the vents and sprues are manually removed for a clean finish.

P1000275 P1000277 P1000278 P1000280

The parting line, vents and sprues are ground flush and every edge is smoothed to remove any irregularities felt while handling the parts.

P1000281 P1000282 P1000283

The grinding marks are sanded smooth with a two stage wet sanding before they are ready for primer.


The parts are then masked off and mounted on a fixture for priming.  This ensures that the fitment area does not get distorted by paint since the tolerance on that area is +/- 0.002”.  Once primed, the surface is wet sanded, cleaned, and carefully inspected prior to the application of an automotive interior grade satin black paint followed by an automotive interior grade satin clear.

P1000289 P1000290 P1000293 P1000295 P1000297 P1000298 P1000300

Once the finish is deemed satisfactory, the parts are then removed from the fixture and the die cut adhesive is applied.

P1000304 DSCF1590_MS

The adhesive is a 3M VHB that has a long term temperature tolerance of 300 degrees and a short term temperature tolerance of 500 degrees.  As such, the adhesive maintains an excellent bond at the elevated temperatures the interior of a car can reach when parked in the sun on a hot summer’s day.  Several other adhesives were tested and found to perform poorly in comparison despite being rated for the expected conditions.

The end result is a part that is aesthetically pleasing and integrates seamlessly with the rest of the interior.  It also provides excellent ergonomics while improving the overall driving experience.

installed car 2installed car 1


Automatic Transmission Cooler Testing and Development Part 3

With cooler weather rolling into Southern California this winter we noticed that the transmission temps were a bit cooler than we’d like to see with our transmission cooler installed. While we plumbed it so that the fluid goes from the transmission to our auxiliary cooler to the factory heat exchanger and back to the transmission, the factory heat exchanger can’t quite add enough heat back to the fluid to warm up sufficiently. In some cases of cruising on the highway in 40 degree weather we saw transmission fluid temps drop as low as 127 degrees at steady state. Considering that our initial testing indicated that the OEM engineers designed the system to run at higher temperatures (189 – 200 degrees) and viscosity is affected by temperature we started working on a thermostat option for people that drive their cars in cooler weather.  Here is a diagram of how the thermostat gets plumbed:

T-Stat Plumbing Diagram_Flat

We started by adding an off the shelf, inline thermostat rated at 180 degrees to the system. Initial testing showed that the thermostat actually started opening at 160 degrees. Since thermostats are generally rated by their opening temperature, we thought that was a little odd. After discussing with the manufacturer of the inline thermostat we opened up the housing to swap out to a higher temperature wax motor. Low and behold it was equipped with a 160 degree wax motor as the temperature was stamped onto the housing. Apparently, this is common practice with the oil thermostat manufacturers for some reason.

While we tracked down a suitable replacement wax motor rated at a higher temperature, we did a little long term testing with the 160 degree thermostat installed. It did help bring the temperatures up a little higher, but they were generally in the 158 – 163 degree range cruising on the highway and up in the 169 – 177 degree during extended street driving. Obviously, the transmission cooler is more effective at highway speeds at shedding heat.

We finally were able to source a 195 degree wax motor that works in the thermostat housing, installed it and started doing some testing. Steady state cruising on the highway is now closer to OEM at 170 – 180 degrees. While one would expect the temperature to be closer to the opening temperature of the thermostat, approximately 10% of the fluid flowing through the thermostat does go out to the cooler circuit when the thermostat is closed. This is to prevent thermal shock when the thermostat opens as well as keep the cooler circuit pressurized.

In addition, we did some quick track testing to verify that the thermostat does not adversely affect the cooling ability of the system. We were fortunate enough to get 4 runs in our session (1 more than when we did the initial track testing). We are happy to report that the transmission outlet temperature stabilized at 223 degrees at the end of the 3rd run and did not exceed that temperature during the 4th run.

In conclusion, we are now offering a thermostatic option as an add on for those who drive their cars in cooler weather and would like their transmission temperatures to be closer to stock when adding our cooler kit.  Now available for purchase from our web store here.

Automatic Transmission Cooler Testing and Development Part 2

Now that we have established the need for additional cooling, we set about selecting a cooler core from Setrab.  Setrab is well known for making high quality competition level heat exchangers.  While we could have selected a less expensive core, those cores are generally not designed to flow air well at the high speeds seen on a race track.  Most of the less expensive cores are designed for towing (high thermal loads at low vehicle speeds).  As such, they are designed to increase turbulence to maximize heat transfer.  At high speeds, however, they are restrictive to airflow and unable to cool as efficiently.  Setrab cores, on the other hand, have a straight through fin design that balances turbulence with high speed efficiency.

Setrab Cooler Core


Alternative Core


Notice that the alternative core requires the air to be diverted at an angle in order to pass through the core.  While this is great for low speed efficiency, it is not the best for racing.

With the Setrab core installed, the transmission maintained normal operating temperatures for a solid 30 minute enthusiastic run up the mountain.  We are pleased to say that this works very well for spirited driving up a mountain pass.  Track testing showed a marked improvement over the factory cooler setup, but has a little room for improvement.  While temperatures peaked out at 233° Fahrenheit and below our established threshold, we would like the temperatures to be more in the 220° range.  Therefore, we did some airflow modifications to the front fog light cover and fender liner to get air in and out more effectively.

IMG_20141001_115804_008 IMG_20141001_155143_633

With the airflow modifications, track performance was precisely where we wanted it to be with transmission temperatures peaking out at 221° Fahrenheit.  During the last session, the temps stayed steady at 210° and only started to go up once we slowed down.

Overall, the cooler works well and does what we need it to do.  For cars that do not see track time, the airflow modifications are not necessary as the cooler gets enough air to do the job extremely well during spirited driving.

Here is a picture of the kit and what it looks like installed:

 Transmission Cooler Logo P1000194

These kits are now available for sale on our webstore here.

Automatic Transmission Cooler Testing and Development Part 1

We wanted to see how well the factory transmission cooler circuit performs under various conditions in the real world.  So we set about splicing in pressure and temperature sensors to the factory transmission cooler circuit.  On the FR-S there is an external heat exchanger that transfers heat from the transmission fluid to the coolant.  The coolant that circulates through the heat exchanger is also part of the heater core circuit and is therefore at a lofty 200 degrees Fahrenheit.  Fortunately, the FR-S uses Toyota WS transmission fluid which is a synthetic fluid and is capable of withstanding much higher temperatures than traditional mineral based fluids that start breaking down at 180 degrees.

Steady state testing at idle showed the transmission fluid to stabilize at 189 degrees.  Once on the highway, steady state cruising yielded 194 degree transmission fluid temperatures.  City driving with its stop and go pattern had elevated temperatures, but even with moderate acceleration, they were by no means alarming at 205 degrees.  Overall, the factory system works as intended for daily driving duties.

Driving up a mountain pass enthusiastically, however, reveals that the factory heat exchanger is unable to shed heat fast enough to maintain reasonable temperatures.  Within 2 minutes of spirited driving, the temperatures shot up to 235 degrees.  While the synthetic transmission fluid can withstand high temperatures, the other components in the transmission cannot.  Generally, seals will start to harden and crack over 260 degrees and the clutches will start to slip at 295 degrees.  Therefore, we really don’t want temperatures to exceed 240 – 250 degrees for longevity.

Track testing with the factory setup naturally showed how woefully inadequate the system is for that kind of stress.  While the car is advertised as being able to carry a set of track wheels and tires in the trunk with the rear seat folded down, it is not truly designed for track duty.  In 8 minutes of track driving, the transmission temperatures climbed up to 244 degrees.  In short, an automatic equipped car needs to have additional cooling if it is to be driven on a track.  See Part 2 for our Transmission Cooler Development and testing.