As we gear up for the 2018 American Solar Challenge, CalSol is excited to reveal our new logo! For 10+ years CalSol has kept our iconic “Flying Cells” logo as the center of our team’s branding. While this logo has served us well over the years, our new logo helps better represent CalSol’s modern existence and demonstrates our continued progression as a team.
When we set out to redesign our logo we wanted to both preserve our previous brand, while modernizing, simplifying, and increasing brand flexibility. Our new logo retains CalSol’s existing default blue and gold scheme, while also continuing the inclusion of solar cells as an integral part. The new logo also gives us more flexibility by including a circular icon, which is a useful form factor for digital media (something not heavily considered when we conceived our last logo in 2005). We are also introducing a number of alternate color schemes beyond our standard blue and gold, affording CalSol additional branding flexibility.
We hope you are a fan of our new logo! We will be rolling it out across our various platforms leading up to the American Solar Challenge. Stay tuned for exciting updates on both Tachyon and our final race with Zephyr.
Circuit boards are an essential part of our solar car, handing all of the controls and communication between the various systems of the car. On CalSol, the electrical team is responsible for designing and programming the printed circuit boards (PCBs) required to operate the solar car. From a complex board that monitors the status of the battery to a simpler board that reads how much the brake and accelerator pedals are pressed down, each PCB reads data from sensors that it’s connected to, processes the information, and transmits it to the rest of the car.
We start designing each board with high-level block diagrams of circuit components, and develop these into a schematic that outlines how all of the parts will be connected.
Next, a layout is made to package everything like how it will be on the circuit board.
Finally, we send the completed board designs for manufacturing. Bay Area Circuits (BAC) is a long-time sponsor of CalSol, and generously manufactures PCBs for us to code, test, and run on the car for no cost. Thanks to them, we can take our designs from ideas to reality and build the best car we can. We encourage you to check out their website at bayareacircuits.com!
We recently switched to using 4 layer circuit boards, as opposed to simpler, easier-to-manufacture 2-layer boards. Using 4-layer PCBs allows us to use smaller boards by accommodating a higher part density along with more resistance to electromagnetic interference (EMI), both of which improve the electrical system of our solar car. While this makes the design easier, they are harder to fabricate. Bay Area Circuit’s willingness to help us with this more complicated process has greatly streamlined our designs, allowing us to spend more time on other aspects of creating a solar car. BAC has also provided us with engineering support and design advice, helping us learn best practices for designing printed circuit boards.
Thanks to BAC’s fast turnaround, we are able to make multiple versions of each board to iterate through designs and correct mistakes before they become a part of the car. We were able to assemble a test bench with our first run of boards to begin testing our firmware. We connected everything in a configuration similar to how it will be on the car, enabling us to test the full system before it is integrated into the actual car.
The test bench helped us verify that all of the boards are functional, but we also had to fix many small mistakes. We’ve received and almost completed soldering the next iteration of circuit boards, which will be the ones powering Tachyon!
Whooo! It’s been a while since our last post. We’ve been hard at work fabricating our molds and laying up Tachyon’s shell and monocoque.
Now for making our molds. As we learned in our design process, “negative” molds are much preferable for making a smooth solar car exterior, and the obvious choice is to shape these directly from foam. This choice was especially tempting when factoring in DUNA-USA’s generous sponsorship of high-density tooling foam. Thank you DUNA-USA for your support!
We considered a few types of foam for this purpose. Polystyrene, the material used in Styrofoam, was one option. It is cheap and readily available, as well as very lightweight. However, it melts at a lower temperature than our prepreg carbon fiber required for curing. It is also chemically incompatible with the primers and gelcoats we intended to use for smoothing the mold surface: we would need additional protective layers, or risk melting the foam with our polyester-based sprays. Thus, it was not a good option for our foam negative mold.
Polyvinyl foam is chemically compatible with polyester primers. It is slightly pricier, but is available in higher densities: we could achieve a better surface finish after milling. It has a higher decomposition temperature than polystyrene, but the softening temperature was still uncomfortably close to our prepreg cure temperature.
That left polyurethane tooling foam, like the blocks donated by DUNA-USA. It is available in very high densities, which leaves a good surface texture after milling. The foam is also chemically compatible with the chemicals we intended to use, and can withstand the temperatures needed to cure our prepreg. However, large blocks of high-density tooling foam can be very heavy: Tachyon’s negative molds could have been over two thousand pounds!
This ruled out lifting the molds by hand for everyday transport, or into our oven for baking. Our self-built oven could only be accessed through a relatively narrow door. Since the molds would be so long and heavy, the forklifts at our workspace could not lift them from the short ends without tipping over. There was also no room for a long ramp at the oven door, for us to roll the molds into the oven.
In short, tooling foam negative molds would be too heavy for us to move regularly and use. We needed a stiffer and stronger material to make molds that we can transport by hand.
The solution was to make composite negative molds, which would be light enough for us to lift by hand. This required a two-step process. First, we would make positive “patterns” from foam, where the smooth, final surface is on the outside of the mold. Next, we would make our composite negative molds from these patterns, exactly transferring the smooth surface to the inside of the negative molds.
The patterns were only needed once; to make the negative molds. They did not require frequent transport, so weight was no longer such a major concern. With this change in process, DUNA-USA’s polyurethane foam again became the material of choice for us.
The foam came to us in smaller pieces, so we had to glue them into a large “blank” for milling. This proved to be an interesting challenge. First, our pieces of foam depended on our sponsor’s available inventory: they were not uniform in dimensions or density. We played “Tetris” in our design files to form the blanks from the foam we had available.
We also had to consider the strength of the foam patterns, so that the assembly would not split when forklifted for transport. To ensure this, we staggered the foam pieces whenever possible: this alternated the location of the glue lines, which were likely to be weak points in the structure. We designed a wooden base made from reinforced, extra-long pallets. This added flexural stiffness, while making the structure easier to forklift.
Finally, it was time to enact our CAD designs. We had never glued tooling foam on such a large scale before, so we ran into some unanticipated difficulties in choosing the right adhesive, applying pressure as the glue cured, and finding enough team members to complete the blanks on time.
Our adhesive had to be strong enough to hold the foam together when it was fork-lifted, and easily millable to leave a smooth bond line. It also had to cure without requiring exposure to air: the closed-cell tooling foam would seal off the glue. Finally, it had to cure reasonably fast, so we could add new pieces to the structure in a timely manner.
Our first choice of adhesive was also manufactured by DUNA: DUNAPOL foaming adhesive was strong and fast-curing, and would also foam up to fill the gap between blocks. Unfortunately, we found the foaming to be more than we could handle: it applied more pressure than we could counteract with weights or clamps. Vacuum-bagging the glued foam pieces ensured a great bond line, but would take too long to enact for our whole molds.
We heard of other teams using Gorilla Glue with great success, but decided to try Loctite construction adhesive as a cheaper alternative. This is what we used to glue the majority of our foam blocks together. Spritzing the glue with water ensured that it would react to form a strong bond. However, the glue did not expand much to fill gaps in the foam: we decided to patch these after milling, and forged onwards.
Applying pressure was our next major challenge. We needed to force the foam pieces together while the glue cured, in order to form a strong bond. This was easier said than done: the whole blank was too wide for off-the shelf clamps, and we found vacuum bagging to be much too slow. We found that ratchet straps worked for applying pressure, but they were harder to use than clamps. Eventually, we settled on a two-step process: use clamps and weights for smaller sub-assemblies of 2-3 blocks, then use ratchet straps to hold the sub-groups into the completed blank.
Finally, we needed a lot of personnel to complete the foam gluing. The high-density foam blocks required multiple members to lift, and the adhesive curing time meant that we could not complete the blank in one sitting. We needed to field a large working crew, day after day.
Fortunately, we could leverage our large team of dedicated members. Our composites team was not able to perform all of the work, so we set up a comprehensive rotation system to share the load between different subteams. This prevented fatigue for individual members as we sent out full cars of members on every day of the week. Thank you to everyone who contributed to this titanic effort!
After weeks of hard work, the completed foam blanks were ready for milling and finishing. Check out Part 3 of this series in the not-too-distant future!
What do all of these have in common? One answer is: they can all be made from carbon fiber. Carbon fiber reinforced polymer composites are strong, stiff, and lightweight, making them ideal for aerospace and other high-performance applications. In addition to these properties, carbon fiber composites can also be molded into complex shapes. This is both a blessing and a curse: while this capability makes it perfect for the compound curves of our solar cars, it also necessitates using a mold. In this series of blog posts, we will walk through the process of designing and manufacturing molds for our 9th-generation solar car, Tachyon.
When we began the process, we already had a simple understanding of molds: the mold needed to match the outside shape of the final vehicle, because the molded carbon fiber composite would take on both its geometry and surface finish. The carbon fiber will be placed on the inside of the mold in order to achieve a smooth exterior for better aerodynamics. A mold like this is known as a “negative” or “female” mold, in contrast to a “positive” or “male” where the molded part will be smooth on the inside.
Furthermore, the mold needs to be removable once our car has been molded. For example, a ball cannot be molded with a single exterior mold: there would be no way to get it out once it’s done! Instead, the mold needs to be split into multiple pieces based on draft angle, which is the difference in angle between a mold surface and the vertical plane. The draft angle cannot be negative, or the part will stick inside the mold. A 2 degree positive draft angle will ensure release in most cases.
Similarly, we had to split Tachyon’s molds into multiple parts in order to get the exact shape that we needed. There are many ways to cut the division lines, and we were lacking in experience: almost all the members who made our last set of molds had already graduated. We started by imitating our previous car’s molds, and divided Tachyon in half horizontally through the widest point.
This gave us mixed results. It worked well for previous vehicles, whose main shells were modelled after wings. They had a defined “widest point” and relatively simple geometry. However, our first-ever cruiser class vehicle had a shape closer to that of a family car. This division line would have been difficult to designate on the mostly-vertical sides of the car, and impossible to implement on the complex nose curvature. Furthermore, the line would have run right through the doors of our car, making each door half difficult to align and join.
Fortunately, we were able to benefit from outside expertise. Goldshield Fiberglass Inc. advised us extensively throughout our design process, and recommended splitting the shell into quarters.
This resolved our problem with draft angles in the nose, and also moved the cut lines to avoid most of our doors and panels. A beneficial side-effect was the reduction in width for each individual mold: this would make it much easier for us to reach the center of the mold when sanding or polishing it. Goldshield Fiberglass also showed us how to account for runout, to allow room for trimming of the molded part. Finally, they suggested a variety of ways for precisely locating parts within a mold, and for aligning adjacent molds in a multi-mold setup.
We would like to thank Mike, Rick, and the rest of Goldshield Fiberglass Inc. for all of their help with our molds design! Their emails and video calls taught us a lot, and greatly improved our mold designs. Goldshield’s logo is proudly displayed on our website as a token of our appreciation for their sponsorship, and we recommend heading over to www.goldshield.com for more information on their custom molded composites.
With our division geometry finished, we added flanges to the sides of the molds to provide space for vacuum tape and other vacuum bagging materials. Stay tuned for our next post to see how we chose materials for our molds and manufactured our mold patterns!
Throughout the history of engineering, people have always been testing things out; whether they’re a caveman clubbing a stick against a wall to see how stiff it is, or an automotive engineer testing the aerodynamics of a sports car in a wind tunnel. However, testing can be cost-prohibitive; industrial-sized wind tunnels cost hundreds of thousands of dollars, not including the models that need to be built for testing. So how does CalSol get the crucial physical and aerodynamic data needed to build safe and efficient solar vehicles? Our answers lie in ANSYS.
ANSYS includes several types of engineering simulation software. Its ability to integrate and analyze models from a multitude of design software such as SolidWorks allows us to examine how our experimental parts would perform on the road, without requiring us to make massive financial investments into physically testing every part ourselves.
Thanks to the generous support of ANSYS Inc., we have been granted full licenses to access their wide variety of analysis software. During the structural design of our latest vehicle, Tachyon, we decided to make the switch from a space-frame chassis to a monocoque design, where some or all of the load on the vehicle is supported by an external “shell.” This shell will be composed of crisscrossed layers of carbon fiber prepreg and aluminum honeycomb core, which are all lightweight and strong materials, but more difficult to analyze. Such a drastic change in design philosophy necessitated rigorous testing and analysis.
Dylan Callaway, a junior in Mechanical Engineering and our monocoque design lead, utilized ANSYS Composite Prep/Post (ACP) to design and verify the structural integrity of our vehicle. Whether we’re optimizing the aerodynamics of the newly designed top shell, or building shell panels for structural strength, our members can rely on ANSYS to provide us with the crucial analysis for a safe and streamlined shell.
For metal parts, we use ANSYS Structural to find how well they would hold under load. Using the variety of the FEA tools within ANSYS, CalSol members like Roger Isied, one of our chassis specialists, are able to conduct a wide range of crash deformation simulations without the pain and cost of building several physical models and testing them one by one.
Now, the fact that we’re able to analyze structural integrity with the help of the software from ANSYS is a massive leap compared to years past. But that’s not all: with ANSYS Fluent, their computational fluid dynamics (CFD) offering, we are able to simulate the aerodynamics of our vehicle. This allows us to optimize the efficiency of our solar vehicles, letting them drive with the lowest drag and resulting power consumption, and giving us a competitive edge.
Airflow modeling isn’t just used for our outer shell’s airflow. We are also able to model the airflow for other parts, such as tire rims, or cooling vents for both our battery packs and our driver. This modeling allows us to place our vents in the best positions to make sure we have enough air going into those critical areas while keeping drag low.
Without ANSYS, getting all this important technical data, whether material strength or ideal aerodynamics, would be grueling. We wish to express our gratitude to ANSYS Inc. for providing us with this valuable resource, and we look forward to a productive partnership for years to come.
CalSol uses carbon fiber composites for many components on our solar vehicles, but we work primarily with rolls of woven carbon fiber or carbon fiber prepregnated with epoxy. Last Thursday, we finally got a chance to see how raw strands of carbon fiber are woven into the fabrics we use for our vehicle.
We started the morning by touring Sigmatex High Technology Fabrics Inc., in Benicia, California. After checking at the front desk, we went on a guided tour of the weaving facilities. It was fascinating to see individual spools of carbon fiber woven, trimmed, and packaged into ready-to-ship boxes. The amount of attention to detail and precision in the cleanroom was inspiring, and we hope to improve on our own processes as we build Tachyon’s composites components.
Immediately afterwards, we embarked on a tour of Patz Materials and Technologies. Joseph Talosig gave us a wonderful tour of the Patz facilities and showed us their honeycomb core and prepreg lines. He was also happy to answer our questions about prepreg manufacturing and materials selection, and even donated several sample rolls of prepreg to us for making test panels and practice layups. Thank you to Patz Materials and Technologies and Joseph for your generosity and the great tour! We look forward to working together for the manufacture of Tachyon’s shell and monocoque.
On Saturday, CalSol had the opportunity to bring Zephyr to Cal’s homecoming game vs Arizona. We had a lovely spot in the Blue and Gold zone where Cal fans of all ages got to learn about the car and the benefits of renewable energy. A huge thanks to Cal Athletics for inviting us!
On Sunday, we displayed Zephyr at the East Bay Mini Maker Faire. While we have attended several Maker Faires in the past, this time we were positioned right at the main entrance. Thank you to East Bay Mini Maker Faire for this great opportunity!
We started our day before the sun had risen with Kingpin donuts and groggy eyes at Hearst Mining Circle, before picking up our trailer and heading out to Oakland Tech High School. When we wheeled Zephyr in, we were amazed to find a chair that walked on eight legs, an amphibious bike, and tiny “robot pod” houses that turned on their sides to let you sleep. There were many, many people; everyone from retired professors to parents to elementary school students came to see what local Makers produced. Best of all, everyone was so curious and genuinely interested in what all the Makers had to present. The amount that people wanted to learn, especially the younger ones, was great to see.
In fact, the kids were one of the greatest parts of the Mini Maker Faire. Many of them got excited and loved to ask how fast we could go, or laughed when we honked the horn. Some of them already knew about solar energy and panels, or even some principles of aerodynamics. We’re excited to go back next year!
Although the school year has just begun, CalSol has already dived headlong into what promises to be an exciting build cycle. We began this past weekend by hosting a joint kickoff and showcase event with Berkeley’s Formula SAE competition team on Friday night.
Prospective members for both teams were treated to a BBQ at our workspaces at the Richmond Field Station, then an exhibition from both teams of the vehicles’ capabilities. Finally, everyone split into groups for a comprehensive tour of both teams’ facilities and subteams. Attendance was high despite a chilly breeze, and we all enjoyed seeing so many enthusiastic new faces.
General Motors, one of our largest sponsors, visited the Field Station the next morning. They treated us to pizza and ice cream, and we got a chance to test-drive a wide selection of GM vehicles. Some of them were capable of blisteringly fast speeds, though we took care to obey RFS speed limits. Still, the acceleration was something to behold.
We offered rides in our solar vehicles as well. However, not everyone was able to fit…
Ford Motor Company, also one of our largest sponsors, visited us on Sunday and brightened our day with burritos. Though there was no test drive this year, experienced Ford employees helped us out by reviewing some of our critical mechanical designs.
We are very thankful for GM and Ford’s continued support, and it was truly a pleasure conversing with both of them this weekend. Many of our alumni are working at both companies, and a lot of us jumped at the chance to catch up with them as well.
This weekend also saw the departure of a long-time fixture in our composites building: Zephyr’s fiberglass molds have left for new horizons! These molds were used to create Zephyr’s entire aerodynamic shell, but are no longer usable for Tachyon. Luckily, the New Jersey Institute of Technology Solar Car Team is happy to use them for their next vehicle.
This transfer was a long time in the planning. We began discussing this possibility with the NJIT team prior to this summer’s Formula Sun Grand Prix, and Jefferson Guerrero from NJIT was able to fly over this weekend to oversee the move.
Zephyr’s molds have served us long and well, and we are confident that the NJIT Solar Car Team will put them to good use. We’ll miss their vivid green presence, but look forward to making use of all the extra space in our composites facility.
Finally, we had a photoshoot with Berkeley Engineer Magazine. They brought out a lot of equipment, and we had fun trying different poses for the camera. Thank you for the opportunity, Berkeley Engineer!!
Zephyr is ready to ship after a year of hard work, and departed for Texas this afternoon! She will arrive at Circuit of the Americas in Texas on Sunday July 2nd. While Zephyr still needs a few final adjustments after her multi-day journey, we are confident she will be ready for Scrutineering on Monday.
The past few weeks have been a flurry of preparation and testing, and many teammates have been putting in long hours at our workshop at the Richmond Field Station. We had time to hold dynamics tests this week to gauge our preparedness for scrutineering, and took careful care to replicate conditions at the race.
Finally, we would like to make a shout-out to everyone who worked so diligently on making Zephyr Rayce-ready this year! Thank you as well to all of our loyal sponsors: without your help none of this would be possible. We are excited to see the fruit of our labors at Formula Sun Grand Prix 2017, and hope to make this Zephyr’s best Rayce yet!
We are excited to announce Mun Manufacturing Co.’s sponsorship for Tachyon! Mun Manufacturing is a machine shop in Oakland, California, and is run by dedicated owners Dennis and Jeanne Lee. They machined the tooling foam plug for our practice mold this summer, and we look forward to a very rewarding and productive partnership with Mun Manufacturing Co.
Dennis Lee is also a Principal Lab Mechnician at UC Berkeley’s Mechanical Engineering Student Machine Shop, and is always ready to help us and other students with machining and advice. He has been essential for the success of many student projects and teams. We have benefited greatly from Mun and Dennis’s generosity, and can’t thank them enough for their help!
Mun Manufacturing Co.’s logo is proudly displayed our website, and will race with Tachyon in American Solar Challenge 2018, World Solar Challenge 2019, and beyond. We encourage you to contact them at email@example.com for more information on their services.