Jeżeli do oprogramowania inżynierskiego można podpiąć własne procedury to NVIDIA udostępnia darmowy driver NVCC za pomocą którego można kompilować kod napisany w C. Można to robić z Visual Studio (od wersji 10) pod Windows lub używając gcc pod Linuxem.
Wówczas oprogramowanie optymalizuje się pod posiadaną kartę graficzną. Warto mieć kartę o CUDA capability większym niż 2.0 ze względu na obsługę typu double.Zobacz Komentarze Newsów
Czy ktoś z forum korzysta może z Elmera? Bardzo ciekawi mnie jak to działa. Bo z tego tutka wynika, że jest OK, a tymczasem ja próbowałem... bezskutecznieZobacz Komentarze Newsów
Postów: 651 Miejscowość: Warszawa, PL Data rejestracji: 18.10.07
Dodane dnia 02/22/2008 15:18
FEA and The World's Fastest Motorcycle
"After 17 years, we finally did it," said Joe Harralson of Sierra Design Engineering in Mount Aukum, California, reflecting on his work with BUB Enterprises of Grass Valley, California to design and build the world's fastest motorcycle. On September 5, 2006, during the 2006 International Motorcycle Speed Trials at the Bonneville Salt Flats near Wendover, Utah, Harralson watched as the BUB team's streamliner, named "Seven", (which looks more like a rocket on two wheels than a traditional street bike) was driven by multi-time American Motorcyclist Association champion Chris Carr to a new world record speed of 350.884 mph. For Harralson and the BUB team, it was the culmination of a decades-long odyssey that included budgetary battles, daunting technological challenges and intense competition in relentless pursuit of a speed record that had stood for over 15 years.
Harralson used finite element analysis (FEA) software from ALGOR, Inc. of Pittsburgh, Pennsylvania to analyze several key components of the streamliner including the engine crankshaft, frame, rear suspension swing arm and wheels. "ALGOR FEA was an invaluable tool in this project," said Harralson. "At almost every step, we were in unexplored territory, and the ability to analyze stress and deflection was essential."
Chasing the Motorcycle Speed Record
In 1989, Harralson attended a meeting of the Society of Automotive Engineers (SAE) in Berkeley, California, where the guest speaker was Denis Manning, owner of BUB Enterprises, a leading manufacturer of motorcycle exhaust systems. Because Harralson was a motorcycle enthusiast, he already knew about Manning's previous speed-racing achievements. (In 1970, a Manning-built streamliner set the world record, which was featured in the documentary film "On Any Sunday".) At the SAE meeting, Manning talked about preparing for yet another attempt on the speed record for motorcycles, and, this time, he wanted to build his own engine. Harralson explained, "Up to that time, all motorcycle speed record attempts were either factory-based efforts or used modified production motorcycle engines."
BUB Team's streamliner "Seven" driven by Chris Carr
BUB Team's streamliner "Seven" driven by Chris Carr
Harralson, who was a professor of mechanical engineering at California State University Sacramento and previously had worked for Mercury Marine and McCulloch Corporation as a design engineer for engines, was so inspired by Manning's speech that he drew up some rough layouts for Manning's proposed streamliner engine. "Denis had said the engine would be a 3 liter V4 with the crank cross ways to the frame," said Harralson. "I had done some preliminary work on a small V8, and I thought it might be adapted to what Denis wanted." Harralson then called upon Manning and showed the drawings. "He was very excited," remembered Harralson. "Denis likes to say that he saw on paper what he had only seen before in his dreams."
Analyzing the Engine
In spring 1989, Harralson began designing the engine and transmission with Manning designing and building the rest of the streamliner. "Seventeen years ago, PC-based CAD was just getting started," recalled Harralson, "and the capabilities of FEA software were rather limited by today's standards. In spite of those limitations, ALGOR FEA was an essential design tool for the engine." Many of the major engine components were analyzed including the crankshaft. "The crankshaft was a particularly critical component because the engine is a 90-degree V4 and has an uneven firing order, which produces strong torsional loads," explained Harralson. "The value of ALGOR FEA has been demonstrated by the fact that there has never been a mechanical failure of the engine."
Harralson's custom-designed engine produced over 420 horsepower naturally aspirated; however, the speed record proved elusive. "After many years of part-time work, computer simulations and testing, it became apparent that the streamliner was too heavy and had several aerodynamic problems," said Harralson.
He took on these new challenges by again relying upon ALGOR FEA. "Work on the frame, swing arm and wheels took advantage of ALGOR's CAD integration with SolidWorks models," he said. "All of these parts were first modeled in SolidWorks and then captured directly in ALGOR's FEMPRO interface for analysis. The time savings and ease of modeling compared to the original work done on the crankshaft was truly remarkable."
Analyzing the Frame
A completely new streamliner body was designed, partly inspired by the shape of a salmon. The new body was fabricated from Kevlar® and carbon fiber in a sandwich construction using Nomex® honeycomb as the core. "Up to that time, land speed record motorcycles had used a tubular steel frame with a separate body, typically fiberglass," said Harralson. "In this case, the body became the frame in a manner similar to modern Formula One race cars. ALGOR FEA played a key role in the development of the frame."
ALGOR FEA of BUB streamliner frame.
FEA of BUB streamliner frame.
Harralson performed comparative analyses between the old steel tube frame and the new composite frame. "One area of particular interest was the torsional stiffness of the two frames," said Harralson. "Streamliners frequently experience handling problems at speed, either weaving or violent shaking of the front suspension, which we believed were due to torsional oscillations of the frame. Once an ALGOR model was made for each frame, the torsional stiffness was determined, and the new carbon fiber design was found to be much stiffer than the original steel tube design although significantly lighter. Both of the actual frames were tested, and the measured torsional stiffness was within 5 percent of the predicted value in both cases."
Analyzing the Rear Suspension Swing Arm
Similar modeling and testing was done on the rear suspension swing arm. In this case, ALGOR FEA helped to identify a design problem. "The original design was a welded steel piece, while the new design was machined from 6061-T6 aluminum," said Harralson. "The ALGOR model was far stiffer than the measured values for the aluminum swing arm, a very disappointing result. However, further investigation revealed that there was a problem with the actual swing arm. In the computer model, the two sides are joined by a large cross piece, which was assumed to be solidly attached. In the actual swing arm, this cross piece was simply bolted in place by a fairly small number of bolts. It became clear that this bolted connection was not stiff enough and was the reason for the low stiffness of the actual part. This also explained bolts that had come loose and, in one case, broke a33; problems that were originally attributed to failing to tighten the bolts correctly. Once the bolted joint was redesigned, the swing arm stiffness agreed closely with the computer model, and no further problems were experienced."
Analyzing the Wheels
ALGOR software was also used to analyze the wheels for the streamliner. "Since the wheels are turning at over 3,500 rpm, centrifugal stresses were a real concern," said Harralson. "Models were run with radial and side loading as well as spinning."
BUB streamliner on the Bonneville salt flats. ALGOR FEA of rear suspension arm and wheel.
BUB streamliner on the Bonneville salt flats.
The proof of Harralson's analyses was evident during the record-breaking runs at Utah's Bonneville Salt Flats: "othing broke and everything worked," he said.
"The team has big plans for the future, including a run on our own record next summer," said Harralson. "Our analysis shows that 400 mph is not out of the question, although there are many hurdles to get over. A V-twin engine is in the works, and, although it will be a street-going bike, there is little doubt that it will also be seen on the Bonneville Salt Flats. ALGOR FEA will continue to play a part in these and future projects."
Joe Harralson earned a Mechanical Engineering degree from California State University Sacramento (CSUS) and a master's degree in Mechanical Engineering from Loyola Marymount University. He retired as a tenured professor of Mechanical Engineering at CSUS and now teaches part-time. Harralson works on high-performance engines through his consulting firm, Sierra Design Engineering in Mount Aukum, California.
Pozdrawiam,
Admin
We live in an age when pizza gets to your home faster than police or an ambulance
Skocz do Forum:
Reklama
Szukaj
Logowanie
Nie jesteś jeszcze naszym Użytkownikiem? Kilknij TUTAJ żeby się zarejestrować.
Tylko zalogowani mogą dodawać posty w shoutboksie.
Bulix 26.01.2022 Witam, szukam osoby która ogarnia program FEMM.
damian14100 25.01.2022 Witam,
Czy znajduje się na forum osoba która ma doświadczenie w obliczeniach wytrzymałościowych wózków wagonowych ?
BE-FEA 31.03.2021 Nie pracuję w Deform, ale może coś podpowiem na zasadzie analogii do innych programów MES. Napisz jeśli nadal aktualne.
daniel8894 31.03.2021 Czy pomógłby ktoś w zrobieniu symulacji procesu skrawania w Deform 3 d
BE-FEA 09.03.2021 Postaram się pomóc, zapraszam do kontaktu jeśli nadal aktualne.
kinia22 09.03.2021 witam, potrzebuje kogoś kto pomoże w przygotowaniu projektu w programie COMSOL - przeplyw ciepła przez ośrodek porowaty!
BE-FEA 02.04.2020 Przerobiło mi jedną literę na emotkę. Chodzi o przycisk new thread
BE-FEA 02.04.2020 Żeby dodać nowy temat wystarczy wejść w odpowiedni dział i kliknąć mały przycisk "ew thread" po prawej stronie.
KrzywaOHIO 30.03.2020 Jak dodać nowy wątek na forum?
Konrad96 02.12.2019 Witam, czy jest tu ktoś kto dobrze zna się na programie Robot struktural i pomógłby mi zamodelować belkę żelbetową podobną do belki użytej w badaniu? Proszę o kontakt jeśli ktoś chciałby pomóc
Użytkownicy Online
othing broke and everything worked," he said.
