Reducing cost of an EVVT system while improving reliability and extending the working lifetime

Magnesense LLC is a research and development company incorporated in 2000 in Maine. The company specializes in electromagnetic actuators and related control technology.
Since its founding, the company has focused on electromagnetic engine valves for spark ignition engines, for both automotive applications and small engines.

“As a small company, we have sought to identify weaknesses in the technology and develop solutions that we can market to industry. Some of our early work was supported by the Maine Technology Institute, while more recent work was supported by the U.S. Environmental Protection Agency,” says a company statement.

Magnesense has designed and bench-tested automotive-scale and small-engine valve actuators, seeking improvements in performance, economy, and low retooling. The company has a total of 10 patents, including key sensorless control patents. Magnesense is particularly excited about a new engine valve application called Electromagnetic Variable Valve Timing or EVVT, which it says opens a new frontier for improving internal combustion engines.

This is accomplished by doing away with the traditional timing belt and camshaft and exerting full computer control over valve timing. Why?

First and foremost, the throttle goes away, along with the breathing losses caused by the throttle. The computer-controlled intake valve “waits” until there is just the desired charge of air in the cylinder, then snaps shut, capturing the charge for compression, fuel injection, and combustion.

Second, in a “full camless” engine, exhaust valve timing is used for controllable internal EGR. Future developments in HCCI and related variations on compression ignition will probably rely on tight cycle-to-cycle control of the combination of intake air, internally-recirculated hot exhaust gas, traditional externally-cooled EGR, and fuel injection. Until highly-controllable, inexpensive VVT systems arrive, HCCI technology will remain stuck in the laboratory. In the near term, multi-fuel automotive systems will rely on the flexibility of EVVT control. While EVVT is widely believed to be far too costly for small engines, Magnesense is developing inexpensive systems to meet the tightening regulatory emission limits for small engines.

The basic EVVT actuator design has changed little since the first patent in the early 1900s. Recently, several R&D teams have built EVVT systems and demonstrated their ability to improve gas mileage, improve performance, and reduce emissions. “So, why aren’t these systems in cars and trucks on the highway?” Magnesense asked this question, identified the major impediments, and now offers solutions.

The greatest problem is hard landing. Electromagnets have a natural tendency to snap together, which is very difficult to overcome with servo-control. Magnetic hysteresis exacerbates the servo-control problem. Furthermore, as noisy EVVT actuators pound away, that pounding work-hardens the magnetic materials, further increasing the hysteresis and precipitating a regenerative spiral to failure. The solution, covered by multiple patents, involves getting “behind” the current-flux hysteresis loop and taking direct servo-control of magnetic flux. The control system then becomes much more manageable and responsive. Carrying this idea further, Magnesense is continuing to refine its techniques to determine both magnetic flux and armature position indirectly, from the current and voltage going out through the power windings without requiring separate sensors or sensor wiring.

Another problem is space: an EVVT actuator has to fit the narrow spacing of engine valves. Additionally, the actuator needs powerful centering springs to help kick the valve rapidly between open and close positions. The demands for narrow spacing and powerful springs has led to vertical growth of the actuators, leading either to a high profile or to a design with large springs that must be set deep into the cylinder head. Magnesense has developed an inexpensive push-pull spring that requires no grinding, minimizes moving mass for fast actuation, and fits compactly into the narrow profile of the actuator. The cylinder head retains a “traditional” valve return spring and a design that is hardly affected by the transition to camless operation.

Magnesense argues that by combining the above innovations with simplified power electronics and efficient computer code that runs on a cheap DSP processor, they can cut significantly from the overall cost of an EVVT system while improving reliability and extending the working lifetime.

Based on extensive road testing, other companies have documented EVVT improvements of roughly 20% in multiple categories: gas mileage, low-end torque, peak horsepower, and emissions of all types. Magnesense intends to lead the way to a generation of actuators that work still better, last longer, are cost-competitive with mechanical variable valve timing (VVT), and are more effective than their mechanical counterparts.

Joseph B Seale founded Magnesense in 2000 with partner Gary Bergstrom. Seale is first-named inventor on 15 U.S. patents, which reveal pioneering work in medical ultrasound, physiological measurements, fluid pumping and flow control, magnetic levitation, and the design and control of solenoids. Deep understanding in these diverse areas, combined with an ability to construct computer models that bring disparate areas together, has resulted in multiple insights and inventions in the area of electromagnetic actuators making optimum use of ferromagnetic materials under sophisticated nonlinear control.

Gary, who is named in 13 patents, develops analog and digital hardware and control software. He has designed products in the medical and industrial test/measurement fields. Sometimes redesigning old products with modern technology and sometimes developing completely new products, the goal has always been a cost effective solution that exceeds the competitions’ devices, often for years to come. For example, he designed an audio voltage controlled amplifier (VCA) that remains among the best after 25 years. A device to test rolling tires is still in daily use, 15 years later.

Automotive Industries spoke with Joseph B Seale, one of the two founders of Magnesense.

AI: What kind of automotive OEM expertise does Magnesense have? Please tell us a little about the technologies your company has to offer automotive companies.

Magnesense: Both Gary and I have spent our careers in product development, and Gary grew up with a dad working in the valve division of TRW. As far as our professional work, however, we are relative newcomers to the automotive field. We think we bring the advantage of “outsiders.” We don’t know “How It’s Supposed To Be Done,” so we find new solutions. We both started out in physics and came to engineering from that angle. Gary, however, is a “practical guy” who works with machinists, cranks out custom circuit boards, has designed sensors for a living, and writes the kind of tight embedded code that is needed to do a big job with a cheap micro or DSP chip. I’m a “theory guy” who’s worked in thermodynamics, fluid dynamics, electromagnetics, and computer simulations from helicopters to medical infusion pumps. We’re very different and complementary.

AI: What are the innovations Magnesense has brought to EVVT?

Magnesense: I think the biggest single innovation is flux control. Everyone else, at least everyone we’ve seen or read about, tries to control the voltage or the current going into a solenoid – any solenoid. Those approaches put all sorts of nasty nonlinear goings-on between the controller and the final result: the motion they’re trying to control. We figured out that it’s relatively easy to keep track of magnetic flux, and then to control it, without having to rely on separate sense coils or Hall-effect devices or anything “extra.” When you know position and control flux, you control force. The control loop is much better behaved. If I can manage to teach the world about flux control, I’ll feel very satisfied.

A related innovation is sensorless determination of position. It’s all wrapped up in the flux control approach. If you know what kind of voltage pulses your valve controller is driving the solenoid with, and if you keep track of the current, then you can figure out both position and magnetic flux, without sensors. The software gets more complicated, but the hardware gets simpler, cheaper, and more reliable.

AI: How have you been marketing this technology to automotive companies and what has the reaction been?

Magnesense: We have made presentations in front of engineering groups at several companies. The engineers have been enthusiastic, but the executives have been more reserved – and for understandable reasons. Innovation is hazardous. If an automaker introduces a technology that has only been around in a few test vehicles for a few years, and five years down the ‘pike they start getting recalls, that can be a killer. We’re working really hard to respond to these concerns by addressing the problems behind the uncertainties and unreliabilities of EVVT actuators. Simplify, simplify, simplify. As you simplify a system, you reduce the number of unknowns. But sometimes to simplify you have to work very hard at understanding what’s really going on.

AI: Do you think EVVT will play an important role only with hybrids or with regular vehicles as well?

Magnesense: The advantage of EVVT is that your engine adapts much better to a wide range of operating conditions. The idea with a hybrid is to operate the engine in a much narrower range of operating conditions and let the electrical system take up the slack, giving you a boost for accelerating and passing, taking over completely when power demand is low, and generally leveling the load on the engine or shutting the engine down. We anticipate that a non-hybrid car with EVVT can be cheaper than a hybrid and can outperform the hybrid on the highway and anywhere except in stop-and-go traffic. So, the advantage of EVVT depends on what kind of driving you do. Now if you add EVVT to hybrid, you make the hybrid even better, but the incremental gain isn’t as spectacular, because there’s some overlap between what the two systems try to accomplish. In the near future, the biggest benefit to EVVT will be with your “regular vehicles.” In the long run, EVVT is going to hook up with HCCI, and hybrid technology will complement that combo. After that, we’ll get to see whether hydrogen technology ever becomes “real” sometime off in the future. I can’t look that deep into the crystal ball. In the future I can foresee, EVVT is going to play a big role.

AI: What have been some of the challenges faced by your company in making your technological breakthroughs known to auto makers? And how do you hope to overcome them?

Magnesense: The problem is credibility. We’re tiny. We’re two people. We’re outsiders. And, there are a bunch of groups with new engine designs, pounding on the doors of industry. We can’t create EVVT technology all by ourselves, from concept to engines running in cars, all coordinated with the software that runs the whole car. Fortunately, we don’t need to do all that. EVVT is going to happen. We decided a long time ago that we could help it happen, and now we have mastered the technologies that we originally identified as critical. So now, we have to focus on communicating what we know and who we are. We’re inventors, and the challenge right now is to do something we aren’t used to: marketing ourselves.