Delphi invests in Tula; Dynamic Skip Fire cylinder deactivation
30 July 2015
Delphi has made a minority investment in Tula Technology, the developer of Dynamic Skip Fire cylinder deactivation technology. (Earlier post.) The software-enabled powertrain technology integrates advanced digital signal processing with advanced powertrain controls to create a variable displacement engine.
DSF allows the engine’s cylinders to fire or skip (deactivate) on a continuously variable basis. In independent tests, DSF has improved fuel efficiency by up to 17% as measured on a CAFE basis when compared to a vehicle (V8 engine) that does not have cylinder deactivation. GM Ventures made an equity investment in Tula in 2012.
|
|
| Comparison of DSF fuel economy gains and costs with competing technologies. Wilcutts et al. (2013) Click to enlarge. |
The significance of this new relationship goes beyond the investment itself. Delphi’s strong expertise in engine management systems and valvetrain components will help us to further optimize the overall performance of DSF and accelerate the deployment of our unique technology for three, four, six and eight-cylinder engines.
—R. Scott Bailey, president and CEO, Tula Technology
Tula’s control approach integrates advanced digital signal processing, culled from consumer electronics technology, with powertrain controls. DSF technology does not rely on fixed cylinder deactivation or switching between fixed patterns. Rather, it incorporates any-time, any-cylinder deactivation and can continuously vary the number of cylinders firing, along with cylinder load.
In a 2013 SAE paper, the Tula engineers described two primary benefits from dynamic skip fire technology:
Fuel economy improvement via removal of pumping losses and optimization of combustion. For each engine speed, a sweet-spot of thermal efficiency for operating cylinders exists which is at high load for throttled engines due to minimized pumping loss. With combustion occurring preferentially in this regime, the engine combustion system can potentially be optimized to match this operating area. With intake or exhaust (or both) valve deactivation in place, inactive cylinders are prevented from pumping air through the engine, which enables effective use of three-way catalyst technology.
Wide authority over generation of vibrational and acoustic excitations. In normal throttled operation of an engine, excitation spectra are tied to the engine speed, and magnitude is determined by the level of throttling. Operating the engine in a dynamic skip fire manner alters the torque excitations on the vehicle powertrain, which could lead to unacceptable noise, vibration and harshness (NVH) characteristics.
With dynamic skip fire technology, the spectra are additionally controlled by the number and sequence of firing cylinders. Full dynamic control of firings and non-firings of engine cylinders means that noise, vibration and harshness (NVH) can be dealt with algorithmically, in a flexible and systematic way.
|
|
| An example of a firing sequence that fires one cylinder followed by two skipped cylinders. The pattern will repeat after three engine cycles for an 8-cylinder engine. Serrano et al. (2014) |
Resources
Chien, L., Younkins, M., and Wilcutts, M. (2015) “Modeling and Simulation of Airflow Dynamics in a Dynamic Skip Fire Engine,” SAE Technical Paper 2015-01-1717 doi: 10.4271/2015-01-1717
Chen, S., Chien, L., Nagashima, M., Van Ess, J. et al. (2015) “Misfire Detection in a Dynamic Skip Fire Engine,” SAE Int. J. Engines 8(2):389-398 doi: 10.4271/2015-01-0210
Serrano, J., Routledge, G., Lo, N., Shost, M. et al. (2014)“Methods of Evaluating and Mitigating NVH when Operating an Engine in Dynamic Skip Fire,” SAE Int. J. Engines 7(3):1489-1501, doi: 10.4271/2014-01-1675
Wilcutts, M., Switkes, J., Shost, M., and Tripathi, A. (2013) “Design and Benefits of Dynamic Skip Fire Strategies for Cylinder Deactivated Engines,” SAE Int. J. Engines 6(1):278-288 doi: 10.4271/2013-01-0359
I had a very similar idea a long time ago.
I'd like to see a skip-fire module implemented as an aftermarket Piggyback ECM. A single module could have a selector for 4, 6, or 8 cylinder engine and a "ECO" button to activate. Ideally it would be programmed to implement the skip-fire based on time averaged power requirements only in Cruise Control On mode. In this way, the drive-by-wire throttle position could be managed as well. By activating only in Cruise Control mode, the skip-fire mode would automatically be cancelled when the brake is pressed and the throttle would be returned to normal position.
The fuel savings would be lower than the execution in the article due to no "air spring" in the deactivated cylinders.
However, the pumping losses would still be lower than normal operation of conventional engines because of the higher throttle openings utilized when the system is activated.
One of the largest obstacles to implementation would be managing the air-fuel ratio. The AFR would be much leaner than normal operation as seen by the oxygen sensor so an aftermarket wideband O2 sensor may be required as well. The piggyback module would need to have the capability to correct the wideband signal to the corresponding appropriate value and feed it to the OEM ECM. If this isn't done the active cylinders would be over-rich as the OEM engine control module tries to achieve stoichiometric AFR.
This extra lean exhaust stream would be a boon to turbocharged engines which could then tune for higher EGR rates and even lower pumping losses as power could be modulated with boost and fuel instead of throttle plate position at lower power requirements seen typically on the highway.
Cylinders seeing the higher loads would not overheat due to the skip-fire and in fact could actually run over-boost since the additional air pumping in deactivated cylinders would pre-cool them in turn. The speed at which the skip-fire rotates cylinders would be fast enough so that the thermal cycling wouldn't be so drastic as to increase material stress in the engine block.
Posted by: Trevor Carlson | 30 July 2015 at 01:48 PM
Back to the early 1980's?
Posted by: HarveyD | 31 July 2015 at 07:23 AM
The issues to implementing such a system in a carbureted vehicle from the 80's versus modern OBDII equipped cars with modern emissions systems intact and control modules connected on a CAN bus system would be completely different.
Besides I'm not interested in patenting the idea I summarized above but still it'd be neat to see a product come to market using it in some form.
Posted by: Trevor Carlson | 31 July 2015 at 10:50 AM
How well would it work with a 4 cylinder engine ?
Posted by: mahonj | 03 August 2015 at 04:09 PM
I can see this software embedded with others in the control module to address many issues. One is levelization of electricity production. If the battery can be relied on to produce spark plug acceleration rather than the alternator, there would be less strain and more optimal engine energy diversion to the accessories -- more fuel savings (GCC had an article on this development). Presumably would would have better on board diagnostics too. How would we tell a power loss from DSF operation?
And I for one am annoyed at that pesky Dynamic Steering Control light on the dashboard that ends up meaning nothing, other than that the old Toyota is about to turn off the DSC. So turn them both off and improve my mileagesome other way!
Posted by: kalendjay | 04 August 2015 at 03:17 PM