Heat to power conversion benchmark

heat2power does not pretend to have the only system that regenerates exhaust energy. Other systems exist already and several more are currently under development. Those systems have their merit but we believe that they do not offer the power density or the efficiency as the heat2power system does. Here follows a selection of known concepts that allow you to compare the heat2power system to these.

You will find below an overview of the technologies as they are known to us but you can also launch a search on the internet for additional information :

Electric turbo-compounding

The electric turbo-compounding as under development by Caterpillar uses the turbocharger shaft for extracting power from the exhaust gas. The power take off is obtained by a little generator that is located between turbine and compressor. The announced fuel consumption reduction is in the order of 3 to 5% on cycle and up to 10% for the peak values. Caterpillar considers this a concept with high customer value.

The electric turbo-compounding as under development by John Deere is slightly different. Here an additional turbine is placed in the exhaust and the energy obtained therein is transformed into electricity by means of a small generator. The announced fuel consumption reduction is in the order of 3 to 5% on cycle and up to 10% for the peak values. Caterpillar considers this a concept with high customer value.

In this project 20% power increase has been demonstrated with little adverse impact on the engine. Higher output seems to be possible. Fuel economy improvements of 10% have been demonstrated at Tier 3 Conditions
Turbo compounding is found to be compatible with emissions and appears to provide benefit. An electrically coupled two-stage architecture offers control, efficiency, emissions, and packaging benefits. It also supports electrification. System costs suggest commercialization potential.

Mechanical turbo-compounding

Mechanical turbo-compounding is a system that uses a second turbine in the exhaust to further obtain usable energy from the exhaust gas. The power is then transferred from the turbine shaft through a cascade to the crankshaft. The system is being used by Scania and Volvo trucks and reduces the fuel consumption by about 5 to 10%.

Scania Mechanical Turbo-compound

Download a benchmark of turbo-compound systems by from Bronislaw Sendyka and Jacek Soczowka. This report shows a graph for fuel economy for loads of 100% and 75% of about 5 to 11% in a narrow band around the optimal RPM.


Turbo-generator Integrated Gas Energy Recovery System

Visteon UK Ltd has created a device called TIGERS (Turbo-generator Integrated Gas Energy Recovery System). It uses the exhaust gas from a car's combustion engine to produce electricity, enough to power the vehicle's electrical system, thus taking a load off the engine; "Parasitic losses from mechanical support systems (i.e., belt-driven) can normally be as high as 6kW or 8hp in a family sedan but can be significantly higher in larger capacity cars and trucks. Moving from those mechanical systems to electrical removes those loses, and fuel consumption could be reduced from between 5%-10%." This technology could be coupled with hybrid technology and the extra electricity could help recharge the hybrid's batteries. "Because the system is fairly simple and partly based on existing technology, it could be fully developed for all car, van, bus and truck engines within a few years."

The small turbo-generator is installed in a by-pass waste pipe fitted just below the engine exhaust manifold. A valve linked to the engine’s management control system allows some of the high-energy exhaust gases to pass through a turbine to drive the generator, depending on engine load conditions.

Typically the 800º C gases have a velocity of 60m/s and a mass flow rate of 0.05 kg/s, providing enough energy to spin the generator at up to 80,000 rpm and create electrical power of up to 6kW—sufficient to handle the car’s electrical systems.

An energy management system will ensure optimal utilization of the available energy. During highway driving, when the available exhaust energy is high, the energy will be captured and the excess power will be stored in a battery. However, at engine idle the penalty for recovery is high and so the vehicle will be operated in battery only mode.

The researchers have looked at placing the new generator at various locations along the exhaust system. Placed too far away from the engine, the waste gases start to lose energy, so in the development stage the generator has been placed just beneath the exhaust manifold to maximize energy recovered. The gases then pass through the catalytic converter after the turbine, to ensure that the gases can still be conventionally cleaned.

By placing it close to the manifold the energy available is optimized. This also allows for shorter runs for control leads and coolant pipes and provides greater protection to the unit. Disadvantages are that the high temperatures mean the generator has to be water-cooled and totally sealed. However, the researchers are convinced that they can fully develop the system.

Because the system is fairly simple and partly based on existing technology, it could be fully developed for all car, van, bus and truck engines within a few years.

The simple design of the switched reluctance generator enables a low cost and easy to manufacture unit to be built that can run reliably at high speeds. It gives the TIGERS device a power density of approximately three times that of a typical alternator. An efficiency in excess of 80% can be achieved compared with 60% for traditional technology.

Dr Richard Quinn, one of the engineers leading the TIGERS project, says the system could be developed to produce anything from 12v to 600v.

The recovered energy could power all of a car’s heating, lighting, air conditioning and in-car entertainment systems. Longer term, the cam belt, drive belts and alternator could be scrapped with the TIGERS-recovered power providing electrical drive instead for further potential for gains in engine efficiency.

The additional electrical energy could be used to power more advanced engine technologies, such as electro-magnetic valve actuation, electric intake charge cooling, electric-powered super-charging or electrical exhaust after-treatment.

Parasitic losses from mechanical support systems (i.e., belt-driven) can normally be as high as 6kW or 8hp in a family sedan but can be significantly higher in larger capacity cars and trucks. Moving from those mechanical systems to electrical removes those loses, and fuel consumption could be reduced from between 5%–10%.

In a hybrid electric car the TIGERS system could feed the extra power directly to the drive motors or back to the battery to increase the range of the vehicle.

On commercial vehicles the extra electricity could be used to power electrical systems to run refrigeration units for chilled food, turn the motors on cement mixers or power pumps on fuel tankers.

The TIGERS group comprises researchers from Visteon UK Ltd in Coventry, Switched Reluctance Drives of Harrogate and The University of Sheffield Electrical Machines & Drives Research Group.

Switched Reluctance Drives is a leader in switched reluctance technology and is developing a high-speed generator to work in this demanding environment. The University of Sheffield research group is applying its knowledge of electrical system modelling and design to optimise the control and energy storage system.

Visteon is the lead partner in the project and is one of the world’s leading Tier 1 automotive suppliers. It is responsible for the system design, testing and implementation.


Two different materials in a heat flow using a temperature difference can generate electricity. This effect was first quantified by Seebeck. The following device shows the typical layout to obtain this.

Efficiencies of conversion depend from difference of temperature and characteristic of used materials. Their relationship in shown in the following formula.

The focus is then to obtain high efficiency by getting the highest possible value for Z as well as a high temperature difference. The following diagram shows the characteristics of ZT for several materials.

With known conventional solids, a limit to Z is rapidly obtained. However, new materials, recently developed show a breakthrough in performance :

Filled Skutterudite as material has shown big progress over previous materials but newer materials come up regularly. The development has accelerated performance considerably over the last two years.

The thermo-electric developments currently focus on finding better materials to convert the heatflow into electricity. The costs per Watt are still very high. In 2005 this was at about 1$ per Watt with Bi2Te3. The latest materials such as Si/SiGe on Kapton Substrate by Quantum Wells are at 0,10$ per Watt. So the cost for a 30kW unit, interesting for larger trucks would already cost about 30.000$ one year ago and now only 3000$....

Caterpillar is working on a thermo-electricity concept to reduce fuel consumption on Diesel engines. They provide a cold side of the thermo-electric element with liquid that is part of a secondary cooling system. This is shown on the following picture.

The system's performance is quantified in the following table and diagram.

From this example one can think that obtaining high power levels and thus important reductions in fuel consumption from thermo-electricity seems quite difficult.

Another project currently going on at GM in cooperation with GE, Oak Ridge National Lab, RTI, University of South Florida and University of Michigan sponsored by the Department of Energy in the United States is aiming at 10% fuel economy without increased emissions and at a cost-effective way.

Another development program presented during the 25th annual conference of Thermo-electrics is a project by NREL, BMW, Visteon, BSST, NASA and West Virginia NETL. Here a BMW 530i is equipped with a thermo-electric generator in the exhaust.

The program goal is 10% fuel efficiency improvement. A vehicle simulation capability has been established in GT Cool that has been validated and may be used to predict and optimize system performance. The thermo-electric generator module has been designed incorporating thermal isolation and high density design principles. A secondary loop has been implemented to improve system efficiency, reduce material usage and to address potential environmental issues. The performance and behaviour of this architecture will be evaluated in phase 2 using build and test results fed back into the GT Cool model.

Some more developments are going on at Quantum Well TE, High-Z Technology Inc., Teledyne Energy systems, NRE Lab, Monash University, Delphi, Clarckson University and others.

The thermo-electric developments currently focus on finding better materials to convert the heat flow into electricity. The costs per Watt are still very high.

Regardless of high cost, the limited level of efficiency and low power density makes the principle not yet appropriate for exhaust energy regeneration. The development has a long history and seems to have many years to go before cost effective solutions may see the market. The threat to this technology will of course come from other concepts of heat to power conversion as discussed in this benchmark. The getting cheaper and more powerful of materials is difficult to predict; thermo-electric developments have been quite slow compared to the market requirements. The Diesel Engine Emission Reduction conferences concentrate a lot of efforts in this field and can keep you updated on the technology. See also this link


The BMW Turbosteamer uses the heat from the exhaust to transform water into steam which in turn is led through an expansion module that transmits torque to the crankshaft. The reduction of fuel consumption of the complete engine system announced by BMW is in the order of 17%. It is not clear from the press communication if this is an average on NMVEG-cycle or on an optimized point. The press release does not explicitly mention the use of water, but the medium most likely used for the process is R245fa or perhaps a mixture of R134a and R245fa. The system then could be classified under the Organic Rankine Cycle (ORC) devices.

The following two pictures show a test at BMW on a test bench with the temperatures along the exhaust line.

The system is built up of heat exchangers in the exhaust, a cooler at the front of the vehicle (where the heat finally exits) and a motor that gives off its power to the main engine:

Integration into a vehicle then looks like this:

Additional information can be found on Autoblog.

Courtousy of BMW for the pictures

Stirling di-thermal engine

The principle of the Stirling engine that uses two levels of temperature allows a high efficiency of regeneration of lost heat. The power density of the Stirling engine is however quite low and makes it therefore not appropriate for vehicle applications even if the efficiencies that can be obtained for the regenerations are in the 50 to 60% levels. The main difference between the heat2power principle and the Stirling engine is the low power density of the Stirling engine whereas the efficiencies are comparable.

Organic Rankine Cycle (ORC)

The Organic Rankine Cycle (ORC) is a thermodynamic cycle according to the Rankine principle but specifically uses organic liquids/gases in order to have a boiling point at relatively low temperatures. The heat is used to make the liquid boil and generate high pressure gases that will then drive a motor able to transmit torque to the crankshaft. There exist two main types of machines able to do this: based on a turbine or based on a reciprocating piston:

AVL has given a very instructive presentation (from which originate these images) in the 2006 DEER conference on ORC and proposes 18% fuel economy for a system that regenerates heat from the turbo-intercooler (CAC), the exhaust and the EGR. The most attractive on-road-vehicle applications as seem by AVL are on heavy-duty diesel trucks for long distance hauling (Class 7-8) because of their enormous amounts of fuel consumed each year and therefore of their quick return on investment. AVL also presented a system that is represented in the following diagram:

The system is designed to regenerate heat from three different parts, each with its specific temperature. The fluid undergoes a cycle in the T-S diagram that then looks like the yellow line in the following image:

The current developments in regulations of exhaust gases will show more and more application of high degrees of Exhaust Gas Recirculation (EGR) which results in higher temperatures. The evolution in exhaust temperatures that follows from this is shown in the following diagram:

The effect of the detent of the exhaust gases can be seen in the decreasing temperatures for higher loads in the following diagram (with EGR):

The above characteristic can be considered as the exhaust temperatures for truck engines with normal cylinder heads in the coming years. This results in the following "efficiency" diagram:

Supercritical WHR-ORC system may be more practical (the high-cost evaporator can be avoided). The WHR-ORC system can function as a LT-coolant loop for EGR and charge air cooling. The expander/turbine can be bypassed when waste heat level is too low - the Rankine loop reduces to a regular LT-coolant loop. If waste-heat temperatures > 400°C, it is possible for the efficiency of the WHR Rankine engine to be 15 - 20%.

Among the liquids for the Organic Rankine Cycle R245fa is one of today's favourites. It is a non-chlorinated hydrofluorocarbon, non Ozone depleting liquid with a low global warming potential. Furthermore it is non-flammable, it has good heat transfer ability, excellent thermal stability and low viscosity. It can work with the existing AC tool set in service shops and runs above atmospheric in its cycle. In its behaviour it is similar to R134a.

Cummins, with the help of the US Department of Energy, also has a ORC project with the ISX engine to recover waste heat energy from the engine´s turbo intercooler (CAC) and EGR. Cummins also opted for R245fa as the working liquid.

The program goals are threefold:

Cummins currently works on a system as shown in the following diagram:

Cummins was to integrate this concept into a Class 8 Tractor and demonstrate this on-highway. They are planning on using the International ProStar for this program

Furthermore Cummins expects to reduce intercooler and EGR heat rejection by the recovery cycle efficiency of about 20% at peak power conditions. This reduction offers a significant benefit to the Vehicle OEM

They finally calculate a 9000$ payback on 18 months on the base of 120,000 miles per year and 3$ per gallon. They conclude that to get the customer to buy this, the system should cost significantly less than 9000$...

General considerations

A presentation of one of the concepts above showed a cost comparison for 18% recovery case on a class 8 truck that runs 200,000 miles per year:
200,000 miles/year / 6 mpg = 33,333 gallons per year
33,333 gallons x $3.15/gallon = $105,000/yr
$105,000/yr x 0.18 fuel economy = $18,900 saved/year
Conclusion : There is big money to be made if double digit fuel economy can be achieved.