Printer Friendly

Concept for improving cost effectiveness of thermoelectric heat recovery systems.

ABSTRACT

The practical application of heat recovery using thermoelectrics requires the realization of reasonable cost effectiveness. Therefore, a thermoelectric generator (TEG) structure that can compatibly increase efficiency and reduce cost was investigated with the aim of enhancing cost effectiveness. To increase efficiency, a method of using a vacuum space structure to reduce the TEG size was investigated to enable installation just after the close-coupled catalyzer, which is subject to many space restrictions. It was found that by making it possible to use high temperature exhaust heat, power generation efficiency can be increased to approximately twice that of the typical under floor installation. In addition, coupled simulation of heat transfer and power generation using FEM, 1D cost effectiveness simulations, and bench tests were performed with the aim of reducing cost. These results clarified that use of a TEG with a heat concentration structure enabled to reduce the number of elements used to 1/4 and the system cost to approximately 1/2 that of the previously studied Honda system, while still maintaining approximately the same size and power output.

CITATION: Mori, M., Matsumoto, M., and Ohtani, M., "Concept for Improving Cost Effectiveness of Thermoelectric Heat Recovery Systems," SAE Int. J. Passeng. Cars - Mech. Syst. 9(1):2016.

1. INTRODUCTION

Various entities have actively investigated heat recovery technology in the past with the aim of enhancing automobile fuel economy. Many of these studies have reported that heat recovery technology helps to enhance automotive fuel economy.[1,2] Honda has also investigated heat recovery technology such as use of a Rankine cycle or thermoelectrics, and has verified that heat recovery is possible in each case.[3,4] However, heat recovery technology has not yet been practically applied to automobiles. A major reason why practical application has not been achieved is that prospects could not be obtained for achieving effects at a suitable cost. Among the studied heat recovery technologies, heat recovery systems using thermoelectric elements feature a simple structure with no moving parts. For this reason, these systems were thought to be advantageous from the viewpoint of cost effectiveness, particularly for small-scale power generation with an average power generation output of less than 1 kW such as heat recovery in a passenger car. Even so, there were still no prospects for obtaining reasonable cost effectiveness. Compatibly achieving a high balance between cost and recovery efficiency is an effective way to enhance cost effectiveness. Therefore, realistic measures that can compatibly increase efficiency and reduce cost were investigated.

2. METHODS FOR INCREASING EFFICIENCY

2.1. Approaches for Increasing Efficiency

Equation (1) gives the power generation efficiency using thermoelectrics. There are mainly two methods for increasing efficiency. One method is to enhance thermoelectric element performance to increase the figure of merit ZT, and the other method is to increase the difference between Th and Tc, that is to say the temperature difference applied to the thermoelectric element.

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1)

[eta]: conversion efficiency

ZT: figure of merit

[T.sub.h], [T.sub.c]: Temperature of TE element at hot and cold side

High expectations are placed on increasing ZT, that is to say enhancement of element performance. However, technical advances are extremely challenging to predict, and the future cannot be known for certain. At present the method for most reliably increasing the recovery efficiency is to increase the temperature difference applied to the thermoelectric element. Therefore, utilization of a high exhaust gas temperature was considered.

2.2. Relationship between Installment Location and Exhaust Gas Temperature

Automobile exhaust gas is higher temperature farther upstream in the exhaust system, and the temperature becomes lower due to heat radiation as the exhaust gas moves downstream. Figure 1 shows the results of measuring the average exhaust gas temperature during US 5 cycles at various locations in the exhaust system of a 13MY US

Accord having 2.4L L4 engine with MT. The measurement locations express the distance downstream from the reference position 30 mm from the inlet of the close-coupled catalyzer substrate. The average exhaust gas temperatures during US 5 cycles measured at the position just after the close-coupled catalyzer and at the under floor position differed by approximately 180[degrees]C. This shows that a higher exhaust gas temperature can be utilized by installing the thermoelectric generator (TEG) just after the close-coupled catalyzer.

2.3. Issues and Measures for Location just after the Close-coupled Catalyzer

Installing the TEG in a location that enables to use a higher exhaust gas temperature is a very natural idea when considering heat recovery. However, TEG is typically installed under floor where the exhaust gas temperature is lower. Figure 2 shows an example of installment by GM.[5] furthermore, as seen in other cases of development by BMW, Ford, Fiat and Renault[6,7,8,9], past development of heat recovery systems using thermoelectrics mainly uses under floor installment.

Figure 3 shows an example of packaging around close-coupled catalyzer in a Honda Civic.

A close-coupled catalyzer is located just after the engine. In order to satisfy emissions regulations, the TEG needs to be located downstream from the catalyzer, so the space available for the TEG is extremely limited. The major reason why an under floor layout is the most common despite it being desirable to use a high exhaust gas temperature to increase recovery efficiency is thought to be the challenge of securing sufficient installment space upstream. Therefore, a method was devised of using a vacuum space structure to reduce the TEG size in order to facilitate installment in even a small space. The feasibility of this method was reported at the 2015 SAE World Congress.[10]

When thermoelectric module is mechanically jointed with hot and colds side heat exchanger, difference in thermal expansion between hot and cold side causes large stress to thermoelectric elements leading to crack to the thermoelectric element as shown in Fig. 4.

To resolve this issue, structure shown in Fig. 5 is typically adopted. Heat exchanger is mechanically disjointed from thermoelectric module to reduce stress of thermoelectric element. However, to conduct heat, heat exchangers need to be strongly pressed against module to acquire good contact, and fasteners were used to press heat exchanger against module as shown in Fig. 5. This complicates structure inhibiting TEG to become compact.

For vacuum space structure, it allows compact size having equivalent ability to reduce stress as the conventional structure. As shown in Fig. 6, cold side heat exchanger is disjointed from thermoelectric module similar to conventional structure. However, space around module is vacuumed, and cold side heat exchanger is pressed against thermoelectric module by atmospheric pressure through flexible plate instead of force created by fasteners. In this structure, fasteners are no longer necessary and it simplifies structure enabling compact TEG.

2.4. Possibility and Effects of Installment just after the Close-coupled Catalyzer

This study newly created a concrete design for applying a TEG with a vacuum space structure to an actual vehicle, and investigated the packaging capability. Figure 7 shows the packaging that assumes installment in a 16MY US Civic. In addition to the generator, the TEG unit packaging also assumed installment of a bypass valve and actuator as shown in Fig. 8. This valve system is purposed to suspend heat recovery by bypassing exhaust gas when water temperature is too high. It is intended to be operated only in special circumstances when vehicle is operated under severe condition and water temperature becomes close to overheat. Therefore, it is not used normally including entire duration of driving cycles evaluating fuel economy since water temperature will not become high enough to operate bypass valve.

Table 1 lists the specifications of the generator used in the installed TEG unit. The use of a vacuum space structure results in an extremely compact TEG, so it was found that although there is size restrictions to generator, the TEG can fit in the space just after the close-coupled catalyzer.

Figure 9 shows the results calculated by simulation of the regenerative power output relative to the exhaust gas temperature at the TEG inlet.

As shown in Fig. 1, the average exhaust gas temperature during US 5 cycles is approximately 630[degrees]C just after the close-coupled catalyzer and approximately 450[degrees]C at under floor, for a difference of approximately 180[degrees]C. As a result, the power generation level due to the difference in the installment position shows that the power generated for the layout just after the close-coupled catalyzer is approximately twice that for the under floor installation. Thus, it was found that by reducing the size to address the issue of installment space in an upstream installation, the TEG conversion efficiency can be increased to approximately twice that of the conventional under floor installation.

Figure 10 shows the simulation results for the relationship between the number of elements and the power output when the TEG is installed just after the close-coupled catalyzer and when installed under floor.

In case of both under floor installment and installment just after the close-coupled catalyzer, the power generation level initially increases together with the number of elements, and then tends to reach a saturation level. This is because the power output increases up to the scale of the recoverable exhaust gas energy, but the exhaust heat of an automobile is limited, so once that scale of recoverable exhaust gas energy is secured, heat recovery cannot increase further even if the number of elements is increased. Conversely, under floor installment and installment just after the close-coupled catalyzer differ in terms of saturation power output. The reason for this is that the exhaust gas temperature differs according to the location, so a difference appears in the conversion efficiency as shown in Fig. 9.

Installment just after the close-coupled catalyzer is subject to severe space restrictions, but it is already clear that power output near the limit recoverable by under floor installment can be obtained given the specifications shown in Table 1. Installment just after the close-coupled catalyzer has high conversion efficiency approximately twice that of under floor installment. This means that comparing TEG having same number of elements, the power generated by installment just after close-coupled catalyzer is twice that of under floor installment. Also from characteristics shown in Fig. 10 indicates that, the number of elements needed to obtain the same power output can be greatly reduced, and the size of the generator, which is roughly proportional to the number of elements, can also be reduced, especially notable when power approaches near the limit recoverable by under floor installment.

3. METHODS FOR REDUCING COST

3.1. Relationship between System Cost and Element Cost

The cost of a thermoelectric heat recovery system consists of the cost of the thermoelectric elements, electrodes, heat exchanger and various other parts that comprise the TEG, and the cost of assembling these parts. In addition to the generator, there is also the cost of a bypass valve and actuator to stop heat recovery when the water temperature is high, a DC-DC converter to supply the electric power generated by the TEG to the vehicle, and other parts. Various cases can be considered according to the specifications of the TEG to be installed. Therefore, a 1D simulation was constructed that can evaluate cost effectiveness by calculating the system cost and the fuel economy enhancement effect across US 5 cycles for these various cases.[11] Figure 11 shows the results of using this 1D simulation to calculate the percentage accounted for by the thermoelectric element cost in the heat recovery system assumed by Table 2.

The element cost accounts for approximately 60% of the overall system cost. Thermoelectric elements contain rare metals and pass through various manufacturing processes before reaching the final product stage, so they typically tend to be expensive. In Honda's case this factor had a large impact on the system cost.

3.2. Approaches for Reducing Cost

The most effective method of lowering the system cost is thought to be to reduce the thermoelectric element cost, which accounts for a large percentage of the overall cost, so this was investigated. There are mainly two methods for suppressing the element cost. One is to lower the element unit cost, and the other is to reduce the number of elements used.

It is considered likely that the element unit cost can be reduced in the future for reasons such as enhancement of manufacturing methods and expanded production scale, but this is unclear at present.

Therefore, this study focused on the other method, that is to say reducing the number of elements used. However, the aim was to reduce the number of elements used without reducing the power generation level. To reduce the number of elements while maintaining the power output, it is necessary to increase the generated power output per element to compensate for the drop in power output due to the reduced number of elements. Equation (2) gives the generated power output of an element.

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (2)

P: electrical power generation

q: heat flux

ZT: figure of merit

[T.sub.h], [T.sub.c]: Temperature of TE element at hot and cold side

This shows that in addition to enhancing the element performance or increasing the conversion efficiency by expanding the temperature difference applied to the element as described previously, the generated power output of an element can also be increased by increasing the heat flux q passing through the element. The heat flux passing through the element can be increased by increasing the heat exchange performance, such as by making the fins finer. However, enhancement of heat exchange performance typically involves an increase in exhaust gas pressure loss, which leads to increased fuel consumption. Therefore, a heat concentration structure using electrodes was devised as a means of increasing the amount of heat passing through the element without a corresponding increase in pressure loss.

3.3. Mechanism for Reducing the Number of Elements by Using a Heat Concentration Structure

Figure 12 shows a schematic of the heat transfer structure of a typical TEG. The exhaust gas heat is transferred first to the heat exchange fins and then to the electrodes and thermoelectric element located directly below the fins. The sum of the combined thermal resistance [R.sub.h] of the heat transfer fins and electrodes and the thermal resistance [R.sub.TE] of the element is the total thermal resistance from the fins to the end of the thermoelectric element.

Figure 13 shows a schematic of a heat concentration structure utilizing electrodes. The mechanism for reducing the number of elements used is described below using as an example a heat concentration effect of three times.

The structure shown in Fig. 13 increases the number of fins per element to three times that of the conventional structure without changing the fin specification by widening the gap between elements compared to the structure in Fig. 12. In this case, assuming the ideal condition in which the thermal resistance Rinplane in the in-plane direction shown in Fig. 13 is small enough to be ignored, the combined resistance of the fins and electrodes that use a heat concentration structure is [R.sub.h]/3, where [R.sub.h] is the combined resistance for the conventional structure. In addition, by reducing the element height to 1/3 as shown in Fig. 13, the thermal resistance of the element can also be reduced to [R.sub.TE]/3, where [R.sub.TE] is the thermal resistance for the conventional structure. The total thermal resistance from the fins to the end of the thermoelectric element of the structure shown in Fig. 13 is 1/3 that of the conventional structure shown in Fig. 12, so the amount of heat that passes through the element can be increased by three times compared to the conventional structure. Regarding the temperature difference applied to the thermoelectric element, the thermoelectric element height is reduced to 1/3, but the heat flux passing through the element is increased to three times, so the values of [T.sub.h] and [T.sub.c] are the same as those for the conventional structure. According to Eq. (2)q is three times that of the conventional structure, and the ZT, [T.sub.h] and [T.sub.c] that affect the conversion efficiency are unchanged, so the power output per element is three times larger. This means that the same module power output can be obtained even when the number of elements is reduced to 1/3. In addition to being able to reduce the number of elements used to 1/3, the element height is also reduced to 1/3 as shown in Fig. 13, so in terms of element weight this is a reduction of the amount of element material used to just 1/9.

3.4. FEM Simulation of Heat Concentration Structure

The above example showing the element reduction effect of the heat concentration structure is the effect assuming that the thermal resistance [R.sub.inplane] in the in-plane direction shown in Fig. 13 is small enough to be ignored. However, thermal resistance is actually present, so the heat concentration effect is assumed to decrease noticeably as the element gap widens. Therefore, a realistic heat concentration effect was investigated by coupled calculation of heat transfer and power generation using FEM. Figure 14 shows the simulation model.

The impact of widening the thermoelectric element gap on the generated power output was calculated for the element and electrodes configuration shown in Fig. 14. Calculations were performed for three hypothetical cases and the results were compared. The first calculation estimated the ideal condition with no thermal resistance in the in-plane direction. The second calculation estimated the case when the element gap is increased as shown on the right side of Fig. 15. Here, the electrodes size is increased in accordance with the wider element gap in order to use copper electrodes with good heat conduction for heat concentration. The third calculation estimated the case when the element gap is increased as shown on the left side of Fig. 15. Here, the electrodes size is increased by only the minimum necessary to electrically connect the elements, so the heat concentration effect using the electrodes cannot be actively obtained.

Figure 16 shows the calculation results. The case when the electrodes size was increased in accordance with the wider element gap in order to utilize the heat concentration effect of the electrodes was predicted to be almost no different from the ideal condition up to an element gap of 3 mm, and then to gradually diverge from the ideal condition at a gap of 4 mm or more. In contrast, the case when the increase in element size was limited to the minimum necessary to support the wider element gap and a sufficient heat concentration effect could not be manifiest was estimated to diverge from the ideal condition immediately after starting to widen the element gap, and this divergence was predicted to increase greatly as the element gap widened further.

3.5. Verification of FEM Results

A thermoelectric module was created with an element gap of 4 mm and a heat concentration structure using electrodes as shown on the left side of Fig. 17 to verify the validity of the heat concentration effect estimated by FEM. The facility shown on the right side of Fig. 17 was used to compare the actual generated power output with that predicted by FEM.

Figure 18 shows the power output results obtained when the respective temperature difference was applied to the thermoelectric module in the experiment and the power output results predicted by FEM.

The experimental results and FEM predictions match closely, which confirmed the validity of the heat concentration effect predicted by FEM.

3.6. Element Reduction and System Cost Reduction Effects of Heat Concentration Structure Using electrodes

The heat concentration effect validated by the FEM and bench test results was applied to the 1D simulation to calculate the system cost effectiveness, and the impact of utilizing the heat concentration effect on the TEG specifications, size and cost was investigated. Figure 19 shows the impact of widening the element gap on the cost effectiveness.

First, regarding groups with the same element gap, even when TEG have the same element gap, the fuel economy enhancement effects and cost vary according to the number of elements, the element height, the heat exchanger specifications and other factors. Dots of the same color in the above graph plot the results for the same element gap with different combinations of other specifications. A Pareto solution of the cost effectiveness for TEG with the same element gap can be obtained by calculating various combinations of specifications.

Focusing on the arrangement of groups with different element gaps, the cost required to achieve the same fuel economy enhancement drops as the element gap widens. This is the result of utilizing the heat concentration effect to reduce the number of elements used as described previously. The conventionally assumed specifications shown in Table 2 correspond to A in the figure. B represents specifications with an element gap of 4 mm that generate the same power output as A. Both specifications A and B are located in the respective Pareto solution of the cost effectiveness for each element gap. Comparison of system cost shows that the cost of A is approximately half that of B. The schematics above the Fig. 19 graph summarize the differences between specifications A and B. The main reasons for this large reduction in cost are a 75% reduction in the number of elements used, and a reduction of 90% or more in terms of element mass due to a reduction in element height. Conversely, a disadvantage of using a heat concentration structure is that the heat exchange area of A is approximately 5% larger than that of B, so the size is larger. This is because the heat resistance in the in-plane direction shown in Fig. 13 is not zero, so the heat exchange area needs to be slightly larger than the conventional area to compensate for this resistance. However, the use of a heat concentration structure enabled to reduce the system cost by approximately half in exchange for an increase in TEG size of only around 5%. Therefore, this is thought to be a very helpful method of securing reasonable cost effectiveness for practical application.

4. CONCLUSIONS

1. The results of this investigation indicate that using a vacuum space structure to reduce the TEG size enable installment just after the close-coupled catalyzer. The results of simulation estimates show that approximately twice the conversion efficiency can be obtained and that twice the power can be obtained by the generator block having same size or the size of the generator block needed to obtain the same power output can be greatly reduced compared to conventional under floor installment.

2. It was clarified that the use of a heat concentration structure using electrodes enables to reduce the number of elements without reducing the power output. It was found that by using a heat concentration structure widening the element gap to 4 mm, the number of elements used can be reduced to 1/4 and the system cost can be reduced by approximately half compared to the conventional TEG used by Honda which does not actively utilize heat concentration.

TEG installment just after the close-coupled catalyzer to increase efficiency, and use of a heat concentration structure to reduce system cost are compatible technologies that can be applied simultaneously without counteracting each other's effects. Prospects were obtained for increasing power generation efficiency by approximately two times and reducing system cost by approximately half. This indicates that cost effectiveness can be greatly increased compared to conventional systems, and is thought to be a large step closer to the cost effectiveness required for practical application.

REFERENCES

[1.] Ringler, J., Seifert, M., Guyotot, V., and Hubner, W., "Rankine Cycle for Waste Heat Recovery of IC Engines," SAE Int. J. Engines 2(1):67-76, 2009, doi:10.4271/2009-01-0174.

[2.] Hussain, Q., Brigham, D., and Maranville, C., "Thermoelectric Exhaust Heat Recovery for Hybrid Vehicles," SAE Int. J. Engines 2(1):1132-1142, 2009, doi:10.4271/2009-01-1327.

[3.] Ibaraki, S., Endo, T., Kojima, Y., Takahashi, K., et al., "Research of a Rankine Cycle On-Board Heat Waste Recovery System", JSAE annual spring congress proceedings, 20065602 No.92-06, 2006

[4.] Mori, M., Yamagami, T., Sorazawa, M., Miyabe, T. et al., "Simulation of Fuel Economy Effectiveness of Exhaust Heat Recovery System Using Thermoelectric Generator in a Series Hybrid," SAE Int. J. Mater. Manuf. 4(1):1268-1276, 2011, doi:10.4271/2011-01-1335.

[5.] Meisner, G, P., "Skutterudite Thermoelectric Generator for Automotive Waste Heat Recovery", 3rd Thermoelectric Application Workshop, 2012

[6.] Mazar, B., "State of the Art Prototype Vehicle with a Thermoelectric Generator", 3rd Thermoelectric Application Workshop, 2012

[7.] Maranville, C., "Thermoelectric Opportunities for Light-Duty Vehicles", 3rd Thermoelectric Application Workshop, 2012

[8.] Magnetto, D., "HeatReCar: First Light Commercial Vehicle Equipped with a TEG", I AV 3rd Conference on Thermoelectrics, 2012

[9.] Aixala, L., and Monnet, V., "Conclusion of RENOTER Project", IAV 3rd Conference on Thermoelectrics, 2012

[10.] Matsumoto, M., Mori, M., Haraguchi, T., Ohtani, M. et al., "Development of State of the Art Compact and Lightweight Thermoelectric Generator Using Vacuum Space Structure," SAE Int. J. Engines 8(4):1815-1825, 2015, doi:10.4271/2015-01-1691.

[11.] Matsumoto, M., Mori, M., Yamagami, T., Haraguchi, T., et al., "Prospect on Commercialization of Thermoelectric Heat Recovery System," JSAE annual spring congress proceedings, 20145152 No.42-14, 2014

Masayoshi Mori, Manabu Matsumoto, and Makoto Ohtani

Honda R&D Co., Ltd.

CONTACT INFORMATION

Masayoshi Mori

Honda R&D Co., Ltd. Automobile R&D Center 4630 Shimotakanezawa, Haga-machi, Haga-gun, Tochigi, Japan, 321-3393

masayoshi_mori@n.t.rd.honda.co.jp

Phone:+81-80-4857-8963

ABBREVIATIONS

FEM - finite element model

fp - fin pitch

L4 - in-line 4-cylinder

MT - manual transmission

MY - model year

RAD - Radiator

SAE - Society of Automotive Engineers

TEG - thermoelectric generator

TE - thermoelectric

1D - one dimensional

Table 1. General specification of TEG

Number of exhaust heat exchanger                     2
Sectional dimension of exhaust heat exchanger       72mm X 10mm
Length of exhaust heat exchanger                    57 mm
Fin thickness and pitch of exhaust heat exchanger  t:0.3mm fp:3mm
Sectional dimension of TE elements                   2mm X 2mm
Gap between TE elements                              4mm
Total number of TE elements                        432

Table 2. General specification of Previous TEG

Number of exhaust heat exchanger                      2
Sectional dimension of exhaust heat exchanger        72mm X 10mm
Length of exhaust heat exchanger                     54 mm
Fin thickness and pitch of exhaust heat exchanger   t:0.3mm fp:3mm
Sectional dimension of TE elements                    2mm X 2mm
Gap between TE elements                               1mm
Total number of TE elements                        1728

TE-elements cost  Other

           (60%)  (40%)

Figure 11. Breakdown of thermoelectric heat recovery system cost

Note: Table made from pie chart.
COPYRIGHT 2016 SAE International
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Mori, Masayoshi; Matsumoto, Manabu; Ohtani, Makoto
Publication:SAE International Journal of Passenger Cars - Mechanical Systems
Article Type:Report
Date:Apr 1, 2016
Words:4458
Previous Article:Investigation of characteristics of gear meshing noise under three-axis driving planetary gear set.
Next Article:Real-world thermal effects on wheel assembly efficiency of conventional and electric vehicles.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters