Final Report Summary - ECOJET (Cost-effective, high-efficient micro gas turbine for micro CHP applications)
Executive Summary:
In the ECOJET project, we built a new engine based on a 10 kWe micro gas turbine gas turbine with an integrated electric generator. The product is aiming at an innovative system solution tailed for environmentally friendly micro CHP (Combined Heat and Power) applications for the residential sector.
The motivation being the essential demand of adopting new measures to reduce energy demand in Europe High-efficiency Micro Combined Heat and Power generation (µCHP) has been identified by the EC as a priority to achieve its energy targets given the potential benefits of this technology with regard to saving primary energy, avoiding network losses and reducing emissions.
In the EcoJet project, we developing a unique micro gas turbine with the aim of an electrical efficiency of at least 25% for a 10kWe power generator. This electrical efficiency can theoretically be achieved through three key characteristics of a micro gas turbine: (1) increasing the combustion temperature; (2) raising the compression ratio; (3) raising the efficiency of the recuperator. We focused on the first item, i.e. raising the combustion temperature, which raises the turbine inlet temperature and the recuperator temperature.
The engine has a uniquely robust design and uses very few advanced components, which makes it inexpensive and simple to manufacture. The micro turbine technology is multi-fuel enabled; has a high power to weight ratio, low emissions; no vibration and low noise. The novelty of this solution lies in cost effective application of micro turbine technology in domestic CHP, currently mainly addressed by reciprocating engines. EcoJet has an innovative design, where all components that need to be kept cold are located in the centre of the engine where the cold intake air passes by, whereas all the hot parts are located in the periphery. In order to achieve this, material studies were made in order to find the right materials for the monorotor. In addition hereto coating possibilities were studied for the monorotor as well as the heat echanger. Material as well as coatings were determined and adequate solutions were found. The process of producing the monorotor was a challenge – more companies tried until a suitable partner was found and a solution for prototyping production was found. During this process a procedure for mass-production was established.
The ECOJET system can be manufactured cost effectively, with few advanced components, and with high power-weight ratio (1/5 compared with a diesel generator) .The micro gas turbine can be produced with a diameter of 40 cm, height 40 cm, and a weight of 35 kg.
The EcoJet solution differentiates from its competitors by the simplicity and cost effectiveness of its innovative design. In addition to ease of manufacturing and comparatively low production costs, the design – with only one rotating part – enables very low cost of ownership due to its operational features and low maintenance cost
Our system will allow overcoming the barrier to produce a cost-effective, environmental friendly and simple product, The EcoJet solution thus has a range of unique selling propositions to a very large group of potential end-users – manufacturers of micro CHP systems and, ultimately, households.
On the long run the project results are expected to have a big impact on contribution to the EU energy efficiency policy. Furthermore, the overall impact of EcoJet will place the EU SMEs in a highly competitive position within the cogeneration market with dominant players currently coming from non-European regions (mainly America and Japan), thereby creating and protecting job, increasing revenues to allow further investment and technological developments
Ecojet aims to base its marketing strategy for the technology primarily on the strong and growing market for distributed micro CHP. At the end-user level, it is estimated that micro CHP can help a consumer save close to 60% on their electricity bills.
Project Context and Objectives:
In the Ecojet project, we designed and developed a new engine – based on a radial gas turbine with an integrated electric generator. The engine has a 10-kWe micro gas turbine power generator in an innovative system solution tailed for environmentally friendly micro CHP (Combined Heat and Power) applications for the residential sector.
The core of the project was the development of a unique micro gas turbine with an electrical efficiency of at least 25% for a 10kWe power generator. This electrical efficiency can theoretically be achieved through three key characteristics of a micro gas turbine: (1) increasing the combustion temperature; (2) raising the compression ratio; (3) raising the efficiency of the recuperator. In the EcoJet project we focused on the first item, i.e. raise the combustion temperature, which raises the turbine inlet temperature and the recuperator temperature.
In order to achieve our goal, we had to design the microgasturbine for the optimal inlet and outlet – we worked on finding the right way of casting the monorotor and we had to study materials in order to find the right material for the microgasturbine. As the gas exhaust heat effects the microgasturbine, we had to find a coating to cover the turbine with possibilities of heat up till 750degC. As the coating would have an influence on the surface of the monorotor it was important to find a way of machine the surface of the monorotor to gain the right surface roughness. Furthermore, we looked into materials and coatings for the heat exchanger.
The project is divided into 5 technical work packages each of them aiming at the specific parts of the final Ecojet solution.
In work package 3 we looked at the system design and functional requirements for the heat exchanger and the monorotor – the two main parts of the ECOJET system and both design and functional requirements were defined and described. Due to manufacturing constrain in the monorotor casting, the nominal flow rate of the gas turbine increased from 140g/s to 200g/s and new specifications were decided.
It was established that the best solution for the recuperator was a GAP heat exchanger type providing the following thermo-dynamical performances (for complete description see deliverable 5.1):
▪ Effectiveness : 0.8
▪ Gas side pressure drop: 220 Pa (0.2%)
▪ Air side performances: 9150 Pa (3.05%)
Due to the flow rate modification, the recuperator effectiveness is lower and the pressure drop higher than the one originally computed for deliverable 5.1. The reason of this is that the recuperator design did not change to compensate this flow evolution. To keep the same performances the flow augmentation would have to be compensated with a heat exchange surface augamentation too but this was not possible considering the overall dimension limitation.
Anyway, it appears that this recuperator design is not the most appropriate one, as there is a plate separation on the exhaust gas side allowing decreasing the gas side pressure drop (this is the reason why this technology choice was made) but implying also a lower effectiveness range due to a reduced heat transfer coefficient on the exhaust gas side.
If we consider normal rules for recuperator design, it appears that this type of heat exchanger i.e. air to gas heat exchanger, has to be compact i.e. without plate separation in order to equilibrate the heat transfer coefficient on both air and gas side. In this way, the energy contained in the exhaust gas can be transferred to the air with less heat exchange surface than previously and finally, the result of this optimization will be a smaller heat exchanger. Moreover, considering the high level of automation in ACTE facilities for producing the COMPACT heat exchanger range, this should also bring a significant cost saving potential.
From the abovementioned, the only way for ACTE to fulfil the optimization purposes i.e. respect the initial specification with a COMPACT recuperator is modifying its plate profile geometry from a triangular shape factor to a trapezoidal shape factor. Indeed, such profile modification will allow keeping the effectiveness target but decreasing both the pressure drop and the weight. After having studied more than 50 different plate profile, the selected one will provide the following performances considering the same overall dimensions i.e. outer recuperator core of 490 mm. the new profile allows a significant gain in the effectiveness that corresponds know to the initial specification but also a smaller global pressure drop i.e. 2.37% instead of 3.25% and also a significant weight reduction. The optimization objectives are then achieved but the final gain in the production cost will be known after knowing the raw material that will have to be used for achieving the corrosion resistance objective.
Finally, note also that this profile is compatible with the elongation limit of all the raw material taken into account by ULB in their corrosion studies. So following their results on this topic the new heat exchanger plate profile will make us able to fulfil all the project objective.
The design of the monorotor was completed within the boundary conditions of the cycle defined. Several iterations of the casting process was performed since the low mass flow resulted in a narrow channel at compressor outlet, constrained by casting to be a minimum of 3mm (POC2 achieved 4mm). At lower specific flow, the flow channels are relatively long and narrow, also a challenge for casting. The castings had a number of issues, but primarily a very poor surface finish on both compressor and turbine sides, probably reflecting an inherent roughness in the rapid prototyping approach used to make the wax masters for investment casting. Some prototypes viewed by PCA had apparent casting ‘debris’ within the channels. Although the basic premise was that the components would run as-cast with the addition of thermal barrier coatings, this was found not to be practicable. Considerable effort was expended by the SME and RTD partners in trying to machine the gas-swept surfaces, with some success on the turbine side, though this is not a viable option going into economic production. However, the compressor channel is relatively inaccessible, the gap between the shroud and hub, combined with the curvature and length of the channel, means only very limited surface dressing is possible. For the gas turbine to self-sustain, the efficiency of both compressor and turbine (the product of the two) must be sufficiently high. As normal practice, and especially for a case where efficiency is of primary importance to the economic viability of the project, the design assumed hydraulically smooth surfaces for both compressor and turbine. PCA did produce a second design with increased flow to 200g/s which may have eased some of the problems, though not to The limited test data is not of much use in calibrating the design since it is so far off-design. The tests ran at low turbine inlet temperature so there is no significant feedback on heat-transfer effects, although there were some signs that the ECOJET layout had some advantages over POC2 due to an air gap between the turbine nozzle and compressor diffuser, which gave some insulation. Although the POC2 prototype had been produced to an acceptable standard, the reduction in channel height for ECOJET has been a step too far. PCA carried out some limited studies to try to produce a design that had ‘line of sight’ through the compressor to afford some access for cleaning, as shown below, but this was not taken up by the lead partner.
In work packages 4 materials and coatings are studied, tested and selected for the two main parts – heat exchanger and monorotor. The design of the turbine is related to the capabilities of these alloys to resist to high temperatures – above 750degC and high temperature corrosive and oxidizing gases. Several HRSA and their properties have been studied and summarized in order to help the turbine designers to choose and justify their decision on the optimal alloy.
The mechanical properties related to the shift in temperatures have been presented in a spread sheet – hereby providing a clearer picture and a justification for using certain HRSA while avoiding using others. This information aims at helping on achieving the optimal alloy to be used in order to keep the best trade performance vs endurance vs cost. This trade is one of the main drivers of the whole Ecojet project.
During the research performed by DAMRC, several properties of the thermal barrier coating were analyzed. The goal was to relate the coating procedure with the final surface properties achievable in order to produce the most efficient surface from the thermodynamic perspective. A better surface will dramatically improve the performance of the turbine as it was stated by PCE (8% improvement of performance of the monorotor from Ra=10µm to a Ra=1,6µm ). Taking this issue in focus, DAMRC researched the theoretically best achievable coating surface in combination with the machining of the raw casted of the part. After this short research was complete, a report labeled 140630 Coating properties vs roughness was created.
The most relevant conclusions of this report are;
• The porosity and hardness of the coating have inverse behaviors when changing the thickness of the coating.
• The higher the porosity of the coating, the lower is the hardness of the coating.
Another relevant conclusion had to do with the roughness of the substrate (casted monorotor) and the adhesion strength of the thermal barrier coating. The adhesive strength of the coating is strongly dependent on the roughness of the substrate material. The Adhesive strength is critical for the reliability of the coating hence, it is also critical for the reliability of the Ecojet concept. If the adhesive strength is poor, the coating might experience delamination problems. These problems will lead to a poor thermal isolation of the substrate material thus; it will lead to poor thermal and mechanical properties and poor reliability. It was also presented a paragraph with the thermal isolation properties of thermal barrier coatings. After combining all these properties of the coating and substrate materials, a decision for using a nominal coating thickness of 400µm (coating material provided by TE&M; Ceria-Yttria-Stablized-Zirconium-Oxide). The next challenge while coating the monorotor is that the coating process does not allow for coating thickness tolerances better than 300µm while our nominal thickness is 400µm. This would mean that we have got a potential geometrical error of 75% of the coating thickness. From the assembly and mechanical properties, this error would make the monorotor unsuitable. So as a conclusion, the coating had to be machined before put up to service.
The machining of Oxide materials such as ZrO2 cannot be achieved by traditional machining methods because the hardness of the coating is superior to the hardness of the tool. This fact would ruin the tool almost immediately. DAMRC developed a state-of-the-art process in order to overcome this issue. This new process was able to perform the machining operations presenting results even better than the requirements of the project (roughness up to; Ra=0.95µm, Rz=8µm). Note: (a) Surface of the turbine when coated. (b)Surface of the coated-monorotor after machining ZrO2. The productivity of this process was also in focus, achieving a material removal rate up to 10 times higher than a grinding process with similar geometrical characteristics.
In work package 5 the focus in on designing and producing the heat exchanger. In D5.1 the mechanical design had been realized accordingly with the thermo-mechanical constrain due to working conditions and transient time (ramp-up & shut down):
▪ Performances assessment with ACTE semi-CFD tool
▪ Creep computation
▪ Thermal shock analysis & fatigue life time
▪ Interface definition (position and type of connection accordingly with Radijet specification)
▪ Recuperator integration in the Radijet frame
D5.2 corresponds to:
▪ The recuperator core manufacturing in the ACTE production line.
▪ The radial collector (headers) machining and mounting on the core
▪ The radial collector brazing on the core
▪ Air inlet interface machining
After manufacturing, this core was shipped for applying the most appropriate coating layer before sending the core for mounting it in the CHP system. However, due to the last event involving the closure of the lead partner Radijet, ULB had not the opportunity to deliver its recuperator for mounting and testing. As a conclusion, we can say that task 5.3 had right been completed but the recuperator has not been integrated into the gas turbine frame.
In work packages 6 the main objective is the design, casting process, coating and machining of the monorotor.
Casting processes are class-divided into the main techniques concerning their mold material into permanent mold and expendable mold processes. Permanent mold processes, e.g. die-casting, are generally used with nonferrous alloys because of their matching thermal resistance. The expendable mold processes are in advantage to cope with high temperatures and therefore are appropriate for casting steel alloys for instance. The lost wax method out performs any other casting process when it comes to superior surface finishes especially compared to sand casting. The lost wax process or investment casting process can be used to produce the most complex shapes in just about any alloy. Advantages of the process are for example liberty of shape, free choice of material, dimensional accuracy and reproducibility of mold details, high surface quality, small up to large quantities possible, flexible regarding piece weight and dimensions. Instead of wax-patterns, RP-models out of PMMA can be inserted into the process when the final design has not been determined and short development time cycles were desired.
For the ECOJET system the various casting methods were evaluated. The specification details of the monorotor were reviewed and the following specifications agreed:
1. The wall thickness must be more than 1 mm
2. The enclosed channels must be at least 3 mm in height
3. Ceramic cores in the compressor channels will give a surface roughness RA between 1.6 - 3.2 μm.
4. The compressor will be designed without splitter blade, due to difficulties with the ceramic cores.
5. The leading edge of the compressor is very important and is designed as a 1:4 ellipse.
6. The shroud must be uniform thickness all the way. 3 mm would be a good compromise.
7. The blades could be as thin as 1 mm. PCA wanted thicker root blade and thinner outer blade shape.
8. This applies to both compressor and turbine. This was a good solution for casting point of view.
9. Different materials suitable for casting were suggested: Inconel 718 and Inconel 738.
10. The design will be according to VDG P690.
The casting process involves up till 12 various tasks and decisions.
1. Choice off master pattern – rapid prototyping has the shortest development cycles and delivery times within lower acquisition costs.
2. Casting design and definition of shrinkage – see the above agreed specifications
3. 3D file format and quality - the software solid works is used and IGES-format has been specified as main interface
4. 3D model printed part material - All rp-models have been printed using a 3D high-performance printer.
5. 3D model printed part surface roughness - the surface quality of rp-models is highly influenced by the layer thickness, the grain size and the infiltration method. Attempts have been made to minimize the surface roughness of the rp-models. Two different methods have been tried. Unfortunately, all results of varying the printing angel had to be rejected. The relation between the printing angel and surface quality is disadvantageous. The best results were achieved by infiltration type 2 material with Ra = 13.93-14.99 μm. Subsequently the surface roughness has been measured on the casting. A roughness of Ra = 2.8 – 3.7μm has been detected. The shift to lower surface roughness presumably is due to sand blasting as finishing step in the production.
6. Strategy of metal flooding and solidification - The monorotor possesses very thin blades and the mold filling has to be ideally balanced. Therefore, e.g. the melting temperature and the preheating temperatures of the ceramic shell mold must be compatibly divined
7. Ceramic mold creation - The ceramic mold is created by iterative dipping and sanding cycles. After the prime cote has been finished, the refractory was filled in the channels of the monorotor manually. This step of the process has to be done very accurate to avoid air inclusions. Otherwise, during metal casting, the thin ceramic prime coat might break and metal infiltrates the air cavity.
8. Removal of 3D plastic/wax model inside ceramic mold - The dewaxing of the casting units can be conducted in two different ways. One method is flash firing and the other on is to autoclave the ceramic shell. Being environmentally friendly, BFG uses an autoclave to remove the wax out of the ceramic shell. However, the 3D-Printed lost model is out off PMMA and cannot be removed by autoclaving indeed.
9. Casting strategy: Air casting – vacuum casting selection – In the ECOJET project it has been decided to use Inconel 718 as casting material. Inconel 718 is a super alloy based on nickel and was especially developed for high temperature applications e.g. needed for jet engines.
10. Casting technic – ideally the pouring is continuous, rapid and non-turbulent so especially the very narrow blade border areas of the mold are filled without could laps. All these criteria can be satisfied by the roll-over casting technology, being a quite simple construction. During the time the basic material is melted, the furnace is in a normal position. To separate the melt from atmosphere argon comes into operation. When the casting temperature is reached, the pre heated and isolated ceramic shell will be mounted with its pouring gate on top of the furnace. For casting the furnace is turned upside down now. This leads to casting times of approximately 2 - 3 seconds and also a sudden mold filling. This technique has been successfully used to cast the monorotors.
11. Removal of ceramic mold - To remove the ceramic of the casting first off all 2 dimensional areas of the shell will be eliminated by a rocking motion with a forge. Afterwards still remaining ceramic on the casting surface will be removed by shot blasting, however shot blasting/sand blasting is not appropriate to withdrawal ceramic out of the turbine channels from the monorotor. For the final finish, the parts have to be sand blasted at the end.
12. Quality control of the part - The first inspection of the casted monorotor has been executed by a visual product inspection. Afterwards the casted monorotor has to pass a crack detection test.
It is possible to cast the Monorotor in an open casting process using the rollover method. However, the process safety by the rapid prototyping process is not very well, the main problem is the introduction of the ceramic mass into the inner spaces of the Monorotor. This can be done only in a manual way with a lot of manpower. Each model is unique. It cannot be determined whether air bubbles are present. By one good part come circa three bad parts. A series production will have a better process safety.
In work packages 7 all these parts are assembled with the rest of the system and test facilites are defined for the test. Specification of the final test was done. Test site specified and selection of measuring devices done. Due to severe problems on delivering good quality casted monorotors along with severe assembling problems several delays occurred. RadiJet performed several tests. Conclusion: it is not possible to have a self-sustained system due to surface roughness of the turbine blades. Combustion system gave problems in the configuration with nature gas as fuel. The bearing system, inverter and the data log system all works fine. Feb16 test run: power supported at 25-30.000rpm by manual control of fuel and rpm, the turbine not installed in the recuperator. NLR worked hard to build/adapt testing facilities to the need of RadiJet and was ready to perform the tests on receipt of the system. The shipment was planned several times but postponed with the consequence that the system never was shipped to NLR. With the closure of RadiJet it was no longer feasible to assemble and ship the system for test
Project Results:
The ECOJET project designed and developed a new radial gas engine.The engine has a 10 kWe micro gas turbine power generator in an innovative system solution tailed for environmentally friendly micro CHP (Combined Heat and Power) applications for the residential sector.
The main project results involve the design and development of
- Radial micro gas turbine consisting of a) monorotor + b) heat exchanger
- Method and matrials to produce the monorotor
- New heat exchanger
- Coating materials for the heat exchanger and monorotor.
Potential Impact:
The ECOJET project designed and developed a new radial gas engine.The engine has a 10 kWe micro gas turbine power generator in an innovative system solution tailed for environmentally friendly micro CHP (Combined Heat and Power) applications for the residential sector. The EcoJet project designed and developed solution differentiates from its competitors by the simplicity and cost effectiveness of its innovative design. In addition to ease of manufacturing and comparatively low production costs, the design – with only one rotating part – enables very low cost of ownership due to its operational features and low maintenance cost. Our approach will allow overcoming the barrier to produce a cost-effective, environmental friendly and simple product, The EcoJet solution thus has a range of unique selling propositions to a very large group of potential end-users – manufacturers of micro CHP systems and, ultimately, households.
The project results are expected to have a big impact on contribution to the EU energy efficiency policy. Furthermore, the overall impact of EcoJet will place the EU SMEs in a highly competitive position within the cogeneration market with dominant players currently coming from non-European regions (mainly America and Japan), thereby creating and protecting job, increasing revenues to allow further investment and technological developments
Ecojet aims to base its marketing strategy for the technology primarily on the strong and growing market for distributed micro CHP. At the end-user level, it is estimated that micro CHP can help a consumer save close to 60% on their electricity bills.
The focus parameters for the microCHP unit have been defined as the following:
➢ Electrical efficiency: >21%
➢ Total efficiency:> 83%
➢ Noise enclosure:<45 dbA @ 1 meter
➢ Noise exhaust:<55 dbA @ 1 meter
These parameters being the key selling parameters when selling a Micro CHP system.
The dissemination activities have mainly been done on fairs & exhibitions, newsletters and via posters.
List of Websites:
The project website was closed by the lead partner as this partner left the project 1 month before closure.
The partners can be found here:
- TEandM www.teandm.pt Contact person Mr. Ricardo Alexandre [ricardo@teandm.pt]
- ACTE www.acte.-sa.be Contact person: Mr. Sébastien DUBOIS [sebastien.dubois@acte-sa.be]
- FRICHS A/S – www.frichs.dk – Contact person. Mr. Ove Munch [omu@frichs.com]
- Danish Advanced Manufacturing Research Center (DAMRC) – www.damrc.com – Contact person Mrs. Charlotte F. Ilvig [chi@damrc.com]
- Stichting national Lucht- enRuimtevaartlaboratorium (NLR) – www.nlr.nl – Contact person: Dr. ir.A.B. (Arjen) Kloostermann [arjen.kloostermann@nlr.nl]
- BFG Feinguss Niederrhein GMBH – www.bfg-niederrhein.de – contact person Mr. Stefan Daub-Klose,CEO
- PCA Engineers Ltd. www.pcaeng.co.uk – Contact person Dr. C J Robinson chris.robinson@pcaeng.co.uk
- Universite Libre Bruxelles (ULB) ww.ulb.ac.be – Contact person: Prof. M.P. Delplancke-Ogletree
In the ECOJET project, we built a new engine based on a 10 kWe micro gas turbine gas turbine with an integrated electric generator. The product is aiming at an innovative system solution tailed for environmentally friendly micro CHP (Combined Heat and Power) applications for the residential sector.
The motivation being the essential demand of adopting new measures to reduce energy demand in Europe High-efficiency Micro Combined Heat and Power generation (µCHP) has been identified by the EC as a priority to achieve its energy targets given the potential benefits of this technology with regard to saving primary energy, avoiding network losses and reducing emissions.
In the EcoJet project, we developing a unique micro gas turbine with the aim of an electrical efficiency of at least 25% for a 10kWe power generator. This electrical efficiency can theoretically be achieved through three key characteristics of a micro gas turbine: (1) increasing the combustion temperature; (2) raising the compression ratio; (3) raising the efficiency of the recuperator. We focused on the first item, i.e. raising the combustion temperature, which raises the turbine inlet temperature and the recuperator temperature.
The engine has a uniquely robust design and uses very few advanced components, which makes it inexpensive and simple to manufacture. The micro turbine technology is multi-fuel enabled; has a high power to weight ratio, low emissions; no vibration and low noise. The novelty of this solution lies in cost effective application of micro turbine technology in domestic CHP, currently mainly addressed by reciprocating engines. EcoJet has an innovative design, where all components that need to be kept cold are located in the centre of the engine where the cold intake air passes by, whereas all the hot parts are located in the periphery. In order to achieve this, material studies were made in order to find the right materials for the monorotor. In addition hereto coating possibilities were studied for the monorotor as well as the heat echanger. Material as well as coatings were determined and adequate solutions were found. The process of producing the monorotor was a challenge – more companies tried until a suitable partner was found and a solution for prototyping production was found. During this process a procedure for mass-production was established.
The ECOJET system can be manufactured cost effectively, with few advanced components, and with high power-weight ratio (1/5 compared with a diesel generator) .The micro gas turbine can be produced with a diameter of 40 cm, height 40 cm, and a weight of 35 kg.
The EcoJet solution differentiates from its competitors by the simplicity and cost effectiveness of its innovative design. In addition to ease of manufacturing and comparatively low production costs, the design – with only one rotating part – enables very low cost of ownership due to its operational features and low maintenance cost
Our system will allow overcoming the barrier to produce a cost-effective, environmental friendly and simple product, The EcoJet solution thus has a range of unique selling propositions to a very large group of potential end-users – manufacturers of micro CHP systems and, ultimately, households.
On the long run the project results are expected to have a big impact on contribution to the EU energy efficiency policy. Furthermore, the overall impact of EcoJet will place the EU SMEs in a highly competitive position within the cogeneration market with dominant players currently coming from non-European regions (mainly America and Japan), thereby creating and protecting job, increasing revenues to allow further investment and technological developments
Ecojet aims to base its marketing strategy for the technology primarily on the strong and growing market for distributed micro CHP. At the end-user level, it is estimated that micro CHP can help a consumer save close to 60% on their electricity bills.
Project Context and Objectives:
In the Ecojet project, we designed and developed a new engine – based on a radial gas turbine with an integrated electric generator. The engine has a 10-kWe micro gas turbine power generator in an innovative system solution tailed for environmentally friendly micro CHP (Combined Heat and Power) applications for the residential sector.
The core of the project was the development of a unique micro gas turbine with an electrical efficiency of at least 25% for a 10kWe power generator. This electrical efficiency can theoretically be achieved through three key characteristics of a micro gas turbine: (1) increasing the combustion temperature; (2) raising the compression ratio; (3) raising the efficiency of the recuperator. In the EcoJet project we focused on the first item, i.e. raise the combustion temperature, which raises the turbine inlet temperature and the recuperator temperature.
In order to achieve our goal, we had to design the microgasturbine for the optimal inlet and outlet – we worked on finding the right way of casting the monorotor and we had to study materials in order to find the right material for the microgasturbine. As the gas exhaust heat effects the microgasturbine, we had to find a coating to cover the turbine with possibilities of heat up till 750degC. As the coating would have an influence on the surface of the monorotor it was important to find a way of machine the surface of the monorotor to gain the right surface roughness. Furthermore, we looked into materials and coatings for the heat exchanger.
The project is divided into 5 technical work packages each of them aiming at the specific parts of the final Ecojet solution.
In work package 3 we looked at the system design and functional requirements for the heat exchanger and the monorotor – the two main parts of the ECOJET system and both design and functional requirements were defined and described. Due to manufacturing constrain in the monorotor casting, the nominal flow rate of the gas turbine increased from 140g/s to 200g/s and new specifications were decided.
It was established that the best solution for the recuperator was a GAP heat exchanger type providing the following thermo-dynamical performances (for complete description see deliverable 5.1):
▪ Effectiveness : 0.8
▪ Gas side pressure drop: 220 Pa (0.2%)
▪ Air side performances: 9150 Pa (3.05%)
Due to the flow rate modification, the recuperator effectiveness is lower and the pressure drop higher than the one originally computed for deliverable 5.1. The reason of this is that the recuperator design did not change to compensate this flow evolution. To keep the same performances the flow augmentation would have to be compensated with a heat exchange surface augamentation too but this was not possible considering the overall dimension limitation.
Anyway, it appears that this recuperator design is not the most appropriate one, as there is a plate separation on the exhaust gas side allowing decreasing the gas side pressure drop (this is the reason why this technology choice was made) but implying also a lower effectiveness range due to a reduced heat transfer coefficient on the exhaust gas side.
If we consider normal rules for recuperator design, it appears that this type of heat exchanger i.e. air to gas heat exchanger, has to be compact i.e. without plate separation in order to equilibrate the heat transfer coefficient on both air and gas side. In this way, the energy contained in the exhaust gas can be transferred to the air with less heat exchange surface than previously and finally, the result of this optimization will be a smaller heat exchanger. Moreover, considering the high level of automation in ACTE facilities for producing the COMPACT heat exchanger range, this should also bring a significant cost saving potential.
From the abovementioned, the only way for ACTE to fulfil the optimization purposes i.e. respect the initial specification with a COMPACT recuperator is modifying its plate profile geometry from a triangular shape factor to a trapezoidal shape factor. Indeed, such profile modification will allow keeping the effectiveness target but decreasing both the pressure drop and the weight. After having studied more than 50 different plate profile, the selected one will provide the following performances considering the same overall dimensions i.e. outer recuperator core of 490 mm. the new profile allows a significant gain in the effectiveness that corresponds know to the initial specification but also a smaller global pressure drop i.e. 2.37% instead of 3.25% and also a significant weight reduction. The optimization objectives are then achieved but the final gain in the production cost will be known after knowing the raw material that will have to be used for achieving the corrosion resistance objective.
Finally, note also that this profile is compatible with the elongation limit of all the raw material taken into account by ULB in their corrosion studies. So following their results on this topic the new heat exchanger plate profile will make us able to fulfil all the project objective.
The design of the monorotor was completed within the boundary conditions of the cycle defined. Several iterations of the casting process was performed since the low mass flow resulted in a narrow channel at compressor outlet, constrained by casting to be a minimum of 3mm (POC2 achieved 4mm). At lower specific flow, the flow channels are relatively long and narrow, also a challenge for casting. The castings had a number of issues, but primarily a very poor surface finish on both compressor and turbine sides, probably reflecting an inherent roughness in the rapid prototyping approach used to make the wax masters for investment casting. Some prototypes viewed by PCA had apparent casting ‘debris’ within the channels. Although the basic premise was that the components would run as-cast with the addition of thermal barrier coatings, this was found not to be practicable. Considerable effort was expended by the SME and RTD partners in trying to machine the gas-swept surfaces, with some success on the turbine side, though this is not a viable option going into economic production. However, the compressor channel is relatively inaccessible, the gap between the shroud and hub, combined with the curvature and length of the channel, means only very limited surface dressing is possible. For the gas turbine to self-sustain, the efficiency of both compressor and turbine (the product of the two) must be sufficiently high. As normal practice, and especially for a case where efficiency is of primary importance to the economic viability of the project, the design assumed hydraulically smooth surfaces for both compressor and turbine. PCA did produce a second design with increased flow to 200g/s which may have eased some of the problems, though not to The limited test data is not of much use in calibrating the design since it is so far off-design. The tests ran at low turbine inlet temperature so there is no significant feedback on heat-transfer effects, although there were some signs that the ECOJET layout had some advantages over POC2 due to an air gap between the turbine nozzle and compressor diffuser, which gave some insulation. Although the POC2 prototype had been produced to an acceptable standard, the reduction in channel height for ECOJET has been a step too far. PCA carried out some limited studies to try to produce a design that had ‘line of sight’ through the compressor to afford some access for cleaning, as shown below, but this was not taken up by the lead partner.
In work packages 4 materials and coatings are studied, tested and selected for the two main parts – heat exchanger and monorotor. The design of the turbine is related to the capabilities of these alloys to resist to high temperatures – above 750degC and high temperature corrosive and oxidizing gases. Several HRSA and their properties have been studied and summarized in order to help the turbine designers to choose and justify their decision on the optimal alloy.
The mechanical properties related to the shift in temperatures have been presented in a spread sheet – hereby providing a clearer picture and a justification for using certain HRSA while avoiding using others. This information aims at helping on achieving the optimal alloy to be used in order to keep the best trade performance vs endurance vs cost. This trade is one of the main drivers of the whole Ecojet project.
During the research performed by DAMRC, several properties of the thermal barrier coating were analyzed. The goal was to relate the coating procedure with the final surface properties achievable in order to produce the most efficient surface from the thermodynamic perspective. A better surface will dramatically improve the performance of the turbine as it was stated by PCE (8% improvement of performance of the monorotor from Ra=10µm to a Ra=1,6µm ). Taking this issue in focus, DAMRC researched the theoretically best achievable coating surface in combination with the machining of the raw casted of the part. After this short research was complete, a report labeled 140630 Coating properties vs roughness was created.
The most relevant conclusions of this report are;
• The porosity and hardness of the coating have inverse behaviors when changing the thickness of the coating.
• The higher the porosity of the coating, the lower is the hardness of the coating.
Another relevant conclusion had to do with the roughness of the substrate (casted monorotor) and the adhesion strength of the thermal barrier coating. The adhesive strength of the coating is strongly dependent on the roughness of the substrate material. The Adhesive strength is critical for the reliability of the coating hence, it is also critical for the reliability of the Ecojet concept. If the adhesive strength is poor, the coating might experience delamination problems. These problems will lead to a poor thermal isolation of the substrate material thus; it will lead to poor thermal and mechanical properties and poor reliability. It was also presented a paragraph with the thermal isolation properties of thermal barrier coatings. After combining all these properties of the coating and substrate materials, a decision for using a nominal coating thickness of 400µm (coating material provided by TE&M; Ceria-Yttria-Stablized-Zirconium-Oxide). The next challenge while coating the monorotor is that the coating process does not allow for coating thickness tolerances better than 300µm while our nominal thickness is 400µm. This would mean that we have got a potential geometrical error of 75% of the coating thickness. From the assembly and mechanical properties, this error would make the monorotor unsuitable. So as a conclusion, the coating had to be machined before put up to service.
The machining of Oxide materials such as ZrO2 cannot be achieved by traditional machining methods because the hardness of the coating is superior to the hardness of the tool. This fact would ruin the tool almost immediately. DAMRC developed a state-of-the-art process in order to overcome this issue. This new process was able to perform the machining operations presenting results even better than the requirements of the project (roughness up to; Ra=0.95µm, Rz=8µm). Note: (a) Surface of the turbine when coated. (b)Surface of the coated-monorotor after machining ZrO2. The productivity of this process was also in focus, achieving a material removal rate up to 10 times higher than a grinding process with similar geometrical characteristics.
In work package 5 the focus in on designing and producing the heat exchanger. In D5.1 the mechanical design had been realized accordingly with the thermo-mechanical constrain due to working conditions and transient time (ramp-up & shut down):
▪ Performances assessment with ACTE semi-CFD tool
▪ Creep computation
▪ Thermal shock analysis & fatigue life time
▪ Interface definition (position and type of connection accordingly with Radijet specification)
▪ Recuperator integration in the Radijet frame
D5.2 corresponds to:
▪ The recuperator core manufacturing in the ACTE production line.
▪ The radial collector (headers) machining and mounting on the core
▪ The radial collector brazing on the core
▪ Air inlet interface machining
After manufacturing, this core was shipped for applying the most appropriate coating layer before sending the core for mounting it in the CHP system. However, due to the last event involving the closure of the lead partner Radijet, ULB had not the opportunity to deliver its recuperator for mounting and testing. As a conclusion, we can say that task 5.3 had right been completed but the recuperator has not been integrated into the gas turbine frame.
In work packages 6 the main objective is the design, casting process, coating and machining of the monorotor.
Casting processes are class-divided into the main techniques concerning their mold material into permanent mold and expendable mold processes. Permanent mold processes, e.g. die-casting, are generally used with nonferrous alloys because of their matching thermal resistance. The expendable mold processes are in advantage to cope with high temperatures and therefore are appropriate for casting steel alloys for instance. The lost wax method out performs any other casting process when it comes to superior surface finishes especially compared to sand casting. The lost wax process or investment casting process can be used to produce the most complex shapes in just about any alloy. Advantages of the process are for example liberty of shape, free choice of material, dimensional accuracy and reproducibility of mold details, high surface quality, small up to large quantities possible, flexible regarding piece weight and dimensions. Instead of wax-patterns, RP-models out of PMMA can be inserted into the process when the final design has not been determined and short development time cycles were desired.
For the ECOJET system the various casting methods were evaluated. The specification details of the monorotor were reviewed and the following specifications agreed:
1. The wall thickness must be more than 1 mm
2. The enclosed channels must be at least 3 mm in height
3. Ceramic cores in the compressor channels will give a surface roughness RA between 1.6 - 3.2 μm.
4. The compressor will be designed without splitter blade, due to difficulties with the ceramic cores.
5. The leading edge of the compressor is very important and is designed as a 1:4 ellipse.
6. The shroud must be uniform thickness all the way. 3 mm would be a good compromise.
7. The blades could be as thin as 1 mm. PCA wanted thicker root blade and thinner outer blade shape.
8. This applies to both compressor and turbine. This was a good solution for casting point of view.
9. Different materials suitable for casting were suggested: Inconel 718 and Inconel 738.
10. The design will be according to VDG P690.
The casting process involves up till 12 various tasks and decisions.
1. Choice off master pattern – rapid prototyping has the shortest development cycles and delivery times within lower acquisition costs.
2. Casting design and definition of shrinkage – see the above agreed specifications
3. 3D file format and quality - the software solid works is used and IGES-format has been specified as main interface
4. 3D model printed part material - All rp-models have been printed using a 3D high-performance printer.
5. 3D model printed part surface roughness - the surface quality of rp-models is highly influenced by the layer thickness, the grain size and the infiltration method. Attempts have been made to minimize the surface roughness of the rp-models. Two different methods have been tried. Unfortunately, all results of varying the printing angel had to be rejected. The relation between the printing angel and surface quality is disadvantageous. The best results were achieved by infiltration type 2 material with Ra = 13.93-14.99 μm. Subsequently the surface roughness has been measured on the casting. A roughness of Ra = 2.8 – 3.7μm has been detected. The shift to lower surface roughness presumably is due to sand blasting as finishing step in the production.
6. Strategy of metal flooding and solidification - The monorotor possesses very thin blades and the mold filling has to be ideally balanced. Therefore, e.g. the melting temperature and the preheating temperatures of the ceramic shell mold must be compatibly divined
7. Ceramic mold creation - The ceramic mold is created by iterative dipping and sanding cycles. After the prime cote has been finished, the refractory was filled in the channels of the monorotor manually. This step of the process has to be done very accurate to avoid air inclusions. Otherwise, during metal casting, the thin ceramic prime coat might break and metal infiltrates the air cavity.
8. Removal of 3D plastic/wax model inside ceramic mold - The dewaxing of the casting units can be conducted in two different ways. One method is flash firing and the other on is to autoclave the ceramic shell. Being environmentally friendly, BFG uses an autoclave to remove the wax out of the ceramic shell. However, the 3D-Printed lost model is out off PMMA and cannot be removed by autoclaving indeed.
9. Casting strategy: Air casting – vacuum casting selection – In the ECOJET project it has been decided to use Inconel 718 as casting material. Inconel 718 is a super alloy based on nickel and was especially developed for high temperature applications e.g. needed for jet engines.
10. Casting technic – ideally the pouring is continuous, rapid and non-turbulent so especially the very narrow blade border areas of the mold are filled without could laps. All these criteria can be satisfied by the roll-over casting technology, being a quite simple construction. During the time the basic material is melted, the furnace is in a normal position. To separate the melt from atmosphere argon comes into operation. When the casting temperature is reached, the pre heated and isolated ceramic shell will be mounted with its pouring gate on top of the furnace. For casting the furnace is turned upside down now. This leads to casting times of approximately 2 - 3 seconds and also a sudden mold filling. This technique has been successfully used to cast the monorotors.
11. Removal of ceramic mold - To remove the ceramic of the casting first off all 2 dimensional areas of the shell will be eliminated by a rocking motion with a forge. Afterwards still remaining ceramic on the casting surface will be removed by shot blasting, however shot blasting/sand blasting is not appropriate to withdrawal ceramic out of the turbine channels from the monorotor. For the final finish, the parts have to be sand blasted at the end.
12. Quality control of the part - The first inspection of the casted monorotor has been executed by a visual product inspection. Afterwards the casted monorotor has to pass a crack detection test.
It is possible to cast the Monorotor in an open casting process using the rollover method. However, the process safety by the rapid prototyping process is not very well, the main problem is the introduction of the ceramic mass into the inner spaces of the Monorotor. This can be done only in a manual way with a lot of manpower. Each model is unique. It cannot be determined whether air bubbles are present. By one good part come circa three bad parts. A series production will have a better process safety.
In work packages 7 all these parts are assembled with the rest of the system and test facilites are defined for the test. Specification of the final test was done. Test site specified and selection of measuring devices done. Due to severe problems on delivering good quality casted monorotors along with severe assembling problems several delays occurred. RadiJet performed several tests. Conclusion: it is not possible to have a self-sustained system due to surface roughness of the turbine blades. Combustion system gave problems in the configuration with nature gas as fuel. The bearing system, inverter and the data log system all works fine. Feb16 test run: power supported at 25-30.000rpm by manual control of fuel and rpm, the turbine not installed in the recuperator. NLR worked hard to build/adapt testing facilities to the need of RadiJet and was ready to perform the tests on receipt of the system. The shipment was planned several times but postponed with the consequence that the system never was shipped to NLR. With the closure of RadiJet it was no longer feasible to assemble and ship the system for test
Project Results:
The ECOJET project designed and developed a new radial gas engine.The engine has a 10 kWe micro gas turbine power generator in an innovative system solution tailed for environmentally friendly micro CHP (Combined Heat and Power) applications for the residential sector.
The main project results involve the design and development of
- Radial micro gas turbine consisting of a) monorotor + b) heat exchanger
- Method and matrials to produce the monorotor
- New heat exchanger
- Coating materials for the heat exchanger and monorotor.
Potential Impact:
The ECOJET project designed and developed a new radial gas engine.The engine has a 10 kWe micro gas turbine power generator in an innovative system solution tailed for environmentally friendly micro CHP (Combined Heat and Power) applications for the residential sector. The EcoJet project designed and developed solution differentiates from its competitors by the simplicity and cost effectiveness of its innovative design. In addition to ease of manufacturing and comparatively low production costs, the design – with only one rotating part – enables very low cost of ownership due to its operational features and low maintenance cost. Our approach will allow overcoming the barrier to produce a cost-effective, environmental friendly and simple product, The EcoJet solution thus has a range of unique selling propositions to a very large group of potential end-users – manufacturers of micro CHP systems and, ultimately, households.
The project results are expected to have a big impact on contribution to the EU energy efficiency policy. Furthermore, the overall impact of EcoJet will place the EU SMEs in a highly competitive position within the cogeneration market with dominant players currently coming from non-European regions (mainly America and Japan), thereby creating and protecting job, increasing revenues to allow further investment and technological developments
Ecojet aims to base its marketing strategy for the technology primarily on the strong and growing market for distributed micro CHP. At the end-user level, it is estimated that micro CHP can help a consumer save close to 60% on their electricity bills.
The focus parameters for the microCHP unit have been defined as the following:
➢ Electrical efficiency: >21%
➢ Total efficiency:> 83%
➢ Noise enclosure:<45 dbA @ 1 meter
➢ Noise exhaust:<55 dbA @ 1 meter
These parameters being the key selling parameters when selling a Micro CHP system.
The dissemination activities have mainly been done on fairs & exhibitions, newsletters and via posters.
List of Websites:
The project website was closed by the lead partner as this partner left the project 1 month before closure.
The partners can be found here:
- TEandM www.teandm.pt Contact person Mr. Ricardo Alexandre [ricardo@teandm.pt]
- ACTE www.acte.-sa.be Contact person: Mr. Sébastien DUBOIS [sebastien.dubois@acte-sa.be]
- FRICHS A/S – www.frichs.dk – Contact person. Mr. Ove Munch [omu@frichs.com]
- Danish Advanced Manufacturing Research Center (DAMRC) – www.damrc.com – Contact person Mrs. Charlotte F. Ilvig [chi@damrc.com]
- Stichting national Lucht- enRuimtevaartlaboratorium (NLR) – www.nlr.nl – Contact person: Dr. ir.A.B. (Arjen) Kloostermann [arjen.kloostermann@nlr.nl]
- BFG Feinguss Niederrhein GMBH – www.bfg-niederrhein.de – contact person Mr. Stefan Daub-Klose,CEO
- PCA Engineers Ltd. www.pcaeng.co.uk – Contact person Dr. C J Robinson chris.robinson@pcaeng.co.uk
- Universite Libre Bruxelles (ULB) ww.ulb.ac.be – Contact person: Prof. M.P. Delplancke-Ogletree