Final Report Summary - BECOOL (BI-PHASE COOLING SYSTEM SUITABLE FOR POWER ELECTRONICS DEDICATED TO MORE ELECTRICAL AIRCRAFT)
The purpose of the more electrical aircraft is to reduce the power consumption of the future airplane by decreasing the weight of embedded equipment. To achieve this purpose, it has been shown that mutual use of part of these equipment to reach a general use of on-board electrical power should achieve this purpose. This is the concept of the more electrical aircraft. Naturally, this must significantly increase the amount of electrical power consumed as the power converter efficiency is still fewer than 100%. Therefore, the thermal dissipation will raise levels that are not assumed by the current embedded cooling system design.
The BECOOL project has been planned on a 2 years development program with the aim to provide a complete cooling system able to absorb a specified amount of power, around 1kW, for a nominal working temperature of the future power converter, around 100°C. The equipment involved in the diphasic loop should be constraint to a maximum size and a maximum weight and be able to work in a harsh environment in terms of acceleration and temperature.
ATMOSTAT has developed, thanks to its internal diphasic programs, a new concept of diphasic cooling system to reach more integrated cooling device. Technological bricks required to design a generic diphasic cooling system have already been identified and have reached their own TRL level but the application, coming from the link between all these equipment specifically designed to fit requirements, have to progress also in its own TRL level.
The challenge of this program is to design a complete cooling system to progress significantly in the maturity level of the application and optimize its size and performance to be compliant with the Topic Manager specification.
The most efficient cooling technologies compared to their embedded size and weight are those working in closed circuit. The simplest configuration used is heat pipes, but we can also find heat sinks, capillary-pumped heat pipes (CPL), LHP... In spite of good performances, these technologies keep major constraints, caused by their architecture: the heat flow must be moderate otherwise the hot surface will be dried and then no longer cooled, drying troubles can occurs because of gravity or during acceleration phase due to the insufficient amount of pressure created by the capillarity mesh, power density is limited due to drying risks, distance of heat transportation are limited and finally unexpected unpriming during duty cycle...
The architecture and technology of the 2-phase loop mechanically pumped that ATMOSTAT has proposed can bring answers to the troubles thanks to the embedded pump, forcing the liquid circulation, and its efficiency at high power density thanks to the innovative technology of its evaporator allowing to push away the drying limits.
The current state-of-the-art, highly represented by the diphasic capillary pumped loop technology, has welcome, in the “diphasic family”, a new concept of diphasic loop: the 2-phase loop mechanically pumped developed by ATMOSTAT. At the end of the project, the advancements of the current state–of-art are coming from the fact that we have moved from the concept of a diphasic loop application, at low TRL level (0 to 1-2) at the beginning of the project, to the level 3 at the end of the project by the experimental proof of the main characteristics and major performance of the loop.
The challenge that might not have been addressed during the program, because of the delay of the project due to the manufacture of the equipment, coming from the transitional regime that couldn’t have been designed properly without preliminary tests. Transitional regimes of the power converter are quite difficult to analyse without the Topic Manager expertise. An ideal transitional behaviour should be defined by the Topic Manager and explained to ATMOSTAT to be able to take its behaviour into account in the future design of the ATMOSTAT diphasic loop.
The maturity of our concept has reach a level to let us consider, now, industrial applications and potential commercialisations in different filed as: defence, civil aeronautics and earth transportation application (train, tram, car), production of renewable energy, telecom (radar), even space application might be considered if we can prove that our pump is reliable enough.
Project Context and Objectives:
Progress associated with the more electrical aircraft relies on the principle of a decreasing consumption of plane obtained by an overall reduction of the weight of equipment, by mutual use of part of these equipment and by growing their efficiency. It has been shown that general use of on-board electrical power might achieve this purpose. This is the concept of the more electrical aircraft.
However this improvement involves using more and more powerful electronic devices as electric power conversion or generator, with higher and higher densities of power. As these equipment have performances less than 100% (about 85-95%), the corresponding heat flow requires heavier and heavier cooling systems, with cold sources providing wider and wider difference of temperature between the fluid and its environment.
Standard cooling systems by coolant or air flow are no longer sufficient and efficient regarding the requirement of weight embedded increasingly restrictive. The changing phase of an evaporable liquid in the useful temperature range is a physical process that consumes a great amount of heat. It is therefore a good way to cool hot source with a reduced amount of liquid and temperature gradient.
The space sector already uses this kind of methods to cool electronic systems installed on satellites of communication. The simplest configuration used is the heat pipe. This is a hermetically vacuum-sealed cavity, a part of which remains cold due to a radiant panel and is filled under vacuum with an evaporable fluid that cannot freeze at low temperatures. The fluid used is ammonia. As the satellite is weightless, the liquid fluid might float in the cooling cavity. But the surface dampening process ensures a more stable position by supplying the lowest power: the liquid is then distributed evenly on the internal surface.
If a hot spot appears, an evaporation flow is generated and the vapour pressure depends on the temperature of the hot spot. Since this vapour pressure is much lower for cold spots, there is a flow of fluid vapour that evaporates over hot spots and condensates over cold spots.
However this simple principle is only compatible with a weightless environment. Moreover, the heat flow must be moderate; otherwise the hot area will be dried and then no longer cooled.
For greater flows, two-phase loops have been developed, again in space applications. In such loops, hot and cold spots are separated and more distant one from the other. They are linked by 2 channels: a liquid channel and a vapour channel. To ensure liquid return to the evaporator, capillaries carry out pressure gradients in the liquid section. Pressure gradients together with the evaporation pressure causes a dissymmetry allowing fluid circulation in a gas form in one section and in a liquid form in the other section. However, these pressure gradients are low and under high acceleration, « blocked » conditions may arise: liquid is not supplied in the vapour chamber which becomes dry. The cooling of the hot spot is not ensured anymore that might cause dangerous failure of the electronic. To cope with these problems, aided pumping is used to overload the capillary from a « short » circuit. It is called a hybrid loop. In all cases these devices are still incompatible with accelerated environments. They cannot transfer heat very far from the source and are limited in terms of power density.
The aim of ATMOSTAT proposal is to provide a Mechanically Pumped 2-Phase Loop (MP2PL) which can:
• Be as compact as possible to limit the mass embedded and the volume occupied.
• Work in such accelerated environment;
• Transfer a great amount of power and;
• Reach a high power density without drying the exchange surface to provide any increase of the working temperature of the electronic component.
At the beginning of the project, the specification of the thermal management of the future power module has been defined in relation with the Topic Manager. Bibliographical researches have been done and registered in the "trade off study" deliverable. This activity confirmed several things: our concept is completely new and therefore no off-the-shelf product exists, it has the potential to solve main troubles of actual diphasic loop system described in the state-of-the-art. Then a preliminary design has been done to provide basic design of the solution proposed for the project. The detailed analysis of selected solutions has then described the requirement in details and expresses it in terms of performances. An intermediate tests phase has been done to tune the parameters of the loop as the volume flow rate of the pump, the working temperature of the expansion tank and the efficiency of the condenser. To do these tests we used our own thermal tools. These tools are able to simulate thermal power spreading of a power module. These tunings allowed ATMOSTAT to present to the Topic Manager these preliminary results and validate them with him in order to pass the 3rd milestone. Thanks to these tests we know perfectly the behaviour of our system.
The final hardware tests have been done with real power modules coming from the Topic Manager facilities. Steady states power tests have been partially passed successfully. We have reached a global amount of power of 600W and keep stable the temperature of the sole under 90°C, temperature limit specified by the Topic Manager. Because of an excessive temperature gradient at the contact interface between the sole and the evaporator due to the necessary added mechanical plate and thermal grease between these equipment to jointly fasten them, the extreme power limit (1500W) for transitional regime have not been reached at the maximum sole temperature (90°C). ATMOSTAT tuning tests have shown that at that power (1500°C), the sole’s temperature have been measured at about 110°C. The transitional regime tests have been done and highlighted that the evaporator had a rapid response to reach the working temperature of the power module when charged. The tests protocol in that type of regime had its limit regarding the quick thermal answers to our system and has to be reviewed to be more significant.
During this 2-years development program, ATMOSTAT got results concerning different main points:
• At the beginning of the program, ATMOSTAT had a concept and equipments designed to prove the feasibility of the concept without performance characteristics due to application specifications.
• At the end of the program, ATMOSTAT has proven that its concept of new diphasic loop architecture can thermally manage the power converter characteristics specified by the Topic Manager.
• Equipments have been designed to reach the performance requested. The global demonstrator has passed part of the thermal tests and shows a great potential of optimisation regarding its mass and size.
• Regarding these results we can say that the TRL level 3 has been reached by the global application system.
Compared to the state-of-the-art at the beginning of the project, the progress is significant regarding TRL level of the application. We moved from 1 to 3 in the TRL level.
Compared to currently available solutions in the diphasic field, the diphasic loop concept of ATMOSTAT have made its first steps to prove its functionality and show first interesting performance allowing us to list new way of improvement design to be able to face high density power. The performance of our system in high density power will be obvious the next development step to compare the major performance of the diphasic loop between the “capillary pumped” one (state-of-the-art) and the “mechanically pumped” one (Atmostat development).
The project foregrounds are composed by the technical characteristics in steady and dynamic state mode of our innovative evaporator mounted on a real power converter. These characteristics are represented by the behaviour curves which indicate the maximum power evacuated by the cooling system versus the maximum temperature at the mechanical interface with the power converter. ATMOSTAT has been able to define a list of technical modifications to improve the behaviour of its cooling system and also to confirm the potential final size and weight of the cooling system. The Topic Manager is now able to manage the power delivery of its power converter versus the capacity of our cooling system at maintain its temperature at a functional value. He also can start to design the technical environment of the power converter thanks to the definition of the size of our system coming from the foregrounds of the project.
The main socio-economic potential impact are coming from the creation of a new industry or the reinforcement of the existing one by hiring development and tests engineers, manufacturing technicians and workers. Qualified people should necessarily be trained to manage the complex technics of this new technology and ensure maintenance capability. If we estimate to 20 (estimation), the number of power module mounted on a plane and 500 plans manufactured per year, we can estimate to 10 000 the potential number of diphasic loop manufactured per year in the only aeronautic sector. If we know that the train transportation but also the renewable energy, the telecommunication with the radars and spatial field are potentially very interested by this type of diphasic cooling system, the potential socio-economic impact is the increase of the need of employment and qualified people and so one. The Topic Manager express also his strong interest in our diphasic solution by the way that it can reduce by 3 the mass embedded of its current, and used, cooling device.
The wider societal implications could come from the habits of the electronic designer dealing with the thermal dissipation of the current power converter (or any electronic cards carrying high dissipative electronic components).
These habits, used in the thermal dissipation’s current design, forced the electronic designer to dissipate the thermal power in a wider surface as required by the electronic size, just to be able to avoid any risk of high temperature gradient due to the high power density. Therefore, the designer build the electrical architecture of the power converter by spacing its component on a larger surface. When the designer will take into account the capacity of our system to evacuate a large amount of power on a lower surface both parties will be able to deeply integrate electronic and its innovative way to cool.
Global results coming from the diphasic activities of ATMOSTAT have been disseminated at the DECIELEC conference in March 2015 (included in the presentation of the Topic Manager) and at the IMAP’s conference at La Rochelle in February 2016.
List of Websites: