Will the Lilium Jet Work? A detailed analysis by an independent expert.

The Lilium Jet is arguably one of the  most iconic and beautiful eVTOL aircraft. However, it is undoubtedly the  most controversial concept, too. On the one hand, it has won numerous  awards, such as the Red Dot Design Award called "Best of the Best"... ... LinkedIn's top rank for startups in Germany, and the IF Gold Award in the  discipline of professional concept. But on the other hand, numerous aviation  experts are skeptical about the Lilium Jet. They question the feasibility  of Lilium's performance targets because they are significantly  higher than competitors. In 2020, the Lilium Jet was marketed as  a five-seater with a 300-kilometer range. A feat that the company  attributed to its unique design— an aircraft with canards, deep  propulsion and airframe integration, and so-called electric jets. In January 2020, Aerokurier,  a German aviation magazine, dismissed Lilium's market claims  based on a technical paper written by an aircraft engineer  who chose to remain anonymous. Aerokurier concluded that Lilium's  promises regarding a range of 300 kilometers and a flight duration of up to one hour   have been completely and  doubtlessly unmasked as unfeasible. Aerokurier was subsequently awarded  the Hugo Junkers Preis for this report. In response to the critics, Lilium  published its own technical paper to defend the feasibility of its concept  and aircraft performance targets. The paper was authored by one of the founders  and reviewed by five independent experts.  Lilium further argues: "Because of its  novelty, some commentators have struggled with properly assessing the Lilium Jet. These misconceptions often lead to errors  when third parties estimate and comment on our aircraft's performance potential." Meanwhile, as these controversies continue,  Lilium's funding has increased considerably. According to Crunchbase, the total funding  amount is, as of now, 945 million US dollars. Additionally, Lilium has built one  of the best teams in the world, featuring top talents from Airbus,  Rolls-Royce, and Honeywell to name a few. So, the question arises, will the Lilium Jet work? Or will it fail? To deeply understand this issue and the debates, we must first cover the fundamental  principles of electric aircraft design, and examine the benefits and  drawbacks of Lilium's concept. Then, we will carefully review the  technical papers by Aerokurier and Lilium, to assess who is right and who is wrong. Finally, we will examine Lilium's history to  understand its trajectory into the future. My name is John. I'm an aircraft engineer from Berlin, Germany. I am the author of this video  and your presenter for today. Lilium's marketing has raised  controversy as early as 2016. Central to the debates is whether Lilium's unique   aircraft concept enables  high aircraft performance, such as the 300-kilometer  range for five occupants. For an in-depth analysis, we need a methodology  for calculating an eVTOL aircraft's range, so we can then assess the feasibility  of Lilium's performance targets. Hence, I will first give an overview of the  fundamentals of electric aircraft design. I will then explain how to maximize the  range and the design trade-offs involved. Finally, I will provide a qualitative  analysis of Lilium's aircraft concept. We estimate the range from  an aircraft’s flight path. This screenshot, taken from Lilium’s website,  shows the Lilium Jet’s flight path and tells us, amongst other things, how long  the hover and cruise flights are. This information is essential because the  cruise flight contributes towards the range,   whereas hover flight reduces it. So how can we quantify their impact on the range? For this, we need to analyze power and energy. Below, we are shown the power  consumption for each flight segment, and the area below the power  curve is the energy consumption. The range ultimately depends on how much energy  is consumed and used to fly the aircraft. For example, the range depends on how much  energy is allocated toward the cruise flight. It also depends on the aircraft’s  drag and energy efficiency. These engineering concepts can seem  abstract and complicated at first, but they are quite easy to understand  once we have put them into a framework. So let us build this framework by starting  our analysis with a simplified flight path. This figure shows Lilium’s original flight path. We will now remove the transition, climb, and  descent flight segments to simplify the problem. As a result, we only have  hover flight and cruise flight. Next, we want to determine how much energy  is allocated toward the cruise flight. Because this helps us determine the range. We can find an answer by using an energy balance, which gives us this equation: the total energy  consumption equals the energy for hover flight   plus cruise flight. The energy balance already gives us two insights. First, we can increase the range by  allocating more energy for the cruise flight. Second, we can allocate  more energy by, for example,   adding more batteries to increase the total energy or lowering the hover time to reduce the  energy consumption for hover flights. These were two high-level insights. Let us now get a more in-depth understanding by examining the cruise flight more deeply. The cruise flight is most interesting because  this is where the aircraft flies the fastest, spends most of its time,  and does most of the range. In fact, we can use the standard definition of   ‘work’ to gain insights into what  we can do to maximize the range. For this, we need to contextualize the general  definition to get the following equation: the aircraft’s propulsive work  equals thrust times range. As the equation shows, we can increase the range  for a given energy budget by reducing the thrust. And we can reduce the required thrust  by lowering the aircraft’s drag. We know that the battery provides our energy, and the propulsion system converts this  electrical energy into propulsive work. We also know that this conversion  process is not 100% efficient, so an efficiency factor of  less than one is needed. Therefore, our next insight is  that we can increase the range by making the propulsion system more efficient at  converting electrical energy into propulsive work. Finally, if we combine the previous  insights, we get one useful equation: the electric aircraft range equation. It shows that we can increase the  range by increasing the battery energy, increasing the propulsion system’s efficiency, and reducing the aircraft’s drag. Maximizing the range of an  eVTOL aircraft is crucial   because it unlocks tremendous business potential. However, design tradeoffs limit the extent to  which the range can be reasonably maximized. So, I will now break down these three  parameters – energy, efficiency, and drag – for valuable insights into these design tradeoffs. Let us start with the battery energy. The battery energy depends on the  battery's energy density and mass. Therefore, we can increase the range by increasing  the battery's energy density, mass, or both. We can also draw more energy from  the battery by using more capacity and discharging it to a lower state of charge. Increasing the battery mass can result   in an overweight aircraft and  problems with certification. For example, the EASA requires these  eVTOL aircraft to be less than 3175kg. At some point, growing the aircraft’s  weight is no longer viable. Therefore, the other design  option is reducing mass elsewhere   to compensate for the  additional batteries we want, thereby preventing the aircraft  from becoming overweight. For example, we can exchange one  passenger for an extra 100kg battery pack, thus trading that passenger  for a higher aircraft range. However, there are several design  tradeoffs for the second option. To understand them, we must first break down  the aircraft’s mass into three categories: namely payload, battery, and empty aircraft mass. We then use the concept of mass fractions, which  is easily explained by the following example. A battery’s mass fraction is the percentage of  the battery mass to the aircraft’s total mass. This parameter helps us understand and  compare different aircraft concepts. For example, you would expect a high-range  concept to have a high battery mass fraction. And what are the design tradeoffs  for having more batteries? As this figure explains, the payload  generates the revenue per kilometer, the battery determines the maximum range, and the empty aircraft mass fraction  affects the aircraft’s functions,   performances, and design margins. Therefore, an optimal aircraft design  must balance these different needs, which all compete for the mass budget. And thus, the design tradeoffs for  more batteries are, for example, lower revenue per kilometer or  lower redundancy for flight safety. Following our previous analysis, we can see  why eVTOL aircraft have low payload and range. Their low aircraft performance  is due to three penalties:   electric propulsion, VTOL  capability, and safety requirements. For example, the battery’s low energy density  means there is less energy for a given mass budget   and, hence, a lower range. Furthermore, the VTOL capability  requires additional systems,   which increases the empty aircraft’s mass. Similarly, safety requirements  need backup systems and components, further increasing the aircraft’s empty mass. And a larger empty aircraft mass results in  smaller mass budgets for the payload and battery. As we just saw, there are no easy design  tradeoff choices for the battery mass. Therefore, besides increasing the battery mass,   what else can we do to  improve the aircraft’s range? We will now turn to the other  options for increasing the range. So let us examine the  propulsion system efficiency. The propulsion system consists of batteries,   other minor electronic components,  electric motors, and rotors. The propulsion system's efficiency is the  total of these component efficiencies. Therefore, if we want to maximize the  propulsion system efficiency for a higher range, we must understand the design  tradeoffs for these components. So let us start by reviewing the rotor  efficiency and its design tradeoffs. Rotors come in many different shapes and sizes,  and they can even have ducts, as in Lilium’s case. However, all rotors operate on the same  basic principles, regardless of their design. And there are two fundamental  principles that govern rotor efficiency. The two fundamental principles are as follows: The Froude efficiency  determines the inviscid losses. The viscous profile efficiency  determines the viscous losses. A more detailed explanation  is given in the figure. Now, let's skip ahead to the design  tradeoffs for these two efficiencies. We can increase the Froude efficiency in two  ways, which all have their design tradeoffs. First, we can increase the rotor diameter,   though this can create issues  in the aircraft configuration, such as rotor tip clearances. Furthermore, an increased rotor diameter  might mean our rotor and motor become heavier, reducing the mass budget  for our battery and payload. The second option is to increase  the aircraft’s velocity. However, this option requires  a larger electric motor since propulsive power increases with  the aircraft’s drag and velocity. A greater power requirement  implies heavier motors, so, as before, the mass budget for  the battery and payload is reduced. And there is another design tradeoff. Due to the higher design cruise speed, the rotor inflow during cruise flight  differs more greatly from hover flight, which can result in a lower  viscous profile efficiency, especially for rigid rotors. We can increase the viscous  profile efficiency in two ways. First, as discussed in the previous slide,  the viscous profile efficiency gets penalized by the different rotor inflows  during the cruise and hover flights, especially for a rigid rotor. We can solve this problem by  using a variable pitch rotor, which ensures the relative wind meets  the rotor blade at an optimal angle for different advance ratios. Our second option is to reduce  the number of rotor blades because rotors with fewer blades  have lower viscous profile drag and higher viscous profile efficiency. As this figure shows, a two-bladed  design is the most efficient. But this figure also reveals that rotors  with fewer blades generate less thrust, which can be a problem for eVTOL aircraft since they require enormous  thrust for vertical takeoff. Hence, the rotor design must  balance various design tradeoffs, and maximizing the rotor  efficiency for a higher range will create costs elsewhere. OK, we just reviewed the  design tradeoffs for the rotor. Next, we can look at the electric powertrain, which consists of batteries,  minor electronics, and motors. The electric powertrain’s efficiency  depends on the motor we select and the electrical resistance  we have in the circuit. First, let us talk about electric motor selection. We can achieve high motor efficiency  by choosing a high-torque motor. However, high-torque motors weigh more, so we must reduce mass elsewhere to prevent  the aircraft from becoming overweight. Therefore, design tradeoffs exist  between a motor’s efficiency and mass. As these figures show, we can select  an 80% efficient electric powertrain or a 50% efficient alternative. But, as discussed, the more efficient  option would result in a smaller mass budget for the batteries and, thus, reduce some of  the range gained through its higher efficiency. The next discussion point is the  electrical resistance of the circuit. We can model the propulsion system  as an equivalent electrical circuit and the motor shaft power as an electrical load. We can then find the following results  using standard circuit equations. First, we can improve efficiency  by increasing the load resistance. However, a higher load resistance  results in a lower shaft power, which can be incompatible with  vertical takeoff and landing. Second, we can improve efficiency  by decreasing the total resistance and keeping the load constant. This approach can be more desirable  because it does not reduce shaft power. However, there are other design tradeoffs. For example, we can decrease  the motor winding’s resistance by using active cooling,  better electrical conductors, or larger cross-sectional areas for the wire. But, as you can imagine, these options increase  the motor's complexity, cost, and weight. We just covered the propulsion system efficiency. Let’s now cover the last parameter: drag. Our aim here is to minimize  the total drag and, hence,   the thrust requirement for the cruise flight, making the aircraft more  efficient and able to fly further. We can approach this problem  by first categorizing the drag and then looking at ways  to optimize each category. I have listed some ideas for  drag optimization in the figure. For example, we can reduce the induced drag  by increasing the aspect ratio of our wing. A case in point is the glider,  which has a high aspect ratio and is strongly optimized for low drag. However, having such a high  aspect ratio for an eVTOL aircraft is not possible due to numerous reasons. For example, an eVTOL aircraft can  have several large propulsion units mounted on its wings, which  generate enormous forces,   bending moments, and torques during hover flight. The wing structure must be designed accordingly, but increasing the aspect ratio makes the wing  weaker, more flexible, and more prone to failure. Hence, an eVTOL aircraft’s  wing cannot have the same   aspect ratio as a glider for structural reasons. OK, so we have just covered all three parameters  contributing to an electric aircraft's range: energy, efficiency, and drag. We saw that optimizing the range  involves many design tradeoffs. And while it is possible to optimize it  strongly, we must be careful of the consequences. After all, safety and noise  are also critical for success. So we need a balanced solution. We also discussed the performance  penalties inherent to eVTOL aircraft. Specifically, electric  propulsion, VTOL capability,   and high safety requirements lower  the aircraft's payload and range. For this reason, the battery is considered  a key enabler of eVTOL aircraft performance. For example, increasing the battery's  energy density can extend the range   without conventional aircraft design tradeoffs. However, we must remember that the  battery must be a balanced solution, too. So we should not let hope inflate our  expectations of the battery's energy density. Finally, besides the battery, many  other parameters could be improved. For example, our previous  analysis showed that motors   with high torque-to-weight ratios  are just as crucial for the range. Since eVTOL aircraft are new, we have  little history and collective experience regarding their design,  manufacture, and operation. Hence, the number of different eVTOL concepts   today shows that there is no  universally accepted solution to this novel and complex optimization problem. With this in mind, what is Lilium’s  approach to solving this challenge? Let us now analyze Lilium's concept. Lilium’s concept is unique for three reasons:   a canard configuration, electric  ducted fans with high disk loading, and thrust vectoring for aircraft control. We will only look at the ducted fans today  because they are essential for range estimation, whereas the other two aspects are more  important for aircraft control and stability. The duct can have several benefits,  such as suppressing noise, protecting the rotor from foreign object damage, and shielding passengers from a spinning rotor. Additionally, some people find  the duct visually attractive, an advantage that cannot be underestimated. Finally, the main benefit of a duct is that it  can increase rotor efficiency for two reasons. We will examine one of them: namely, how  the duct improves the Froude efficiency. If we compare an open rotor to  a ducted fan, we will find that the ducted fan only needs 71% of the open  rotor’s power to generate the same thrust. In other words, the ducted  fan is 41% more efficient. This result agrees with Lilium’s  statement on its Technology Blog: ‘During hover, ducted fans  have roughly 40% increased   efficiency compared to an open propeller...’. The duct can increase the Froude  efficiency in the following way. First, it can guide the airflow and remove  the vena contracta from the slipstream. As you can see from the figures,   the ducted fan does not have a vena  contracta, but an open rotor does. As a result, the duct can increase the  mass flow and reduce the exit velocity. Finally, lower exit velocities yield higher  Froude efficiencies, as this equation shows. So, the duct increases the Froude efficiency  by effectively acting like a larger rotor: increasing the mass flow and  reducing the exit velocity. The previous comparison  used the same disk loading. However, the Lilium Jet uses a much higher  disk loading than all other eVTOL concepts. For example, this figure from Lilium’s  Technology Blog shows that the Lilium   Jet’s disk loading is more than ten  times that of a tiltrotor concept. The previous efficiency benefit no longer applies   if Lilium’s disk loading is  higher than an open rotor’s. Consequently, the figure also shows that Lilium’s   concept is 50% less efficient than  a tiltrotor during hover flight. Thus, a natural question arises: why did Lilium forgo the duct’s efficiency  benefits and choose such a high disk loading? The answer has to do with design  tradeoffs for ducted fans. The Lilium Jet uses the duct as an  engine nacelle, lifting surface (wing), and control surface (aileron and elevator). The duct also provides structural  support for the rotor and stator. Combining several functions into the  duct makes it more difficult to design. For example, the duct design must  take care of more design constraints, and the viable design space becomes smaller. If the duct size is small, some weight and  surface area (i.e., drag) can be saved. However, the consequence is a higher disk loading. Lilium has considered these design  tradeoffs and selected their approach. So, there must be a design penalty worse than  high disk loading and low hover efficiency. What is it? What happens  if the duct size is large? We can find an answer in Lilium’s Technology Blog. From the highlighted text, we can guess  that ducted fans do not scale well because structural challenges and  aerodynamic drag increase with duct size. Thus, using ducts for VTOL aircraft can result in  high disk loading because the ducts must be small, and the thrust requirement is high. Conventional airplanes with ducted fans  do not have high disk loading because they do not need to do VTOL. As a result, their Froude efficiency for  takeoff can be much better than Lilium’s. However, ducts provide another benefit... consider an open rotor with high disk  loading and inadequate fruit efficiency Consider an open rotor with high disk  loading and inadequate Froude efficiency. As we know from before, we can use  ducts to improve its Froude efficiency. Therefore, ducts can enable rotor  designs that have high disk loading. There are several benefits of high disk loading. For instance, a high disk loading  means smaller rotor diameters,   which can lead to lower rotor torque. Hence, ducts can potentially remove  gearboxes or some motor mass. Furthermore, smaller rotors can  benefit the aircraft configuration. For example, it can enable low-  and mid-wing configurations. These configurations can have lower center of  masses than the popular high-wing configuration, potentially shortening the landing gear  and lowering its drag contribution. In addition, the mid-wing configuration  can also have a lower trim drag since the thrust line runs closer to the  center of mass than the other configurations. However, the downside is that conventional  low- and mid-wing configurations can have poor cabin access, creating  a problem for air taxi services. Lilium solves this problem  with a canard configuration,   which, combined with small ducted fans, allows excellent cabin access for the passengers. Furthermore, as we saw earlier,  the ducts can compensate for the   drawbacks of small rotors and high disk loading by improving the Froude efficiency. But we must check if this compensation is enough. Lastly, ducts can improve the boarding experience  by shielding people from spinning rotors. We just covered the benefits of high disk loading. But, of course, there are drawbacks, too. The main one is that it results in  poor rotor efficiency for hover flight due to more inviscid losses and,  potentially, viscous losses. In other words, high disk loading  results in poor static thrust efficiency. But why? Why does high disk loading  result in more inviscid losses? Well, due to the same principles  that govern Froude efficiency: lowering mass flow and increasing the downwash  speed results in more inviscid losses. Increasing the disk loading results in higher   downwash speeds and lower  static thrust efficiency. We can see from the equation  at the bottom left corner, that increasing disk loading,  T/A, results in a larger power. Furthermore, as mentioned earlier, higher  disk loadings mean higher downwash speeds, achieved by either spinning the rotor  faster or raising the thrust coefficient. Unfortunately, higher blade tip  speeds usually lead to more noise, which limits this design  option for Urban Air Mobility. So we must look at the alternative: using  rotor designs with larger thrust coefficients. However, such rotor designs  usually come with more blades, and their disadvantage can be lower rotor  efficiencies due to greater viscous losses. Lilium has chosen the second  option, as shown by the next slide. As you can see from the picture, Lilium's ducted fan has over 25 rotor blades, whereas a typical number  for an open rotor is only 5. Lilium needs more blades because its design has  a high disk loading and low blade tip speed. However, as discussed earlier, more blades  can lead to lower rotor efficiencies. And due to the high disk loading  and low viscous profile efficiency, the Lilium Jet requires tremendous  power for hover flights. What are the consequences of such  enormous power demands for hover flights? Again, there are pros and cons. First, let us look at this screenshot  taken from Lilium’s website, shown earlier in the presentation. As we said previously, this figure tells us  the power consumption for each flight segment. And according to Lilium, the hover power is  about ten times higher than the cruise power; thus, there is a huge power discrepancy  between hover and cruise flights. In other words, a very powerful propulsion system   is needed for the enormous  demands of hover flight, while only a fraction of that power is  necessary for all other flight segments. Such powerful propulsion systems can  have significant benefits and drawbacks, which we will examine next. One benefit is that the aircraft  becomes very maneuverable. We know from the previous  slide that the Lilium Jet   has an excess of ~2000kW during its cruise flight. This excess power can be  converted into more thrust, benefitting the aircraft in numerous ways. For example, the Lilium Jet  can accelerate more quickly, leading to shorter transitions  from hover to cruise flight. In addition, more thrust can  offer better angular acceleration, improving aircraft control. Lastly, additional thrust  can be used to climb faster. In short, the benefits of a large  propulsion system are similar to   having a car with a big and powerful engine: the vehicle becomes very maneuverable. However, there are drawbacks, too. Obviously, more powerful propulsion  systems are generally heavier. And as mentioned earlier, this reduces the  mass budget for the payload and battery. Additionally, larger hover powers mean  less time can be spent in hover flight, before the battery runs out of energy. And hover endurance can be important  for ensuring safe landings. Finally, the enormous power  discrepancy mentioned earlier creates significant challenges for  the rotor and battery designs. These two drawbacks are often not  fully understood by the public, yet they are critical. First, let us talk about  the rotor design challenges. Larger power discrepancies  between hover and cruise   flights make the rotor design more challenging, and likely penalize the  viscous profile efficiency. Why? We know that if we want to increase  power, we need to spin the rotor faster. Therefore, the rotor is spun rapidly at hover  flight but much more slowly at cruise flight. Additionally, the aircraft’s forward  velocity at hover flight is zero, whereas, at cruise flight, it can  be up to 300km/h for Lilium’s case. These two aspects combined mean the  relative wind meets the rotor blade at drastically different angles of  attack for hover and cruise flights. And, as we know, an airfoil operates optimally  only at a narrow range of angles of attack. Conventional design approaches use mechanisms  to control the blade pitch and, thus, solve the problem created by  large power discrepancies. For example, this figure shows the cross-section   of a constant-speed propeller and the  mechanism for blade pitch control. This mechanism enables the airfoil to  be operated at optimal angles of attack. However, as we can see, the cost of improved  efficiency is greater mechanical complexity. For this reason, conventional solutions  are difficult to adopt for the Lilium Jet because of its large number of rotor blades. After all, the mechanical complexity  likely increases with the number of blades. Hence, Lilium has opted for a variable nozzle   to achieve high rotor efficiency  for both hover and cruise flights, and thus address one of the problems  a high power discrepancy brings. Another significant challenge created by  large power ratings is the battery design. Amongst many issues, the most notable one is the  battery’s specific power at a low state of charge. Towards the end of the flight, the  battery will be at a low state of charge, and due to the battery’s voltage drop with its  state of charge, specific power will be low. However, the Lilium Jet’s vertical  landing requires enormous power. As we learned from Lilium’s Technology Blog,   the hover power is ten times  greater than the cruise power. And Lilium’s technical paper,  which we will analyze in Chapter 2,   tells us that the Lilium  Jet’s hover power is 2.57MW. Therefore, combining ‘electric’ with ‘VTOL’  leads to the following design challenge, pertinent to all eVTOL concepts  but more so to the Lilium Jet: the eVTOL aircraft must conduct its vertical  landing before the specific power becomes too low. Otherwise, it cannot land as planned,  potentially compromising safety. This unique eVTOL problem  leads to the following dilemma: We can either choose to vertically land   the aircraft when its battery's  state of charge is still high, thus reducing our aircraft's  performance potential, like its range. Or, like Lilium, we can specify  an advanced battery that provides   both high energy density and high specific power. However, the battery design must  also be a balanced solution. For example, we also want the battery  to be safe, reliable, and long-lasting. In conclusion, the challenges created by high  power discrepancies are not easy to solve. A case in point is Lilium's concept.  Its ducted fans cause high disk loading, and paradoxically, they are somewhat of a  cure to the problems of high disk loading. On balance, data shows that  the cause outweighs the cure. This is because the ducted fans need  significantly more power for hover flight. However, one question remains: is  Lilium's concept a well-balanced solution? We have found some answers up to now. As we saw, Lilium decided to specify  advanced batteries instead of penalizing the aircraft's performance potential. This decision shifts the burden from  the eVTOL concept to the battery design. Thus, the power imbalance in Lilium's  concept potentially results in imbalanced battery design requirements. We spent a lot of time  reviewing Lilium’s ducted fans. So far, we have covered many theories. But there are problems with  theory versus practice. All of this is fine as long as we are  aware of the limitations of theory, and the uncertainty it brings. Uncertainty in predicting aircraft  performance can create problems, but we can deal with them in two ways. First, we ask many ‘what if’ questions, such as: What if the battery energy density  improves significantly in ten years? And what if the improvement is less than we hoped? This exercise of asking questions and using  a range of input values for our models   is known as a parametric study. It helps us understand the consequences  of uncertainty in a systematic way. Second, depending on how confident we feel about  our predictions, we can leave ourselves margins. These provide a buffer for critical interfaces. So conservative design approaches  use larger margins and vice versa. There have been several parametric studies  of the Lilium Jet over the past seven years, some private and others public. Aerokurier’s paper, arguably  the most famous study,   estimates a range of 18.4km for the Lilium Jet. In stark contrast, Lilium  estimates a range of 261km. Next, I will review each paper and  analyze the reasons for such differences. As mentioned previously, Aerokurier’s article from   January 2020 raised severe  doubts about Lilium's range. Based on the calculations of an aerospace  engineer, who chose to remain anonymous, the Lilium Jet was estimated to achieve less than  7% of its marketed 300 km range (only 18km!). Besides the contentious results, what  was striking about this technical paper was the sarcastic tone used  by the anonymous author. For example, the paper concludes that if the young  man from the YouTube channel "The Real Life Guys" makes some further improvements  to his flying bathtub,   with which he flew 1.5km to the bakery and back, then he would be more advanced than the Lilium  Jet in terms of range and energy efficiency. This is how Frank Thelen, one of Lilium’s  earliest investors, responded to the Aerokurier: “It seems you have already made your  decision to negatively bias this article.” “Among other things, you question our  competence and credibility as a tech VC,   since we stand behind the parameters  of the aircraft communicated by Lilium. We have been supporting the company for almost 4  years and have examined the technology in depth. We have critical information and parameters that  your professors should not have, which means that their assessment of the feasibility of  the jet cannot be sufficiently supported.” Frank Thelen concludes: “Instead of  supporting innovative, young companies and visionary founders, we nip their innovative  ideas in the bud with our skepticism.” “You rely on the opinions of some professors  who do not know the necessary parameters to evaluate the Lilium's technology   and consider it completely out of the question  that several top-class engineers and experts, who have worked on the Lilium Jet for years,   have achieved a breakthrough  in the history of aviation?” To which the anonymous aerospace  engineer responded: “It is astonishing   how Mr. Thelen thinks so little of the professors and how superhuman he seems  to think the Lilium team is.” It was not until more than a year after  Aerokurier's article, in April 2021, that Lilium published its own technical  paper to defend its eVTOL concept. This chapter examines the two  sides of the debate - arguments   that both question and defend  the Lilium Jet's capabilities - by reviewing the technical papers by  Aerokurier and Lilium, respectively. Aerokurier's paper uses a simplified version  of the methodology I described in chapter 1. When I reviewed the paper in detail, I found that  the author modeled the hover power incorrectly. This is due to two fundamental mistakes  related to the physics of propulsion,   not a typo or rounding error. The first mistake is related  to modeling hover efficiency. The second mistake is related  to modeling ducted fans. Consequently, this paper significantly  overestimates the Lilium Jet's hover power. This leads to higher energy  consumption in the hover phase and, as a result, a very low  aircraft range of only 18.4km. Let us now examine these  two mistakes in more detail.  The author estimated 110kWh of battery  energy for the 5-seater aircraft and assumed an efficiency of 0.2 for the  propulsion system during the hover phase. The first mistake lies in this  efficiency assumption of 0.2. However, before we delve deeper into this, let us  examine the plausibility of the author's results. With these values, the battery  power for hover is 5849kW. Other comparable eVTOL  aircraft require around 1000kW. A Tesla Model S has a vehicle  power of around 761kW. So the author's power estimate seems  far too high by a factor of 3 to 6. Overestimating the hover power has an  enormous impact on the aircraft range. For the author's assumed hover time of 60  seconds, the energy consumption is 97.5kWh, which is 89% of the 110kWh battery. This leaves only 11% of the  battery for the cruise phase, resulting in a range of only 18.4km. So, what went wrong with estimating  the propulsion system efficiency? The author correctly states that the  propulsive efficiency is zero during hover. Since the aircraft is not moving in any  direction, it is not doing any propulsive work. However, stating that momentum theory is   insufficient for determining the  hover efficiency is incorrect. In fact, as many standard textbooks  explain, momentum theory is used to find the hover efficiency,  also known as the figure of merit. In fact, we determine the figure of merit by  finding the ideal power with momentum theory and then dividing that ideal  power by the actual power. Since losses, such as the rotor  blades' viscous profile drag,   are not considered in momentum theory, the actual power is higher than the ideal power. Typical values for the figure  of merit are 0.4 to 0.7. And what did the anonymous author use? The author used a value of only 0.2, based  on a so-called 'time-averaged' method. However, the paper does not  describe this method in detail, and no justifications or references  were provided for the value of 0.2. I think that a hover efficiency  of 0.2 is far too low, a sentiment held by other experts, too. The Aerokurier reports that another professor's  estimation for the hover efficiency is 0.62, higher by a factor of more than three. This paper mistakenly used the  power equation for an open rotor. As explained in chapter 1 of this video,   the power requirement of a ducted  fan can be significantly lower. The absence of a vena contracta would make a  ducted fan more efficient than an open rotor. In conclusion, the author seems to have confused  propulsive efficiency for hover efficiency and has modeled an open rotor  instead of a ducted fan. The author used a value of  0.2 for the hover efficiency, which is far lower than the  technological capabilities of today. The justification of a value of 0.2  was not based on empirical evidence, but on an unexplained 'time-averaged' method; therefore, I find this value to be unjustified. I support my arguments with our  cross-checking of the results, which revealed an enormous power  value of 5849kW for the hover phase, which is highly implausible  for a vehicle of this size. The implications of these  mistakes are significant, especially considering how  the author used the results to criticize Lilium's concept in a sarcastic tone by comparing it to the  performance of a flying bathtub. Nonetheless, I consider this  paper’s qualitative analysis and subsequent discussions on the  Lilium Jet logical and reasonable. As rightly raised by Aerokurier to Lilium: “How do you come about  [Lilium’s] aircraft performance, as postulated for years,  of transporting five people at 300 km/h over a distance of 300 kilometers?”. Does that not seem too good to be true? What has been Lilium’s immediate response? Oliver Walker-Jones, the Head of  Communication at Lilium at that time, said: “The performance data we publish is based on  the synthesis of ground-test performance data of the aircraft’s individual components  as well as data obtained from seven months of flight testing and  state-of-the-art simulation software.” Furthermore: “The data we are  getting from our flight test   campaign continues to confirm our predictions.” The long-term response came more  than a year later in the form of   a technical paper by one of Lilium’s co-founders. So let us now examine Lilium’s paper. Let us begin this review by examining the paper’s  context and understanding its significance. This paper by Lilium is a primary reference  for supporting the performance claims. For example, this paper was  cited no less than six times   in Lilium’s investor presentation from March 2021. We can also find it on the front page  of Lilium’s investor relations website. And it is cited multiple times in the Tech FAQ  section on Lilium's investor relations website. Let us stay on this FAQ section for a little  longer to understand the paper's context. Near the top, it states: "Compared to other eVTOL  concepts, the Lilium Jet has a novel architecture, characterized by its ducted fans, or  Ducted Electric Vectored Thrust (DEVT)." It continues: "Because of its  novelty, some commentators have   struggled with properly assessing this  innovative aircraft architecture (...). These misconceptions often lead to errors when   third parties estimate and comment on  our aircraft's performance potential." Further down, Lilium furthers: “Essential physics about ducted fan aerodynamics  are occasionally misunderstood and lead to underestimating efficiencies in the hover phase of  flight for an architecture like the Lilium Jet’s. These misunderstandings can lead to overestimation  of the power requirements and further miscalculations of physical range capabilities.” These comments by Lilium are valid and  fair if they are connected exclusively to the mistakes we found in  Aerokurier's technical paper. When we continue, the arguably  most important FAQ reads: “Lilium appears to be making slow progress with   its demonstrator aircraft towards  the range target. Why is that?” Lilium answers: “The purpose of a technology demonstrator is  not to achieve the final range target but,   instead, to validate the flight  physics and control laws.” This statement means that Lilium is  focused on getting the steering right - using a car as an analogy, ensuring the  car turns smoothly, safely, and as desired when the steering wheel is used. The answer continues: “Since the architecture of our  demonstrator is very similar to   that of our anticipated certification aircraft, overall performance metrics can be  transferred with high accuracy (...). By tracking the power demand per flight  phase of the demonstrator, we can make predictions of the certification  aircraft’s performance due to similarity of the architecture and the proven  knowledge we have about the final battery system.” What can we deduce from this statement? Lilium states that their demonstrator  and certification aircraft are very similar in architecture  so they can estimate the certification aircraft’s range  with the demonstrator’s power. However, Chapter 1 showed that range depends  on the propulsion system’s efficiency, battery energy, and aircraft drag. So from the three parameters, what is causing the significant performance  gap despite the architectural similarity? Is it the batteries? Or are  there other reasons, too? This question, amongst many,  leads us to Lilium’s paper. Its purpose, according to  Lilium’s website is to provide “a more thorough and transparent  technical explanation of the Lilium Jet”. Considering the content we have just reviewed,  the significance of this paper is enormous. So let us now talk about this paper.  I have reviewed it in great detail, and I will go over my main findings first,  following which, I will support my statements with a detailed analysis of  the paper, and a conclusion. My main finding is that the paper’s  aim is not supported by its content. As the introduction states: “The aim of this paper is to demonstrate the  mission capability of [Lilium’s concept].” Similarly, the conclusion says: “The aim of this paper is to demonstrate the  technical feasibility of [Lilium’s concept].” However, Lilium’s paper uses theory instead  of in-house test data to achieve these aims. Consequently, this paper is more  like an initial concept study because it has more theory and  uncertainty than evidence and confidence. If Lilium's paper is viewed as an initial concept  study, then its methodology is reasonable. A concept study aims to compare many  different concepts quickly and narrow down the number of candidates with 'quick and  dirty' calculations for further investigation. However, Lilium's paper aimed to demonstrate  the technical feasibility of the Lilium Jet and its marketed aircraft performance. In that  case, these ‘quick and dirty’ calculations have too much uncertainty and  representative test data is needed instead. The paper does not use the  demonstrators' test data, despite Lilium mentioning flight test  campaigns and component ground tests to defend their performance targets  against Aerokurier's questioning. Considering Lilium published its paper more  than a year after Aerokurier's criticism, questions arise about Lilium's decision for  theoretical 'quick and dirty' calculations. Why does Lilium’s paper not use the  power test data from its demonstrators? As this clip shows, some flight  test data was already available a year and a half before  Lilium’s paper was published. Is it because the test  results are not good enough? Is the aircraft’s power higher than expected?  Is it performing worse than required? Not using the concept demonstrator’s  test data to demonstrate the technical feasibility of the concept  cannot be for intellectual property reasons. Sharing performance data is not  sharing intellectual property because performance data  does not contain ‘know-how’.   While it can be justifiably argued that geometric data is intellectual property,   basic performance data (e.g.,  aircraft power and energy efficiency) are not because it is impossible  to replicate the Lilium Jet by   knowing how efficiently it performs in tests. It is impossible to replicate a design by  knowing the tested value of the design’s benefit. Furthermore, as of the time of making this video,  Lilium’s aircraft demonstrators do not have a core technology:   the variable nozzle, which ensures high  efficiency for both hover and cruise flight. Lilium’s most recent demonstrator, the Phoenix 2,   has chevrons, which are  features for reducing noise. These two figures show the design differences  between chevrons and variable nozzles. Here is a close-up view of the differences. Dr. Nathen, the author of Lilium’s  paper, was asked on LinkedIn whether the variable nozzle was used for the  prototypes. He answered that the variable nozzle “would not have benefitted the intention  of [Lilium's] technology demonstrator.” His statement contradicts with what is said  on Lilium's investor relations website, which states: "overall performance  metrics can be transferred with high  accuracy, from the demonstrator to certification aircraft”. The absence of variable nozzles  in Lilium’s prototype means that its power characteristics differ  significantly from the certification aircraft’s. Lastly, Lilium makes several assumptions  for the aircraft's empty weight, drag, and system efficiencies that bias  towards a higher aircraft range but reduce the feasibility of the design. These assumptions were based on other studies,  which, as the detailed analysis shows, provide weak or no support for Lilium's case. Therefore, Lilium's assumptions in their paper   were found to be overly optimistic,  improbable, or unfounded. Since these assumed values are  multiplied for the range estimation, small individual overestimations can compound  to create a much larger range overestimation. This effect is similar to  how interest rates compound:   a 10% increase in a monthly interest rate leads to a significantly  larger annual interest rate. In conclusion, it can be said that Lilium’s paper  is a concept study that explores with low-fidelity the Lilium Jet's potential range under  different but mostly favorable circumstances. The results show that  Lilium's concept could achieve a range of over 250km if certain assumptions  are made, though it was found that the references for these assumptions provided  weak or no support for Lilium’s case. With this paper, Lilium aimed to demonstrate  the technical feasibility of its concept but did not use the test data  of its concept demonstrator.   This observation raises significant questions about the viability of the  numbers used in this paper. I support this statement with my seven  findings from reviewing this paper. Finding one: Take a look at the mass  fractions of 0.22 and 0.48, which Lilium supports with  reference to Kim et al. (2018). However, if we read that reference, we find  that it does not mention mass fractions nor uses values adjacent to  0.22 and 0.48 as Lilium does. This reference, therefore, does  not support Lilium's assumptions. Finding two: Lower down, Lilium supports the  battery mass fraction of 0.3 with a reference from Bacchini and Cestino  (2019). So, let us check this reference. The reference studied three aircraft: a multicopter concept by Ehang, a lift and cruise concept by Kitty Hawk, and a Ducted Vectored Thrust  Concept by Lilium itself. It is inappropriate to base the  Lilium Jet’s battery mass fraction on significantly different aircraft concepts   because of the differences  in aircraft architecture and required performance. The Kitty Hawk, for example, was  stated to achieve only 100km. Therefore, of the three concepts,  only Lilium’s own 2-seater version can be used for Lilium's paper. A problem thus arises when  we see the reference use a battery mass fraction of  0.49 for the Lilium Jet, whereas Lilium’s paper states 0.30. This reference, therefore, provides little support to Lilium's battery mass fraction assumption. Finding 3: As we continue reading, we find Lilium assuming a minimum state of charge of 10%. Besides the question of whether  delivering such high specific power is possible at such a low state of charge, discharging batteries to such an extent  is detrimental to their cycle life. Replacing batteries more frequently, of course, costs more money. For these reasons, a more reasonable  minimum state of charge is 20%. If a maximum SOC of 90% is also  considered because of charging time, the available battery  energy reduces even further. Lilium also assumes a battery  energy density of 320 Wh/kg. Consequently, we find that Lilium’s  aircraft range is increased by at least 13% to 29% through, in my,  overly optimistic battery assumptions. Finding 4: Next, we come to section four. It is the longest section,  spanning from pages 12 to 24, almost a third of the entire paper. The second paragraph reads: “The  analysis starts by calculating (...) the power use in each phase of flight.” Calculating the power? Considering the statements  on Lilium’s investor relations website, the skepticism raised by Aerokurier, and the replies by Lilium that mention test data, why is “calculating power” necessary? Let us examine this section  with this question in our minds. Finding 5: Section four leads with an  estimation of the Lilium Jet’s drag. Like we saw in chapter 1, this is  one of the three important parameters for maximizing the aircraft range. It proceeds to state that “The  total drag of the aircraft in cruise is the sum of the drags of  each part of the aircraft”. A drag estimation method like this  is fine for an initial concept study but insufficient for an aircraft concept that  has been under development for several years. Where are the wind tunnel test results? Furthermore, when we examine  equation 15 and the subsequent text, we find that the landing  gear drag was not considered. The canard’s drag contribution  was omitted entirely or partially. Lilium’s drag estimation only  mentioned the canard in section 4.1.1., as shown in this screenshot. The highlighted sentence “wing interference  drag increases the cabin drag by 30%” does not account for other types of drag. Furthermore, no reference was  provided for the 30% value. For the lift-induced drag, Lilium  only accounted the wing area, despite their Technology Blog stating that  only 60% of the lift comes from the main wings. Hence, the lift-induced drag from  the canards and cabin were not added. Among other details we did not cover, these omissions likely lead to  an underestimation of the drag, as reflected by Lilium’s lift-to-drag ratio. Lilium’s value of 18.26 seems too high, though I cannot be certain of this  conclusion without test data as evidence. Still, if we compare Lilium’s  lift-to-drag ratio to other aircraft, we find that Lilium’s value of 18.26  is significantly above average. The result of a potential  drag underestimation is that the Lilium Jet’s range is increased. Let us now examine the  propulsion system efficiency by starting with the ducted fan. Lilium’s paper reads: “The design  flow coefficient of the fan [as] 1.09 is higher than those typically  reported in the literature where the optimal flow  coefficient is approximately 0.6. The choice of a high design  flow coefficient is deliberate. It allows both the fan diameter to be kept low, and the blade tip Mach number to be below 0.45.” In other words, Lilium’s high disk  loading and low blade tip speed push the fan design into  unconventional design spaces. The paper continues: “High fidelity  analysis (...) give Lilium confidence that ducted fans with high  flow coefficients [like] 1.31 can be designed during cruise  flight at efficiencies of 0.83, while achieving peak performance in hover  flight with a fan efficiency of 0.88.” Note that Lilium’s flow coefficient for cruise is more than double the typical  optimal flow coefficient of 0.6. Uncertainty is shown from the words ‘analysis,’  ‘gives confidence,’ and ‘can be designed’; instead of ‘tested,’  ‘demonstrated,’ and ‘was designed.’ A reference to Howell (1945) is provided, who purportedly measured isentropic  efficiencies of 88 and 85%. Let us examine this reference. Figure 82 of Howell's paper  shows a tested stage efficiency of around 88% at a flow coefficient of 1.0. If Lilium's isentropic efficiency and Howell's  stage efficiency are considered equivalent, then the 88% reported in Howell's paper  corresponds to the value quoted in Lilium's paper. However, for a flow coefficient of 1.3,  Howell reports a stage efficiency of only 72%, which does not correspond to Lilium's 85%. The difference between Howell's  72% and the Lilium's 85%, which was made in reference to Howell's  work, was not explained in Lilium’s paper. Finding 7: Next, let us talk about motor efficiency. Lilium's paper assumes motor  efficiencies of 92% to 95%. Typical motor efficiencies range from 70 to 85%, so Lilium's values are at  the high end of the spectrum. Achieving such a high efficiency  would require very low friction between the shaft and its bearings,  minimal ohmic resistance in the motor, and a high torque motor. As discussed in chapter 1, a high torque  motor enables a higher motor efficiency but adds more weight. Since the Lilium Jet's total  shaft power is more than double that of other eVTOL aircraft due  to its fans' high disk loading, its motors must be sized for both  high power and high efficiency and are thus likely heavy. Though some motor mass is reduced by  increasing the number of propulsors, as explained in my previous  video, I remain skeptical about how Lilium plans to achieve  such high motor efficiencies, shaft power, and low weight simultaneously. Lilium states that “the viability  of these numbers is demonstrated and discussed in more detail in Section 5.2”. If we examine section 5.2., we see Lilium mentioning aerospace suppliers but ultimately referencing a  paper by Seitz et al. from 2012. So let us examine this reference. The introduction states: “the International  Air Transport Association (IATA) and the European Commission announced ambitious  emission reduction targets for the year 2050”. It continues, “To this end, the authors  have recognized that a forward-looking pre-concept review is warranted  in establishing, firstly, what potentially can be realised with   electrically-powered aircraft  propulsion in the future”. Does this reference’s forward-looking  pre-concept review demonstrate the viability of Lilium’s  motor efficiency numbers? To further answer this question, let us look at the technologies studied in this reference. This screenshot shows that  this reference is a study of high-temperature superconducting technologies, which feature motors with cryogenic rotors equipped with superconducting coils. It is highly unlikely that Lilium is using such advanced technologies. The efficiency values used by this reference are not representative of  near-term technology capabilities and do not support Lilium’s case. So, again, we come back to the question: why does Lilium reference such  futuristic technology capabilities instead of using its own test data? Let us now conclude our  analysis of Lilium’s paper. As we saw in chapter 1, component  efficiencies are multiplied to give the total efficiency. We multiply the total  efficiency by the battery energy and divide it by drag to determine the range. Therefore, overestimations  for individual parameters compound like interest rates do. For example, if every parameter  is inflated by 10 percent, then these inflations compound to give  a much larger range overestimation. What we have seen in chapter 2 is that Lilium has likely overestimated  numerous parameters, such as the battery energy  density, L/D ratio, fan efficiency, and motor efficiency, to achieve a range of 261km. Furthermore, the paper does  not consider the degradation of battery performance with age. Therefore, the Lilium Jet’s range  is likely significantly below the 261km estimated by Lilium’s technical paper. The methods presented in this  paper are simple and low-fidelity. Though the paper claims its methodology  “allows a range to be calculated accurately”, no model validation or  error analysis was presented to substantiate this claim. Furthermore, the references used by  Lilium to support its assumptions were found to be irrelevant, taken  out of context, or weak support. Above all, Lilium does not reference its test data despite mentioning this intent on its website. In conclusion, this paper fails to  demonstrate the technical feasibility of the Lilium Jet because it is not based  on the concept demonstrator’s test data. What do other experts say about Lilium’s paper? Volker Gollnick, Professor of aeronautics at the  Institut für Lufttransportsysteme of Hamburg, concluded the following (paraphrased): The reviewers of Lilium’s  paper have been contacted and interviewed about their  reviews and assessments. However, they state that they  do not confirm the assumptions made by Lilium concerning, for example,  efficiencies or energy densities. Therefore, questions have to  be raised about the estimations and assumptions of various figures, especially the efficiencies and discharge level, which are considered as too optimistic. So, as we can see, Professor  Gollnick’s statements agree with our findings and support our conclusions. The paper we just analyzed was  published in early April 2021, a few days after Lilium’s announcement  of the development of the 7-seater jet. Lilium’s announcement to develop a  7-seater caught many analysts by surprise due to the expectation for a 5-seater  to be the certification aircraft. It is one of the numerous key events  in Lilium’s evolution as a start-up. To better understand where Lilium is today and where it could possibly be in the future, we must understand Lilium’s history. Lilium was founded in 2015 by four students  from the Technical University Munich. These four students were on a mission  to change the world with Lilium. As Sebastian Born, one of  the co-founders, explained: “We are doing something here that has  the potential to change the world. In terms of the difference it could make, we think it’s similar to the  transition from horses to cars. While commercial flying has come down in  price over the last couple of decades, nobody has ever succeeded in  truly democratizing flight – making it available to anyone, anywhere, anytime.” In the founding year, the company’s mission  was to give people personalized air transport. The marketed aircraft performance  was even higher than they are today. Watch this clip of one of the co-founders  of Lilium speaking at ECO15 in London. So it combines the flexibility of a helicopter  with the speed and comfort of a business jet. It gives you everything  that traveling lacks today:   a speed of up to 450 kilometers per hour  and all electric range of 500 kilometers. No traffic jams, no delays  and probably the best view. Besides the higher aircraft  performance, the timeline was shorter. in May next year, we will do the first manned  hover flight with a full-scale prototype, in 2018 the aircraft will  be ready for certification and Market entry is planned for 2019. The manned hover flight, where a test pilot sits  inside the vehicle, never occurred in May 2016. Skepticism from the establishment came soon  after, a year after Lilium’s founding year. In 2016, the Wired magazine asked: “Can  those narrow wings generate the lift needed to carry two adults 300 miles at an average  speed of 180 mph? If the max takeoff weight is just 1,322 pounds, how much  battery could possibly be in there?”.   It further reports: “Based on the  renderings, the ‘jet’ will fly, but not far, says Charles Eastlake, an aerodynamicist  at Embry-Riddle Aeronautical University. Concepts like this often soar on hopes  and dreams before slamming into hard  realities, he says. “‘In general, the public has a  hopelessly optimistic view about   how straightforward and wonderful  electric vehicles are,’ he says – a view that often doesn't consider the  challenges of safety, weight, and cost.” There was strong pushback  from the people at Lilium. Watch this clip of one of  Lilium’s co-founders speaking at the Hello Tomorrow Summit in  2017, pushing back against criticism. And when you reach this top of this mountain,  there will be this wall of disruption. And this is the wall - this  consists of non-believers, naysayers, investors who don't like you, and there will be tons of, I promise you, and the only thing that you have to  do now is to smash this wall down, and the only way how to do this: When you disrupt the world, it  would always run against it. So take your hat, smash against it, smash  against it, and it will hurt, literally. At that event, Lilium explained that both their  business model and aircraft concept had changed. Last year, I was actually  presenting you our answer   to the individual on-demand transport problem - to get a person from point A to point B, and I was showing this bad boy here: a two-seated  all-electric vertical takeoff and landing jet. Last year, I demonstrated or said in some years  that you will fly from Charles de Gaulle to here and this will be will be not changed,   but the whole business case behind it  and what we want to do change completely. My name is Patrick and I'm one of  the co-founders of Lilium Aviation and this is our answer to  the urban mobility problem: a fully electric vertical takeoff and landing jet   with five seats - an air taxi  and the first of its kind. In five years from now, I want all of  you, when you arrive at Charles de Gaulle, that you take out your mobile phone, order a jet  and fly within five minutes here, to Le CENTQUATRE The new vision was an air taxi service. Perhaps as a consequence of the changes, the  market entry was then moved from 2019 to 2022. The aircraft performance was additionally  lowered to 5 occupants, a 300 km range, and speed of 300km/h. From 2017 onwards, Lilium’s new mission was  “to make air taxis available to everyone and as affordable as riding a car.” The team envisioned that the Lilium  Jet could “takeoff and land anywhere” and customers can “take off with the  push of a button” on their mobile phones. You can book the jet on your smartphone,   it's going to come, pick you  up, fly to your destination, land and drop you off again. And the great thing   about this is it's going to be  the same price like an Uber, but it's 300 kilometers per hour fast. Those selling points remained  consistent for many years afterward. What did not remain consistent,  however, was Lilium’s timeline. The first manned flight was  pushed from 2016 to 2019 and the market entry was pushed from 2022 to 2025. In 2019, the company claimed that fares  would be low and affordable to everyone as part of their vision to democratize flight. Watch the following clip of one of Lilium’s  co-founders speaking at the Web Summit in 2019. So what we have here is a comparison of the energy  cost of one person for a 300-kilometer trip, assuming the load factors of these  different transportation systems. So for a helicopter, we are at 108 Euros. This is not really great for one of us. Then for the electric car, we have  9 Euros in a ground-based service and for the Lilium jet, we  sit somewhere around 6 euros. But how will they do this? Lilium plans to achieve low  fares by using high load factors, which means the aircraft operates at  nearly full capacity for every trip. Furthermore, Lilium assumes that their  aircraft will be more productive than a car and achieve more passenger kilometers per hour. In 2019, the Lilium Jet was  awarded the RedDot design award. Watch this clip titled ‘the  inspiration behind the Lilium Jet’. Super happy to introduce you to Mathis, one of our top designers who designed the  Lilium Jet that you can't see right now. And it won recently the Red Dot award, one of the most prestigious Design Awards  in the category Best of the Best [Music] So the goal was really to  inspire myself from the nature and really catch this elegant  line that manta ray has when she glides in the water and to integrate that into the prototype. That's what happened in the beginning with me. They really give me all the responsibility  for the the design of the aircraft. [Music] So I came to Lilium to recreate  something completely new to really improve the world,  to design something. [Music] 2020, Aerokurier published scathing  articles on the Lilium Jet, overall questioning the viability of the concept. These articles were based on the  technical paper we analyzed in chapter 2. Though critical calculations were incorrect,   the central question drawn from  the results still holds true today: where is the test data from  Lilium’s concept demonstrators? However, this doesn’t seem to  have affected Lilium’s rise. In October 2020, Lilium was named the  number one start-up in Germany by LinkedIn. By now, this was one of the dozens of  accolades Lilium has received over the years. Lilium’s stable of world-class  talents continued to grow. At the end of 2020, Lilium appointed veteran  Rolls-Royce engineer Alastair McIntosh as their Chief Technology Officer. A few months later, in January  2021, Tom Enders, former Airbus CEO, joined the Lilium board. ‘How would they  lead Lilium?’ many commentators wondered. In February 2021, Forbes reported that Lilium had  been secretly working on a seven-seater concept which was unveiled officially  by Lilium in March 2021. That Forbes article also revealed the following: First, Lilium aims to begin flight  testing of the new aircraft in 2022   and to win safety certification  by the end of 2023. Second: “two former Lilium employees told Forbes  that development of the five-seater prototype was dogged by problems and that the  flight test campaign made minimal progress before it was incinerated in a  battery fire in February 2020.” And third: “an internal investigation  concluded the fire was likely caused by a bent pin in a battery  connector that caused a short, two of the former employees say – to save  weight, connectors weren’t fully covered.” Evidently, Lilium struggled with achieving the   incredibly low weight target  of its five-seater prototype. “‘You don’t account for all the  screws you need ... or you need   more carbon fiber than in  your original CAD model,’ the former employee says. ‘The  grams, they just add up and up.’” Lilium has made changes to fix this problem. Here is what two former Lilium employees  have said about the new 7-seater concept. “the design, for the most part, is a blown  up version of the five-seater that would have a maximum take-off weight of  3,175 kilograms (7,000 pounds), 2.4 times that of the five-seater’s  1,300 kg (2,866 pounds).” As the data shows, the 7-seater concept has  a significantly higher empty aircraft mass. In other words, the mass budget was increased  for the aircraft structure and systems, improving the technical feasibility and accounting  for the mass of the screws, carbon fiber, and connector coverage mentioned earlier. Later that year, in April 2021,   Lilium published a paper to "demonstrate"  the viability of their 7-seater concept. Though more plausible than the 5-seater concept,  the skepticism continued as not a single test data from the concept demonstrators  was actually referenced. Still, Lilium has begun to  move in the right direction,   and these sensible changes to Lilium’s concept were presumably made under the new leadership,  such as their new CTO, Alastair McIntosh. However, besides the aircraft concept, Forbes  unveiled problems in Lilium’s management. “Three former Lilium employees  (...) said that they and a   number of their colleagues left over the past year out of frustration over the  management of the company by Wiegand, who they say is overcontrolling and has held to an   unrealistic timeline in the  face of development delays, pushback from his engineers and a planning  process that was throwing up red flags.” “management repeatedly said that  first flight would be in two weeks, which became a running joke among workers.” “The unrealistic assertions were at the  insistence of Wiegand, the ex-employees say, who, in the manner of many startup founders, they describe as unyieldingly  optimistic and deeply involved   in every aspect of the aircraft’s design.” “With a keen memory for physics formulas,  Wiegand would overwhelm his engineers with rapid-fire calculations on his iPhone  calculator to sketch out design specifications that they say others might take  hours or days to arrive at.” There seemed to be friction  inside the Lilium management, generated by presumably novices and  veterans rubbing against each other. “Airbus veteran Yves Yemsi, who joined the  company as chief program officer in 2019, brought in a team of project managers and  attempted to implement more rigorous work and planning processes,   but three of the ex-employees say  Wiegand resisted the conclusions.” Despite this backdrop of problems, the year 2021  concluded with Lilium winning the IF Gold Award in the discipline of Professional Concept. A  manta ray was depicted in their aircraft sketches. The old management would not last, as  the Financial Times reports in 2022: “Wiegand was replaced as chief  executive in August by Klaus Roewe, who worked at Airbus for 30 years. “There was a bit of ego involved, but my ego  is happy with where we are,” says Wiegand” From 2021 to 2022, notable changes  were made to the Lilium Jet. Amongst them are increasing the  number of seats from 5 to 7, increasing the aircraft’s empty mass fraction, adding the variable nozzle technology, changing from a retractable  to a fixed landing gear, and introducing a conventional landing capability. These are mostly sensible  changes that make progress   toward the performance and certification targets. In the Development Update video from June 2022, a  stated maximum range of just 175 km is mentioned. Compared to Lilium’s paper from 2021,  the range was lowered by a third, from 261 to 175 km. Finally, a presentation from December 2022 shows  that Lilium’s goal of a 300km range aircraft has been pushed to be beyond 2030. These are all sensible changes  that increase the feasibility. However, while the 7-seater concept  was making much-needed progress,   presumably under the new leadership, in catching up to the reality,   the marketing department was seemingly one  step ahead but in the opposite direction. In a presentation for investors in March  2021, Lilium marketed a 16-seater concept. The pictures from Lilium’s presentation suggest   that its concept does not scale  with wingspan or rotor disc area. Is that 16-seater design feasible and sensible? We know by now that the thrust requirement for  vertical takeoff must be significantly higher than the aircraft’s weight.   And the propulsion power scales with  thrust raised to the power of 1.5, even for a ducted fan design. Hence, the power requirement for the 16-seater  concept is likely significantly more than double the 7-seater’s 2.5MW. How the 16-seater is projected to be  launched by 2028 was not explained and its technical feasibility is a mystery to me. Going forward, Lilium must  finish developing and testing   a full-scale prototype of its 7-seater concept, which is the anticipated certification aircraft. The 7-seater prototype significantly differs  from their current one, the Phoenix 2, notably in weight and the variable nozzle. If the variable nozzle technology  is not matured in time, the aircraft will experience drastically different   rotor inflows during hover  flight and cruise flight, resulting in either poorer hover  efficiency, cruise efficiency, or both. The consequences of low  efficiencies are significant: a lower hover efficiency results in a higher shaft   power requirement and thus  larger propulsion systems; a lower cruise efficiency  penalizes the aircraft range,   reducing the current target of 175km even further. Other tilt-eVTOL aircraft achieve high efficiency  for hover and cruise with variable pitch rotors, a technology Lilium stated they  would avoid for simplicity. However, variable pitch rotors are much   more common and arguably better  understood than variable nozzles. Hence, Lilium’s quest for an efficient  design does not seem so simple after all. Currently, the success of the Lilium Jet  depends heavily on advanced battery technology and favorable regulatory requirements,  such as a low hover time. Lilium also faces challenges in  developing the variable nozzle   and certifying the aircraft before 2025. Considering these significant  uncertainties and difficulties,   here is what I would do going forward. First, going for such a high range  and high payload concept generates unreasonable requirements for the battery and the aircraft’s empty weight. From my visual impression, Lilium’s  concept could have lower drag and a higher minimum drag  speed than its competitors. These characteristics would mean that the concept  is better suited for high speed and range, advantages that must be maintained at  all costs – even at the cost of payload. The 7-seater’s target range  of 175km is unimpressive. Considering that time savings (i.e., service  value to the customer) increase with range, the simplest change would be to convert the  7-seater to a 5-seater or even a 3-seater by exchanging passengers for  additional battery packs. This change increases the battery mass by 200  to 400kg and the range disproportionally more by around 40-80% (assuming the  energy expended for everything   other than cruise flight remains constant). From a mission perspective, the Lilium Jet   design is essentially trading away its  hover capabilities for aircraft range. However, I believe hover capabilities are  more important for certification than range, due to the need to maximize hover  endurance for safe landings. Therefore, the potentially better but more  difficult solution to Lilium’s problems would be to lower the disc loading by  reducing the overall aircraft size and mass while keeping the rotor disc area the same. For example, I would consider  reducing the MTOM to around 2000 kg by lowering the payload,  battery mass, and disc loading. Though these changes reduce the  revenue per trip, they compensate by increasing the technical feasibility,   reducing the operating cost for trips  with low passenger load factors, and increasing the aircraft range  for important market differentiation. Lastly, Lilium plans to use a variable nozzle  but has yet to demonstrate this technology. I would consider shelving the  variable nozzle for the future and going for a simple rigid duct  for the certification aircraft. The fan blade design would consequently need to be   a compromise of the demands  of hover and cruise phases. Though this would lower aircraft performance, this  change would increase the technical feasibility and maximize the company’s chances of meeting its  ambitious market entry target of the year 2025. As we have seen in chapter 1, optimizing the range  of an electric aircraft involves design tradeoffs between aircraft drag, system  efficiency, and weight. Finding the optimum solution is  challenging, especially when other issues like safety and noise are also considered. The challenge is further increased by the  absence of an eVTOL development history, detailed regulations, and  market-validated requirements. The vast number of unique  eVTOL concepts in research and   development reflects the wide range of approaches taken in interpreting, optimizing, and  implementing these design tradeoffs. Lilium’s approach is a ducted vectored  thrust concept characterized by a canard wing configuration, electric  ducted fans with a high disc loading, and thrust vectoring for aircraft control. Their concept is unique due to the implementation  of numerous unconventional technologies. Uniqueness, however, can be a double-edged sword. On the one hand, it achieves important product  differentiation in a crowded market space. And original, innovative,   and bold designs are often lauded during  early research and development phases. On the other hand, bringing innovation to an  industry where safety is paramount can be risky. Therefore, the benefits of such innovations   must be weighed carefully and  candidly against their risks. In chapter 2, we reviewed  Aerokurier’s and Lilium’s papers. We found the Aerokurier  significantly underestimated   Lilium’s range because of two fundamental mistakes in the power model for hover flight. Nonetheless, the subsequent questions  raised by Aerokurier were still reasonable. In particular, how Lilium planned to achieve  its highly ambitious performance targets. Lilium’s response was their own technical paper. According to Lilium, its paper 'aims  to demonstrate the mission capability' and 'the technical feasibility' of its concept. However, when we analyzed Lilium's paper,   we discovered that it does not include  test data from Lilium's demonstrators. Not using the demonstrator's test data  cannot be for intellectual property reasons because performance data  does not contain 'know-how'. It is impossible to replicate the Lilium Jet by  knowing how efficiently it performs in tests. And, according to Lilium's investor relations  website, 'overall performance metrics can be transferred with high accuracy, from  the demonstrator to certification aircraft'. In that case, why is Lilium not using test  data as evidence to support its technical paper and market claims and answer  reasonable skepticism? Instead of test data, Lilium’s paper used theory   and assumed values for aircraft  drag, mass, and system efficiency to estimate an aircraft range of 261 km. Lilium based many critical  assumptions on other references. We examined these references  and found that almost all of   them provide weak or no support for Lilium’s case. Some references used by Lilium  were taken out of context. Others do not have the value  quoted by Lilium’s paper. Overall, certain values are theoretically  possible when viewed in isolation but unfeasible when other design  tradeoffs are also considered. The nature of physics and engineering is that you can be exceedingly good in one aspect  or another but not all at once. Therefore, Lilium’s assumed  values were found to be overly optimistic, improbable, or unfounded. And since these values are  multiplied for the range calculation, small individual overestimations compound to  create a much larger range overestimation. In chapter 3, we viewed this paper by Lilium  in the context of the company's history. We saw that Lilium has consistently reduced the  aircraft range and flight speed over the years. What once was 500 was reduced  to 300 and is now just 175 km. Meanwhile, the number of passengers  increased from 2 to 5 to 7. This trend is counter-intuitive to me because   the advantage of a high-speed  aircraft (i.e., time savings) is expressed through higher aircraft ranges  and, as a consequence of more batteries to increase the range, lower passenger numbers. It's been two years since Lilium announced the  7-seater as its certification aircraft in 2021. And Lilium is two years away from  its market entry target in 2025. Going forward, Lilium must still  build its 7-seater prototype and demonstrate the feasibility  of the aircraft's 3-tonne weight and the variable nozzle, which is  crucial for achieving the 175 km range. So, will the Lilium Jet work? In my view, it is extremely  unlikely that the Lilium Jet   can be made to work within the  timeline given by the company. This view is shared by many other  industry experts around the world. For example, Mark Moore, a prominent figure widely  credited for kick-starting the eVTOL industry, has severely questioned  Lilium’s certification timeline. He said this about recent  statements made by Lilium’s CEO: “I find it amazing that the CEO of a public  company would say that he’s (...) 100% certain that they will have entry into service by 2025.” He continues: “The fact [is] that they haven’t  even built or flown a full-scale demonstrator as a basis to know when they will be certified.” He furthers, “I would think shareholders will have   litigation opportunities with these  statements [by Lilium’s CEO].” Some analysts go even further. Iceberg Research, a firm notorious for  finding accounting irregularities and   fraud in public companies,  concluded the following: “Lilium should have remained a school project   and never come to market to  raise hundreds of millions. There is no redemption for companies when  their technology is structurally flawed.” “We believe it’s high time to end this charade  before more investor money is incinerated.”

Channel: John Lou

Views: 103,323

Keywords: lilium, eVTOL, UAM, AAM, electric aircraft, VTOL, aircraft

Id: Hk6kF2eOxOQ

Length: 109min 16sec (6556 seconds)

Published: Thu Apr 27 2023