Skip to main content

News & Events


SENS4ICE newsletter No 4: Feedback from the SENS4ICE Icing Wind Tunnel campaign

(Left) Collins Aerospace - Applied Research and Technology team in Cork Ireland: Rohan Chabukswar, Giancarlo Gelao and El Hassan Ridouane. (Middle) Collins Aerospace – Ice Protection team in Ohio USA: Matthew Hamman, Andrew Taylor, Galdemir Botura and Michael Beyman. (Right) Bora Venkatesh (TUBS) Elias Paul-Quentin (ONERA) and Rafael Martins (ONERA) at work at the TUBS IWT.
(Left) Collins Aerospace - Applied Research and Technology team in Cork Ireland: Rohan Chabukswar, Giancarlo Gelao and El Hassan Ridouane. (Middle) Collins Aerospace – Ice Protection team in Ohio USA: Matthew Hamman, Andrew Taylor, Galdemir Botura and Michael Beyman. (Right) Bora Venkatesh (TUBS), Elias Paul-Quentin (ONERA) and Rafael Martins (ONERA) at work at the TUBS IWT.

One of the main objectives of SENS4ICE is to develop and mature different ice detection technologies. Ten different ice detectors have been developed by the consortium partners. Most of these ice detectors target detection of and discrimination between the Appendix O and Appendix C conditions.

Extensive icing wind tunnel tests have been performed in the project to evaluate the detector’s performance in relevant environment at icing wind tunnel. This icing testing reproduces natural icing conditions in compliance with aeroplane certification specifications (CS-25/14 CFR Part 25, formerly known as FAR 25)  App. C and the extension to Supercooled Large Droplets (SLD) cloud conditions (App. O). 28 weeks of testing time have been distributed over four facilities participating in SENS4ICE. The results of these icing wind tunnel tests provided data for technology selection and development towards natural icing flight tests.

The SENS4ICE project works with five icing wind tunnels (IWT) worldwide: the Collins Aerospace Icing Wind Tunnel in the US, the Technische Universität Braunschweig Icing Wind Tunnel in Germany, two TsAGI IWT in Russia i.e., the Climatic-type Icing Wind Tunnel AHT SD as well as the Year-round Icing Wind Tunnel, and the Altitude Icing Wind Tunnel research facility of the National Research Council Canada.

In this article, we offer complementary points of view of the IWT operators (Collins Aerospace and Technische Universität Braunschweig) as well as that of the sensor developers (Honeywell and DLR) on the SENS4ICE IWT testing campaign. Indeed, the IWT operators focus on the steps they had to take to prepare and perform a successful icing wind tunnel test campaign, whereas the sensor developers highlight the actions necessary to prepare their ice detectors for the IWT test campaign and deliver the results.


Interview with the Icing Wind Tunnel operators: Collins Aerospace and the Technische Universität Braunschweig

Question 1 (Q1): The icing wind tunnels of Collins Aerospace and the Technische Universität Braunschweig are state-of-the art facilities. However, they needed to enhance some technological features to fit the testing requirements set in SENS4ICE. What were the challenges of upgrading the icing wind tunnels to the needs of the project?

Answer 1 (A1) / Collins Aerospace: Collins Aerospace IWT was inaugurated in 1988.  Since that time the tunnel has received several upgrades - from the structure to the cooling system used to maintain the performance required for the airworthiness authorities. The IWT has been used for a number of OEM, FAA, NASA, university, small businesses and internal Collins ice protection developments. One of the key challenges was in achieving the smallest Liquid Water Content (LWC) value possible for the tunnel’s App. O conditions. Throughout the process an additional priority for Collins Aerospace was establishing conditions with high Median Volume Diameter (MVD) levels (i.e. MVD > 100µm) in order to capture the most critical aspects of Supercooled Large Droplets (SLD). Achieving these two objectives simultaneously - that is, generating conditions with high MVD but low LWC levels - is a difficult task. Because of this, the tunnel went through a series of studies, analyses and characterization efforts to achieve large MVDs and reduce the LWC to a more desirable level.  Studies are currently underway to further improvement and to find a solution to increase the tunnel icing condition inside the App. O icing envelope.

A1 / Technische Universität Braunschweig (TU Braunschweig): TU Braunschweig , established in 1745, is the oldest technical university in Germany with a core research focus in mobility, metrology, infection and therapeutics and future cities. The Institute of Fluid Mechanics at TU Braunschweig, renowned for its expertise in wind tunnel testing, designed and commissioned the Braunschweig Icing Wind Tunnel (BIWT), a state-of-the-art icing tunnel with proven App. C capability conforming with SAE ARP-5905. The active engagement of the BIWT in several international icing research projects in collaboration with research organizations like DLR, CIRA and ONERA has contributed significantly to understanding the icing physics and the development of numerical icing codes. The perseverance of the aviation community in making aviation safer has resulted in the new icing certification requirements listed in App. O, which are more hazardous and differ substantially from App. C conditions. To simulate App. O conditions, BIWT is upgraded with a new spraying system with an independent control system. The key challenge in augmentation to App. O is to obtain the characteristic wider spectrum of droplets (1µm to 2000 µm) while satisfying the conflicting lower LWC limits. The calibration of the upgraded tunnel is a further challenge due to the lack of accepted instrumentation and the large uncertainties produced by large droplets with a very low probability. Extensive in-house calibration and collaboration with DLR enabled the realisation of Freezing Drizzle conditions in BIWT.

Q2: Which ice detectors have been tested by Collins Aerospace and the TU  Braunschweig? 

A2 / Collins Aerospace: The Collins Aerospace IWT was extensively used by the Collins Ice Differentiator Detector (Collins-IDS). Collins Aerospace also allowed the sensor developers in the consortium to test at the company’s IWT. Unfortunately, some ice detector developers had to change plans not to use the Collins Aerospace IWT due to travel restrictions caused by the Covid-19 pandemic. With the objective to improve icing cloud characterization, the Cloud Combination Probe (CCP) and Nevzorov probe were also tested in the Collins Aerospace IWT as part of the reference measurement activities led by DLR.

A2 / TU Braunschweig: TU Braunschweig was pleased to welcome several SENS4ICE sensor developers, including five development sensors and two reference sensors. The testing of CM2D sensor developed by DLR was carried out along with cloud characterisation with the CCP and Nevzorov probes. The novel conceptual sensors tested in the SENS4ICE project are the Local Ice Layer Detector (LILD) sensor from DLR and the Atmospheric Hydrometeor Detector based on ELectrostatics (AHDEL) from ONERA. IWT tests of the Primary in-Flight Ice Detection Sensor (PFIDS) and the Appendix O Discriminator (AOD) sensors from Safran Aerotechnics were postponed to October 2021 due to the Covid-19 situation.

Q3: A wide testing envelope represents a key asset in the experimental effort. What were the main testing conditions?

A3 / Collins Aerospace: The test conditions were down-selected by the SENS4ICE consortium and tunnel operators. This task was led by Collins Aerospace with the objective of maintaining a high degree of confidence with commonality among other tunnel participants. A comprehensive matrix was used with priorities for different conditions to cover the icing envelopes.

A3 / TU Braunschweig: The test points were derived to enable the testing of sensors in multiple aspects like their sensitivity to the LWC and their ability to discriminate between App. C and App. O type of clouds. The response of the sensor to the changes in the operating temperature, particle size, LWC and velocity are studied. The App. C distributions in most wind tunnels are identical, however, the droplet distributions and LWC of App. O clouds are found to be different across the tunnels. In order to produce a common test ground, the expected response times are scaled to produce 0.3 mm thick ice on a cylinder with 25 mm diameter, which is an accepted indicator in the aerospace industry and defined within ED-103. With the consensus of the SENS4ICE partners, a robust test process is employed that essentially tests the reliability of the sensor in terms of positivity and false negative rates, annunciation ability, reproducibility and endurance when subject to harsh conditions for longer durations. In addition to the agreed test matrix, several tests were carried out rigorously with the individual sensor developers to enhance the understanding of the ice accretion and the specific sensor’s behaviour.

Q4: The sanitary measures taken in the framework of the Covid-19 pandemic put a stop to international business travel which means that not all sensor developers in the project could witness the icing wind tunnel tests on-site. In this context, how did you manage the test campaign? 

A4 / Collins Aerospace: Unfortunately, the unprecedented Covid-19 pandemic caused delays and cancellations in the test schedule. However, Collins Aerospace successfully implemented all the safety procedures and protective equipment for continuation of the tests and tunnel calibration to satisfy the needs of the SENS4ICE project. Collins Aerospace completed the calibration remotely, making use of collaboration tools such as web cameras and video conferencing technology to connect with our international consortium peers. Organizationally, Collins Aerospace had a clear mission to exceed SENS4ICE expectations and support the project as well as we could. Truly, a job well done by all involved.

A4 / TU Braunschweig: Certainly, Covid-19 hindered the planning and levied additional costs in time and efforts. As most sensors are in a maturation stage, remote testing was not a viable option. To fully harness the benefits of the proposed IWT tests, the presence of sensor developers was deemed necessary. TU Braunschweig has given utmost priority to the safety of employees and guests and ensured it by practicing a detailed hygiene plan drafted in consultations with the university health department and the city health office.

Q5: What are the preliminary observations/results/conclusions of the tests from the point of view of the icing wind tunnel operator?

A5 / Collins Aerospace: The tunnel delivered what was expected. The results from the Collins-IDS were exceptional, with clear differentiation between dry, App. C and App. O with higher LWC. The tests were successful, and most importantly Collins Aerospace was able to safely meet the consortium’s needs and expectations.

A5 / TU Braunschweig: Unique innovative icing environment detection technologies have been developed in SENS4ICE. The successful IWT tests have demonstrated the soundness of the concepts. The sensors tested have shown exemplary performance in detecting the icing conditions, the response in most cases is as instantaneous as the cloud. Evaluation of these diverse technologies in controlled environment should have enhanced the understanding of the detection principles and shortcomings. The tests will enable the developers to improve the architecture to be accepted as primary ice detection systems. Besides, the concepts have enormous potential to serve applications beyond icing.

Q6: What is the way forward for Collins Aerospace and the TU Braunschweig after the SENS4ICE icing wind tunnel test campaign?

A6 / Collins Aerospace: Continuous improvement, testing and investment into ice detection technology is the mission of the Collins Aerospace Ice Protection program. The Collins IWT will pick back up with our internal programs, technology advancement and research that were temporarily put on hold while supporting the SENS4ICE initiative. We expect to have the Collins-IDS back in the tunnel some time before the natural icing flight tests in the first quarter of 2023 to refine the logic for the specific application. It will be another busy year in the tunnel.

A6 / TU Braunschweig: The Multiphase Flows and Icing Group of the TU Braunschweig has experience with several measurement technologies, and participation in the SENS4ICE testing has further strengthened the group’s interest in novel measurement principles. SENS4ICE and other EU projects enabled the upgrade of BIWT to SLD conditions on par with leading world class facilities. The calibration of the SLD conditions is challenging, the TU Braunschweig will continue its collaboration with DLR in this regard to ascertain the uncertainties and contribute to establishing acceptance criteria. Further, the upgraded facility will be utilized in studying the critical aspects of SLD icing like droplet splashing and its influence on the accretion rate, extended icing limits due to large droplets and heat transfer investigations. All experiments will help to get a deeper understanding of the physics of ice accretion and will lead to improved numerical models.

Q7: Do you have other comments that you would like to share with the members of the SENS4ICE community?

A7 / Collins Aerospace: Collins Aerospace would like to thank SENS4ICE for the opportunity to participate in such an important consortium for improving safety in aerospace.  The importance of collaborative projects emphasizes that 1+1 is more than 2 when everyone works together with the same purpose and objectives. Collins Aerospace expects to continue these collaborations and hopes to support others in the future.

A7 / TU Braunschweig: TU Braunschweig is delighted to be a part of the SENS4ICE project and thanks the European Union for supporting the activity. The consortium brought together niche expertise across the world; the seamless collaboration of the partners has enabled a  definition of new standards for future testing. We look forward to future collaboration.


Interview with the sensor developers: Honeywell and DLR

Question 1 (Q1): In the SENS4ICE project, Honeywell has been developing the Short Range Particulate (SRP) sensor, whereas DLR has been developing the Cloud Multi Detection Device (CM2D). The first two years of the project were spent to further increase the technology readiness level of the sensors and prepare them for the icing wind tunnel tests. What were the challenges in developing and enhancing the ice detectors?

Answer 1 (A1) / Honeywell: The SRP developed under SENS4ICE was a variation of an existing technology under development at Honeywell, which is focused on smaller droplets (< 50 µm). Both sensors utilize backscattered light to detect and size individual droplets. There are many challenges associated with such an optical technology to detect the larger droplets, such as balancing the need for a larger sample area (due to their rarity in the atmosphere) and the increase in background light contamination. Due to the low light levels associated with backscatter, the sensor is always pushing the limits in terms of signal to noise ratio for accurate droplet detection and sizing.

A1 / DLR: The CM2D consists of two flight proven sensors, the Nevzorov probe for measurement of liquid water content and the Backscatter Cloud probe with Polarization Detection (BCPD) for the measurement of droplet sizes from 2 - 42 µm. However, no information about the presence of SLD is directly available from the output of either of the two sensors. During the first two years of the SENS4ICE project, we investigated how the outputs of the two sensors could be combined to obtain information on the presence of SLD. We performed first icing wind tunnel testing at TU Braunschweig in App. C conditions as well as some further tests to improve our understanding of the performance of the Nevzorov probe in App. O conditions. From the results of the testing, we developed the algorithms for the App. O detection.

Q2: The sanitary measures taken in the framework of the Covid19 pandemic put a stop to international business travel which means that not all sensor developers in the project could witness the icing wind tunnel tests on-site. In this context, how did you manage the test campaign?

A2 / Honeywell: Planning for two IWT test campaigns was a big part of our strategy in the event that one of the campaigns could not be held. The SRP was scheduled to be at both the Collins Aerospace tunnel (in person) and the NRC tunnel (as a remote test). This combination of remote and in person test mitigated several risks: (1) remote testing mitigated the risk that the test would need to be cancelled due to travel restrictions and (2) personal test mitigated the risk that equipment was not a fully autonomous system and might require some on-the-fly adjustments. In the end, we utilized the Collins Aerospace test as our primary test under SENS4ICE.

A2 / DLR: Fortunately, the CM2D wind tunnel testing was scheduled in summer 2020, when only few Covid-19 restrictions were in place in Germany. With a coordinated hygiene concept, we were able to travel to Braunschweig and the campaign was carried out with only reduced personnel but achieving all scientific goals.

Q3: As several icing wind tunnel facilities and several ice detectors are involved in SENS4ICE, the consortium defined standardised icing wind tunnel conditions and standardised reporting requirements based on the icing wind tunnel tests. This is to ensure a comparability of results. What are the preliminary observations/results/conclusions of the icing wind tunnel tests of your sensors?

A3 / Honeywell: Overall, the results from the Collins Aerospace IWT test were quite good with respect to detecting all App. C and App. O test conditions within the required response time calculated with ED-103B. There were some systematic errors in the droplet sizing, but, as the errors are systematic, this is promising in terms of finding a root cause which is correctable. Investigations are currently underway to address this issue.

A3 / DLR: Our sensor combination employs scientific instruments. Results from the IWT testing show that we are able to provide highly accurate information on atmospheric parameters, like the liquid water content and the droplet size distribution. Data from the BCPD sensor, which is a fairly new sensor and had not been extensively tested before, showed that its measurements are comparable to those of other atmospheric instruments like the Cloud Droplet Probe (CDP). Furthermore, it is able to distinguish between water and ice. The detection of icing conditions is almost instantaneous and App. O can be discriminated from App. C in most cases.

Q4: What are the current limits of your sensors and possible improvements in the framework of SENS4ICE and beyond?

A4 / Honeywell: The main developments that are planned for SRP in the coming years are items which will lead to a product down the road – enhanced ruggedization, full qualification against DO-160, software and analysis improvements.

A4/ DLR: The CM2D sensor combination is designed for scientific measurement campaigns. For commercial aviation, the Nevzorov probe is too heavy and produces too much drag. Nonetheless, the combination of a hotwire sensor and an optical sensor is able to provide a lot of information on the atmospheric conditions and may give pathbreaking incentives for the use on commercial aircraft.

Q5: What is the way forward for Honeywell­-SRP and DLR-CM2D in the framework of SENS4ICE?

A5 / Honeywell: As SRP passed the SENS4ICE sensor selection process, which is the final selection of direct sensors towards flight testing, Honeywell will focus on addressing the calibration issue observed at the Collins Aerospace IWT and implementing a few small hardware alterations to support flight test. A more automated and robust software system will be developed which will allow for autonomous operation and improved live data processing. Flight test is a key step towards the maturation of the SRP sensor and the opportunity to flight test under SENS4ICE will be a valuable addition to the SRP development.

A5 / DLR: The CM2D will go to flight testing on the ATR-42 operated by SAFIRE. We are currently discussing the installation with SAFIRE and ATR. We also aim to publish the findings that we obtained from the icing wind tunnel testing in the near future.


DISCLAIMER: The information, statements and opinions in the above interview are personal views of the individuals involved in the SENS4ICE project and do not necessarily reflect the views of the SENS4ICE consortium as a whole, nor of the European Commission. None of them shall be liable for any use that may be made of the information contained herein.

Johannes Lucke (DLR) and Raphael Märkl (DLR) in the TUBS IWT.