## [Moisture ingress protection for hydrogen refuelling nozzles](/spotlight/student-project/moisture-ingress-protection-for-hydrogen-refuelling-nozzles) /sites/default/files/spotlight-images/2026-05/mtrl.jpg Erez Kassirer, Harvard Tang, Rehaan Raother and Rushil Bhatt Department of Materials Engineering - Materials Engineering - Bachelor of Applied Science Vancouver ## Our project Our client, [IRDI System](https://irdisystem.com/about/), observed that certain hydrogen refuelling nozzles experienced reduced reliability after extended service. We were asked to redesign the moisture protection system inside the nozzle so that the electronics would be better able to withstand exposure to humidity and the significant temperature swings (from −40°C to ambient conditions) experienced during the hydrogen fuelling process. [IRDI System](https://irdisystem.com/about/) ## Our design solution We approached the problem by evaluating the three components that make up the protection system: the potting compound, the adhesive that bonds the infrared lens to the housing, and the conformal coating on the PCB. > Our goal was to recommend materials that would reduce moisture ingress, minimize mechanical stresses during thermal cycling and maintain compliance with industry safety standards. We started with a literature review to decide which candidate materials had the required structural specifications and thermal properties to improve the protection system. What we found is that some of the materials have a glass transition temperature inside the nozzle’s normal operating range. When the material crosses that temperature, its coefficient of thermal expansion jumps dramatically, and because the various surrounding materials expand and contract at different rates, the resulting tensile stress leads to delamination, cracking and ultimately moisture pathways. > This information guided our material selection. We focused on materials that have glass transition temperatures well outside the operating range so as to minimize the coefficient of thermal expansion difference. The compound we recommended for the potting compound has a transition temperature of 82°C (which is far outside the operating range) and absorbs 80% less moisture. For the adhesive, we recommended a cyanoacrylate that has a fast cure time, low moisture absorption, high tensile strength, transition temperature of 80°C and a coefficient of thermal expansion similar to the materials it would be in contact with. We also recommended a new conformal coating that has improved electrical reliability in humid environments. [![Moisture Ingress Protection Project Capstone Poster.](/sites/default/files/inline-images/image_16.png)](https://engineering.ubc.ca/sites/default/files/2026-05/moisture_ingress_protection_capstone_poster.pdf) ## The technical challenges we faced One of the biggest challenges was validating our solution. To do this, we designed and 3D-printed a custom test jig that fits onto our client’s robot arm. This rig is capable of testing 10 enclosures simultaneously between a freezer at -30°C and a humidity chamber at 25°C, ultimately replicating a year of real-world use of 12,500 cycles. Image ![Client's robot arm with a 3D-printed custom test jig.](/sites/default/files/styles/original_image/public/2026-05/screenshot_2026-05-20_at_11.49.26_am.png.webp?itok=Y7uQeEMJ) Microscopy analysis was another challenge. Our sponsor provided us with some non-conforming units so that we could characterize the failure modes. We started out with optical microscopy, but this was unsuccessful as the curved and recessed epoxy made it difficult to get any usable images. We then switched to [scanning electron microscopy (SEM)](https://ncstate.pressbooks.pub/advancesinpolymerscience/chapter/sem-scanning-electron-microscopy-characterization-of-polymers/), but the samples were too large to sputter coat, which is normally required to prevent charging. > After conducting additional research, we decided to use low vacuum SEM with variable pressure. By capturing micrographs between 20x and around 1000x magnification, we could see two main failure modes: delamination at the interface between the potting compound and the housing and localized brittle failure. ![Lab workstation showing electron microscope analysis with micrographs displayed on dual monitors.](/sites/default/files/inline-images/image_1.jpeg)## How we validated our solution Our experimental validation included moisture absorption testing, which showed that our proposed potting compound absorbs 80% less moisture. We also assembled 10 prototype nozzle receiver assemblies using different combinations of compound and coating. The assembly process spanned several days to accommodate the curing schedules of each material, progressing through conformal coating of the PCBs, infrared lens bonding, and finally potting compound injection. > Through an iterative process, we refined our potting procedure. An initial attempt using top-injection proved incompatible with the enclosure geometry, and switching to bottom-injection through the cable hole produced consistent, well-filled assemblies across the remaining units. Nine units will undergo thermal cycling testing at IRDI. ![Lab setup with wires fed into a test enclosure for potting process trials.](/sites/default/files/inline-images/image_2.jpeg) ## What’s next for our project The sponsor will complete the temperature cycling tests to quantify the performance of our candidate materials and prototypes. If the results align with our early findings, they may implement our proposed redesign into their units to enhance robustness, extend service life and improve the operational reliability of hydrogen fuelling infrastructure. ## What we’re most proud of This was a very hands-on project and we spent hours in the lab troubleshooting, iterating and refining our material choices and procedures. We gained experience using SEM to analyze polymers and adapted our technique to produce meaningful images. > We also gained experience designing test plans, procuring materials and developing procedures from scratch. Finally, it was very rewarding to see that our recommended potting compound performs five times better in moisture absorption testing. ![Lab setup with samples sealed in a box during moisture absorption testing.](/sites/default/files/inline-images/image_3.jpeg) IRDI System Inc. - [Engineering](/topics/engineering "Engineering") - [Student Experience](/topics/student-experience "Student Experience") Community Partner Link