ANL-ART-318 ANL-METL-67 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Nuclear Science and Engineering Division About Argonne National Laboratory Argonne is a U.S. Department of Energy laboratory managed by UChicago Argonne, LLC under contract DE-AC02-06CH11357. The Laboratory’s main facility is outside Chicago, at 9700 South Cass Avenue, Lemont, Illinois 60439. For information about Argonne and its pioneering science and technology programs, see www.anl.gov. DOCUMENT AVAILABILITY Online Access: U.S. Department of Energy (DOE) reports produced after 1991 and a growing number of pre-1991 documents are available free at OSTI.GOV (http://www.osti.gov/), a service of the US Dept. of Energy’s Office of Scientific and Technical Information. Reports not in digital format may be purchased by the public from the National Technical Information Service (NTIS): U.S. Department of Commerce National Technical Information Service 5301 Shawnee Road Alexandria, VA 22312 www.ntis.gov Phone: 800-553-NTIS (6847) or 703-605-6000 Fax: 703-605-6900 Email: orders@ntis.gov Reports not in digital format are available to DOE and DOE contractors from the Office of Scientific and Technical Information (OSTI): U.S. Department of Energy Office of Scientific and Technical Information P.O. Box 62 Oak Ridge, TN 37831-0062 www.osti.gov Phone: 865-576-8401 Fax: 865-576-5728 Email: reports@osti.gov ANL-ART-318 ANL-METL-67 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 by A. Grannan, E. Kent, H. Belch, S. Bakhtiari, T. Elmer, H. T. Chien, and C. Grandy JUNE 2026 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 ABSTRACT The Gear Test Assembly (GTA) is an experimental test apparatus built and installed in the Mechanisms and Engineering Test Loop (METL) at Argonne National Laboratory (ANL). While designed to accommodate a variety of components intended for use in advanced compact in-vessel transfer machine, the focus of GTA been the testing of large radial spur gears. The performance of the components used in GTA motivates the choices made in the design, components, and material choices for the forthcoming METL test articles; the Gripper Test Assembly (GrTA), and the Bearing Test Assembly (BTA). To date, GTA has completed nine experimental testing campaigns and achieved over 21 million revolutions under prototypic sodium fast reactor loads and operating conditions. This report provides an overview and results for the two most recent experiments Campaign #8 and Campaign #9. `Campaign #8 was the first campaign to make use of a set of nickel alloy radial spur gears, made from Hastelloy C-22HS, and combined with heat-treated tapered roller bearings. The bearings and torque profile used allowed Campaign #8 to be directly compared with Campaign #1 where Inconel 718 gears were used instead. Campaign #8 attained 1,058,880 shaft revolutions within 23% of the 1,314,855 revolutions achieved in Campaign #1. These results demonstrate that the heat-treated tapered roller bearings show consistent increased lifetimes, approximately three times longer than for non-heat-treated taper rollers, and that the particular material of the gear does not play a substantial role in performance. Similar to Campaign #7 where bearing failure resulted in fragmentation and transport of bearing material into the gear teeth led to tooth damage in the Inconel gears, a similar process led to the damage of the Hastelloy gears. Subsequent NDE analysis showed that the damage was only limited to the tooth surface and does not extend into the gear interiors. Based on the NDE reports, the Hastelloy gears can be reconditioned and put back into service for future campaigns as was done for the Inconel gears. Reconditioned Inconel 718 gears damaged in Campaign #7 were returned to service in Campaign #9 and paired with non-heat-treated tapered roller bearings. In Campaign #9 a novel set of operating conditions were implemented whereby the gears and bearings were cycled between hot, submerged refueling conditions where fuel handling operations occurred and cold, dry, and inert conditions where the sodium was drained and the vessel was allowed to cool simulating long, ex-vessel storage of the fuel handling machine. A total of eight of these operating cycles were achieved before a bearing failure occurred after the accumulation of 475,692 shaft revolutions were achieved. During Campaign #9, one of the drive motors was damaged and over the course of a 240 day standby period, the entire GTA drive system was upgraded from a 480VAC system to a 240VAC system that provided the opportunity to consolidate wall-space in the METL facility and provide more space for additional test articles being installed in the coming year. The over 475,000 shaft revolutions achieved in Campaign #9 eclipses the previous record for lifetime of non-heat-treated tapered roller bearings held by Campaign #3 of 392,000 shaft revolutions. Further analysis and future campaigns hope to shed light on the cause of this approximately 17% increase in lifetime. ANL-ART-318 ii Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 TABLE OF CONTENTS ABSTRACT............................................................................................................................................... ii Table of Contents ...................................................................................................................................... iii Table of Figures ........................................................................................................................................ iv Table of TABLES ...................................................................................................................................... v 1 Introduction ........................................................................................................................................... 1 1.1 Motivation for Gear Test Assembly .............................................................................................. 1 1.2 Gear Test Assembly Overview ...................................................................................................... 3 2 Summary of Previous Campaigns ......................................................................................................... 5 3 Campaign #8.......................................................................................................................................... 9 3.1 Campaign #8 Results ..................................................................................................................... 9 3.2 Campaign #8 Failure .................................................................................................................... 10 3.3 Over-Pin Measurements of Gears ................................................................................................ 12 3.4 NDE Method: Eddy Current Sensing .......................................................................................... 12 3.5 NDE Method: Ultrasonic Analysis .............................................................................................. 17 4 Campaign #9........................................................................................................................................ 22 4.1 Campaign #9 Results and Failure ................................................................................................ 27 5 Summary and Path Forward ................................................................................................................ 31 Acknowledgement ................................................................................................................................... 32 References ................................................................................................................................................ 33 ANL-ART-318 iii Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 TABLE OF FIGURES Figure 1.1. Compact In-vessel Transfer Machine – Based on Pantograph Design .................................... 2 Figure 1.2. Perspective View of Gear Test Assembly ................................................................................ 4 Figure 1.3. Layout of GTA gears and bearings submerged in the sodium ................................................. 4 Figure 3.1. Torque Profiles used in GTA Campaigns. ............................................................................... 9 Figure 3.2. Location of damage to the bearings in Campaign #8. ............................................................ 11 Figure 3.3. Evidence of damaged gears and bearings from Campaign #8. .............................................. 11 Figure 3.4. Damaged gear teeth and example of the Over-Pin Measurement Method ............................. 12 Figure 3.5. Representative ECT inspection data collected on tooth #8 Hastelloy gear L1AT. ................ 13 Figure 3.6. ECT Inspection of T1BT: Tooth #14. .................................................................................... 15 Figure 3.7. ECT Inspection of T1IT: Tooth #17....................................................................................... 16 Figure 3.8. ECT Inspection of T1IT: Tooth #5......................................................................................... 17 Figure 3.9. Ultrasonic imaging system and data acquisition setup. .......................................................... 19 Figure 3.10. 3D Volumetric images of Gear T1IT after testing. .............................................................. 20 Figure 3.11. 3D Volumetric images of Gear T1BT after testing. ............................................................. 21 Figure 4.1. Vessel Temperature in Campaign #9. .................................................................................... 24 Figure 4.2. Cumulative Shaft Revolutions in Campaign #9. .................................................................... 25 Figure 4.3. Consolidation of GTA Electrical Components....................................................................... 26 Figure 4.4. Comparison of torque profiles before and after the Standby period of Campaign #9. .......... 26 Figure 4.5. Location of damage to the bearings in Campaign #9. ............................................................ 28 Figure 4.6. Extensive passivated sodium found in Campaign #9 disassembly. ....................................... 29 Figure 4.7. Damage to Cylindrical Pin Thrust Bearing in Campaign #9.................................................. 30 Figure 4.8. Damage to tapered roller bearing T1IT-Top in Campaign #9. ............................................... 30 ANL-ART-318 iv Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 TABLE OF TABLES Table 2.1. Summary of the bearings and material treatment used in each campaign ................................. 5 Table 2.2. Timelines for each experimental campaign ............................................................................... 7 Table 2.3. Duration breakdown of activities during the experimental phase of each campaign ................ 7 Table 2.4. Duration and performance of gears and bearings across all GTA Campaigns. ......................... 8 Table 4.1. Summary of detailed testing dates for Campaign #9 ............................................................... 23 ANL-ART-318 v Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 1 Introduction The Gear Test Assembly (GTA) is an experimental test article built for use in the Mechanisms Engineering Test Loop (METL) at Argonne National Laboratory (ANL). GTA was the first test article to operate in METL, with the initial experimental campaign beginning in early 2019. A total of nine experimental campaigns have been completed to-date with the seven previous campaigns discussed in the referenced reports [1], [2], [3], [4], [5], and [6]. There are two sections of new results provided in this report. The first part provides a summary of experimental operations and the results of Campaign #8 along with bearing and gear analysis in the form of detailed photographs and non-destructive evaluation (NDE) analysis using eddy current sensing and ultrasonic testing. The second part consists of a summary of the experimental operations and the results from Campaign #9. 1.1 Motivation for Gear Test Assembly The purpose of the GTA is to evaluate the performance of mechanical components intended for use in an ANL designed advanced compact in-vessel transfer machine (IVTM). The ANL IVTM, Figure 1.1, will be used in pool-type sodium fast reactors (SFR) and is designed such that all fuel assemblies can be reached using a single rotatable plug. This is compared with the more traditional use of multiple eccentric rotatable plugs and a straight pull IVTM that allow for a single point on a single plug to cover the whole of the reactor core. The ANL IVTM design allows for a reduction in the diameter of the reactor vessel which translates into a smaller containment facility overall, reducing the overall capital costs of the reactor plant. To accomplish this, the design for the IVTM requires that the mechanically driven components to be fully submerged in the primary coolant. These components include radial spur gears, mechanical bearings, ball screws, bushings, and universal joints. At the start of this project, there was a lack of sufficient literature regarding the design and performance of these mechanical components when submerged in liquid sodium and operating under the anticipated loads of the IVTM. Therefore, it is crucial to fill the gap in our knowledge that informs the development of the full-scale GrTA and before proceeding with a complete fabrication and testing of the full IVTM. ANL-ART-318 1 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Figure 1.1. Compact In-vessel Transfer Machine – Based on Pantograph Design The Gear Test Assembly (GTA) represents the first device designed and fabricated for testing the major components needed in order to realize the complete IVTM. GTA is currently equipped to test radial spur gears up to six inches in diameter and mechanical roller bearings with an internal bore diameter of 2.5”. GTA was robustly designed so that it could be modified to accommodate a variety of gears, e.g., helical spur, worm, straight, and/or spiral bevel, as well as a variety of bearing designs including ball, tapered roller, needle, thrust, hydrostatic, and/or hydrodynamic variants. ANL-ART-318 2 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 The performance of the gears and bearings analyzed over the nine campaigns of GTA has informed on the design of the Gripper Test Assembly (GrTA) and the Bearing Test Article (BTA). The GrTA is an experimental assembly currently under construction that will evaluate the performance of a full-scale fuel assembly gripper head that also operates and is driven by mechanical components submerged in sodium [7], [8]. The BTA is designed, and is currently being fabricated, in order to provide a dedicated testing environment for the bearings that have been found to be the primary source of failure in GTA. 1.2 Gear Test Assembly Overview GTA, shown in Figure 1.2 and Figure 1.3 was designed to operate in the smaller, 18” nominal diameter, of the two vessels used in the METL facility. Thus, the foundation for the design of the GTA is a blank 18” Class 300 ANSI flange. Two drive shafts penetrate the flange to provide the prescribed torque to the tested gears and bearings. An internal support structure is used to position the test gearboxes in the liquid sodium pool, below the argon cover-gas and sodium vapor space. Two test gearboxes contain two sets of radial spur gears customized for this experiment. One large pair of gears has 15-teeth, and a second smaller pair has 22-teeth. To date, these gears have been supported by tapered roller and ball bearings while the shafts themselves are supported axially by cylindrical pin thrust bearings. Essentially all submerged components are manufactured from 316 stainless steel to ensure known material compatibility with the liquid sodium test environment. The 316 stainless-steel also provides sufficient strength for the load bearing components to operate in this environment at elevated temperatures. The only submerged components not made from 316 stainless steel are the Inconel 718 gears or Hastelloy C-22HS gears, steel bearings under investigation, and Inconel 718 shaft sleeves between the gears/bearings and drive Shaft A and Shaft B. ANL-ART-318 3 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Figure 1.2. Perspective View of Gear Test Assembly Figure 1.3. Layout of GTA gears and bearings submerged in the sodium ANL-ART-318 4 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 2 Summary of Previous Campaigns GTA has completed nine experimental testing campaigns with the two most recent campaigns being the subject of this report. Each campaign follows a similar set of operational phases. In Phase 1, the specific combination of gears and bearings is chosen and GTA is fully assembled on a custom-made assembly positioner that allows GTA to be rotated to make assembly easier. The four spur gears used in the previous campaigns are constructed from Inconel 718 (ASM 5664). The more varied test components have been the bearings which have been the primary source of failure in all previous campaigns. A summary of the bearings used is provided in Table 2.1. Two types of bearings have been used in testing. The first are commercially available tapered roller bearings made from 52100 bearing steel by Timken. The off-the-shelf versions of these bearings without additional heat-treatment have been used in Campaigns #2, #3, #5, and #7 and are the most economical (lowest cost) and readily available (shortest lead-time). A heat-treated version of these bearings that pushes its stability limits from 150°C to 350°C and has been used in Campaigns #1 and #4 but come at higher cost and a lead time of 14 weeks (3.5 months). The other type of bearing used was a custom set of ball bearings fabricated from heat- treated M50 tool steel. These bearings have been used in Campaign #6 and were the most expensive and had the longest lead time, 32-36 weeks (8-8.5 months) of all variants tested. Table 2.1. Summary of the bearings and material treatment used in each campaign Gear Bearings Lead Material Campaign Material Bearing Type Mfr. Material Time Treatment (wks) 1 Inconel 718 Tapered Roller Timken 52100 Bearing Steel Heat Treated 14 2 Inconel 718 Tapered Roller Timken 52100 Bearing Steel Standard 1 3 Inconel 718 Tapered Roller Timken 52100 Bearing Steel Standard 1 4 Inconel 718 Tapered Roller Timken 52100 Bearing Steel Heat Treated 14 5 Inconel 718 Tapered Roller Timken 52100 Bearing Steel Standard 1 6 Inconel 718 Angular Ball Custom M50 Tool Steel Heat Treated 32-36 7 Inconel 718 Tapered Roller Timken 52100 Bearing Steel Standard 1 8 Hastelloy C-22HS Tapered Roller Timken 52100 Bearing Steel Heat Treated 14 9 Inconel 718 Tapered Roller Timken 52100 Bearing Steel Standard 1 In the next phase, the fully assembled GTA is installed in Vessel #1, one of the 18” stainless steel test vessels that comprise the METL facility. The vessel and test article are heated to 250°C and the vessel is filled with nuclear grade sodium with an oxide content below 5-wppm. The METL facility is equipped with a cold trap that operates between 150°C – 170°C and is run periodically over the course of the experimental campaign to maintain the sodium’s purity. The dates during which the major phases of each campaign took place are given in Table 2.2. In the experimental testing phase, a cyclic loading is provided to the gears and bearings with one of the motors acting as the driver, while the other motor acts to provide a resistive loading force that mimics the maneuvers necessary for handling core assemblies over the lifetime of a reactor’s operation. The experimental phase can be further broken down into more granular detail as summarized in Table 2.3. The periods specified are: ANL-ART-318 5 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 • Normal testing where the experiment runs without issue. • Standby: the experiment is placed into a safe non-operating state for over 10 consecutive days • Maintenance: experimental testing is stopped due to sodium filling and draining or periodic purification of the sodium inventory where the test vessel is included in the purification piping circuit, or a short repair is performed to the METL facility. • Repair: experimental testing is stopped to perform a repair to the test article itself. • Idle: experimental testing is stopped primarily due to weekends or other periods of less than 10 days. Fortunately for most campaigns the primary period is that of Normal operation. The exceptions are Campaign #2 where GTA was placed in a 69-day Standby due to the Covid-19 pandemic and Campaign #4 where repairs to the METL facility and the Laboratory’s closure over winter holidays caused GTA to be placed in a 38-day Standby. For the longest-lived experiment, Campaign #6, GTA operations were ceased for METL maintenance only two times where no operations were performed in a 24-hour window. GTA has only ceased operations to make repairs to the test article for two days in Campaign #1 where an upper gear block bearing was replaced and for two days in Campaign #6 when a graphite shaft seal had to be repacked. The Idle phase accumulates over time for longer campaigns as the period of normal operations is ceased on weekends, short duration holidays like Thanksgiving, or other activities that pause GTA operations. Starting in Campaign #6, more advanced monitoring software allows for extended operations that promote up time where GTA operates and limits the amount of idle time. The experimental phase ends when either a failure of a component, as indicated by a change in noise during operation or variation in the signals from vibration sensors is detected or the campaign has been halted after a sufficient period of testing. A failure was detected for all previous campaigns except for Campaign #6 which was halted before failure occurred. All failures have been due to the fragmentation of the bearings due to lack of material stability for the bearing inner and outer races at the temperatures where testing occurred or, in the case of the heat-treated bearings, the cage used to maintain spacing between the roller elements has not received the same heat treatment resulting in a failure. A summary of the operational conditions, time at torque, and total revolutions for all campaigns is compiled in Table 2.4. the lifetimes of the campaigns is primarily dictated by the bearing type and bearing material choices and not by gear material. For the non-heat-treated tapered roller bearings, the lifetime in revolutions is ~300,000 shaft revolutions at torque. Implementing heat-treated tapered roller bearings with their increased stability at higher temperatures provides ~3x increase in lifetime to approximately one million shaft revolutions. Finally, the switch to angular(ball) bearing geometry provides an ~10x increase in lifetime compared to the heat-treated tapered rollers or approximately ~20 million shaft revolutions. Despite such failures the data collected helps inform the maintenance schedule to be used should such bearings be deployed in the in-vessel fuel handling machine. Our testing finds that while no bearing need be precluded from installation in an IVTM, however, the shorter-lived bearing variants would necessitate a more numerous replacement schedule during the lifetime of refueling operations for sodium fast reactors. ANL-ART-318 6 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Table 2.2. Timelines for each experimental campaign Campaign Campaign Campaign Campaign Campaign Campaign Campaign Campaign Campaign Activity 1 2 3 4 5 6 7 8 9 Sodium Fill 02/01/19 03/03/20 02/09/21 11/22/21 06/15/22 11/10/22 08/31/23 03/19/24 03/17/25 Testing Start 02/05/19 03/18/20 02/16/21 11/23/21 06/21/22 11/11/22 09/06/23 03/21/24 03/20/25 Standby X 03/20/20 X 12/06/21 X 12/22/22 X X 05/31/25 Restart X 05/28/20 X 01/12/22 X 01/03/23 X X 01/25/26 Fault/Stopped 03/07/19 06/15/20 03/04/21 01/27/22 06/27/22 05/10/23 09/08/23 04/19/24 03/12/26 Drain 08/01/19 07/28/20 03/05/21 02/03/22 06/28/22 05/12/23 09/18/23 04/25/24 03/23/26 Removal 08/06/19 08/07/20 03/18/21 02/21/22 07/07/22 05/25/23 10/04/23 04/30/24 04/16/26 Carbonation 08/06/19 08/10/20 03/19/21 02/22/22 07/08/22 05/26/23 10/05/23 05/01/24 04/20/26 Start Carbonation End 09/17/19 08/18/20 04/02/21 03/21/22 07/23/22 06/05/23 10/11/23 05/06/24 04/24/26 Cleaned and 11/07/19 09/02/20 05/05/21 04/29/22 08/22/22 07/07/23 10/25/23 05/31/24 05/31/26 Disassembled Table 2.3. Duration breakdown of activities during the experimental phase of each campaign Campaign Activity 1 2 3 4 5 6 7 8 9 Total Duration Submerged in 182 148 25 74 14 184 19 38 167 851 Sodium (Days) Duration of Experimental 31 90 17 66 7 181 3 30 358 783 Period (Days) Duration of Normal Testing 29 15 13 16 5 124 3 16 12 233 (Days) Duration of Standby Period 0 69 0 38 0 12 0 0 240 359 (Days) Duration of Maintenance 0 0 0 0 0 2 0 4 21 27 (Days) Duration of Repair (Days) 2 0 0 0 0 2 0 2 0 6 Duration of Idle (Days) 0 6 4 12 2 41 0 8 85 158 ANL-ART-318 7 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Table 2.4. Duration and performance of gears and bearings across all GTA Campaigns. Campaign 1 2 3 4 5 6 7 8 9 Total Inconel Inconel Inconel Inconel Inconel Inconel Inconel Hastelloy Inconel Gear Type 718 718 718 718 718 718 718 C-22HS 718 HT- No HT- No HT- HT- No HT- No HT- HT- No HT- Bearing Type HT-Ball Roller Roller Roller Roller Roller Roller Roller Roller High Torque [Nm] 400 450 450 450 450 300 Low Torque [Nm] 150 150 150 150 150 150 150 150 120/200 Time in Sodium [hr] 4344 3528 576 864 312 4392 432 888 4008 19344 Experimental 10343 2768 3136 2532 1150 41685 57 8824 99.85 Operating Cycles Set Experimental 120 240 240 240 240 240 240 240 240 RPM Revolutions per Experimental Phase 1 60 125 125 125 125 125 3600 120 3600 Experimental Cycle Total Phase 316,50 1,058,88 620,580 346,000 392,000 143,750 5,210,625 205,200 359,460 Revolutions 0 0 High Torque Period 3 10 10 10 10 5 [s] Low Torque Period [s] 30 28 28 28 28 40 910 35 902 Time at High Torque 8.62 7.69 8.71 7.03 3.19 12.26 [hr] Time at Low Torque 86.19 21.53 24.39 19.69 8.94 463.17 14.41 85.79 25.02 [hr] Total Time at Torque 94.81 29.22 33.1 26.73 12.14 463.17 14.41 98.04 25.02 [hr] Experimental 9,257.00 3,445.70 645.75 Operating Cycles Set Experimental 150 240 120 RPM Experimental Phase 2 Revolutions per 75 3600 180 Experimental Cycle Total Phase 12,404,52 694,275 116,235 Revolutions 0 High Torque Period 3 [s] Low Torque Period [s] 30 910 90 Time at High Torque 7.71 [hr] Time at Low Torque 77.14 871 16.14 [hr] 1,314,85 316,50 17,615,14 1,058,88 21,868,02 Revolutions 346,000 392,000 143,750 205,200 475,695 5 0 5 0 5 Time at High Torque 16.33 7.69 8.71 7.03 3.19 12.26 55.21 [hr] Totals Time at Low Torque 163.33 21.53 24.39 19.69 8.94 1334.16 14.41 85.79 41.16 1713.40 [hr] Total Time at Torque 179.67 29.22 33.1 26.73 12.14 1334.16 14.41 98.04 41.16 1768.63 [hr] Total Time at Idle [hr] 4164.33 3498.78 542.9 837.27 299.86 3057.84 417.59 789.96 3966.84 17575.37 ANL-ART-318 8 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 3 Campaign #8 For Campaign #8, GTA was reconstituted with Hastelloy C-22HS radial spur gears and Timken tapered roller bearings made from 52100 bearing steel and heat-treated for enhanced stability at higher temperatures. The goal of this campaign was to replicate the components used and torque profile applied in Campaign #1 with the only difference being the Hastelloy gear material replacing the Inconel 718 gear material used in all previous campaigns. GTA was installed in METL’s Vessel #1 and was filled 03/19/2024. Testing began on 03/21/2024 and was concluded after a failure was evidenced on 04/19/2024 that caused the gear and motor to train to lock in place, unable to move any further. During the operational phase of Campaign #8, the torque profile implemented had been used previously in Campaigns #1, #2, #3, and #5 and is shown in Figure 3.1.a. This profile consists of a short duration, 2-10 seconds, high torque ~400Nm phase, shown in green, followed by a longer duration, 30-60 seconds, lower torque ~200Nm phase, shown in red, concluding in a dwell or idle phase after which the process restarts by rotating in the opposite direction. During both torque phases, the rotation rate was fixed at 240 RPM. The temperature of the vessel and internal sodium temperature was fixed at 250°C. After failure the vessel was drained and cooled and GTA was removed on 04/30/2024. From 05/01/2024 to 05/06/2024 GTA was placed in the carbonation cleaning system and after this was completed, Campaign #8 was disassembled and the cause of failure and subsequent damage was identified. Figure 3.1. Torque Profiles used in GTA Campaigns. 3.1 Campaign #8 Results Campaign #8 was submerged in sodium for 37 days and experimental testing took place over 30 days as indicated in Table 2.3. Over the course of this testing period, GTA operated at low torque for 85.79 hours and 12.26 hours at high torque as shown in Table 2.4. Over this duration, GTA accumulated 1,058,880 shaft revolutions at torque. The dates for the operational activities are provided in Table 2.2 and Table 2.3 and a summary operating hours, cycles completed, and revolutions completed are provided in Table 2.4. ANL-ART-318 9 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Looking back at previous campaigns where heat-treated bearings were used, it is helpful to directly compare Campaigns #1 and #8, Although heat-treated tapered roller bearings were used in Campaign #4, cylindrical thrust bearings that provide axial load support at the ends of the shafts were not used and it is suspected that the additional loading taken on by the tapered roller bearings led to a significantly shortened lifetime. Thus, as was the intent of the planning for Campaign #8 the most glaring difference is the gear material choice. The approximately one million shaft revolutions achieved in Campaign #8 differs from the 1.3 million shaft revolutions achieved in Campaign #1 by about 23%. As shown in Table 2.4, Campaigns #1 and #8 operated at torque for approximately 180 hours and 98 hours respectively. However, owing to the fact that the rotation rate used in Campaign #1 was approximately half, 120-150 RPM, of the RPM used in Campaign #8 each campaign was still able to achieve 1.3 million and 1.0 million shaft revolutions respectively, Although we don’t have enough data to perform statistical analysis the general agreement shows that under nearly identical operating conditions the Inconel gears used and Campaign #1 and the Hastelloy gears used in Campaign #8 likely do not play a strong role in affecting the lifetimes of performance when compared with the bearing installed and the bearing material choices made. 3.2 Campaign #8 Failure Final disassembly of Campaign #8 revealed that the two tapered roller bearings above and below the test gear, T1BT, on Shaft B had fragmented with rolling elements being released from the bearings into the teeth of the test gears. In addition to the damaged tapered roller bearings, the cylindrical thrust bearing at the bottom of Shaft B was also found disintegrated. The locations and images of the damaged components are shown in Figure 3.2. The outer races for these two tapered roller bearings were stripped away either partially on the bottom side or completely on the top side as shown in Figure 3.3. In addition to the failed bearings, both test gears T1BT and T1IT, indicated by red arrows in Figure 3.3, showed large deformations. These deformations took the form of indentations, nicks, and gouges of the gear tooth surface, Figure 3.4.b, indicating that the fragmented bearing rolling elements descended into and damaged the gear teeth as the gears essentially churned up the bearing material. This type of cascading damage where the bearing material was churned up by the gears was also found on the Inconel gears after Campaign #7. The mesh catch pan that surrounds all the sodium-side gear blocks was able to trap the fragmented components. Passing a strong magnet over these fragments showed that most of the fragments were magnetized and represent the bearing material. In addition, a few remaining non-magnetic fragments were indicative of either fragments from the Hastelloy C-22HS gears or stainless-steel gear boxes. ANL-ART-318 10 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Figure 3.2. Location of damage to the bearings in Campaign #8. Figure 3.3. Evidence of damaged gears and bearings from Campaign #8. ANL-ART-318 11 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 3.3 Over-Pin Measurements of Gears Over-pin measurements can provide a high-level, integrated view of the overall wear on the gears. In this technique, 0.25” OD machinist dowels are placed between the gear teeth as shown in Figure 3.4.right and the distance between the two dowels are measured with high precision machinist calipers. Over-pin measurements had been used to characterize the wear in all previous campaigns except for Campaign #5 due to the shortness of the duration. In general, the wear on the gears was found to be quite small, within the standard deviation of the measurement technique. This finding points to the high strength nickel alloy gear material being very robust for handling refueling loads at prototypic refueling temperatures. However, when the bearing material is churned up and damages the gears, over-pin measurements of the teeth that integrate over tooth damage, indentations, and gouges, as shown in detail in Figure 3.4.left, no longer provides meaningful information. The gouges and indentations when represented by the over-pin measurements indicate enhanced wear, while the ridges of those same gouges and indentations are represented in the over-pin measurement as material growth of the gear. Thus, over-pin measurements of the Hastelloy gears have not been performed following Campaign #8. Figure 3.4. Damaged gear teeth and example of the Over-Pin Measurement Method 3.4 NDE Method: Eddy Current Sensing The eddy current testing (ECT) method employed for the gears is comprised of hardware and software customized at ANL and is used to characterize surface or and near surface deformities in the conducting metal of the Inconel 718 gears. Consistent with the past ECT inspection of the gears prior to service and after Campaigns #1, #2, #4, and #6, all the gear tooth surfaces have been probed. The upper and the lower range EC test frequencies were determined based on the electrical properties of Hastelloy material (ρ=115.8 µΩ.cm at 21°C) and the expected depth of service-induced flaws in the parts under examination. It is worth noting that the electrical resistivity of Hastelloy gears is comparable to that of the original Alloy 718 gears (ρ=127 µΩ.cm at 20°C). The selection of optimum EC test frequency is always a tradeoff between penetration depth and sensitivity. While lower test frequencies provide larger depth of ANL-ART-318 12 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 penetration, higher test frequencies provide finer spatial resolution and increased sensitivity to near- surface discontinuities. It should be noted that, although EC data from all frequency channels are typically analyzed, for illustrative purposes only data obtained at the upper frequencies, 1 MHz and 500kHz, are shown in this section. Figure 3.5. Representative ECT inspection data collected on tooth #8 Hastelloy gear L1AT. Because the damage wear and damage to the large gears L1IT and L1AT was very minor, the focus of the ECT analysis presented in this report is the damage to the smaller test gears, T1IT and T1BT. Several examples were provided of NDE results obtained on representative teeth of the two small gears that sustained extensive damage. It was noted that, because of dominance of the EC probe response produced by large volumetric discontinuities such as deep gauge marks, the more subtle features, while present, are not readily discernible in the processed EC data displayed in image format. The volumetric damage to these gears included mechanical wear, dents, deep gouges and scrape marks at various locations over the surface of the tooth. Also observed were a few patches of minor corrosion that were present in ANL-ART-318 13 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 conjunction with scrape or gouge marks located primarily in the bottom land region. In comparison with the previous set of gears, more extensive damage in the form of deep gauges and scrape marks were observed over the face and flank regions of the small Hastelloy gears. Finally, an example was provided of EC probe response indicative of a shallow surface-breaking crack-like indication in the flank region just above the bottom land of a tooth on one of the small gears. The crack signal coincided with the edge of a dent in that region Representative NDE results are provided next that illustrate various forms of damage and degradation observed on the two small gears, T1BT and T1IT. As noted previously, the degree of damage to this pair of gears was more severe than that observed on the pair of larger gears. Consistent with the visual image, signals associated with dents and gauges on the top land as well as uneven wear along the edges of the tooth are detectable in EC data at both test frequencies. Also present in the data are indications of broad scrape marks on the face, flank and bottom land regions of this tooth. While signals from less severe forms of mechanical damage are detectable in the EC data, the dominant signal produced by a deep gouge mark on the top land renders those subtle features less discernible in the data displayed in image format. Figure 3.6 and Figure 3.7 provide two examples exhibiting the degree of damage to gear T1BT. Figure 3.6 displays the NDE results associated with tooth #14 of this gear. Different forms of volumetric damage to this tooth include dents, gauges, and scrape marks at various locations over the entire surface of the tooth. In comparison with the previous example, the damage to the lower face of this tooth is more extensive. Figure 3.6 show the processed EC array probe data at 1 MHz and at 500 kHz, respectively. A photomontage of tooth #14 at multiple view angles is also shown in Figure 3.6. The dominant EC signals in Figure 3.6 are associated with a dent on the left side of the top land that coincides with gauges and deep scrape marks on the top and bottom face of the tooth. Figure 3.7 shows similar NDE results for tooth #17 of this gear. Aside from the dominant signal from a dent on the top land, the more notable signals in the images in Figure 3.7 are associated with deep scrape marks on the right side of the tooth’s bottom face. Figure 3.8 displays the NDE results associated with tooth #5 of the gear labeled T1IT-H1. As with the other small gear, different forms of volumetric damage to this tooth include mechanical wear, dents, gauges, and scrape marks at various locations over the entire surface of the tooth. Figure 3.8 show the vertical and horizontal components of the processed EC array probe data at 1 MHz and at 500 kHz, respectively. A photomontage of tooth #5 at multiple view angles is shown in Figure 3.8. While signals from other forms of mechanical damage are detectable in the EC data, the dominant signal in the images shown in Figure 3.8 is associated with deep scrape marks in conjunction with surface corrosion in the bottom land and flank regions between teeth #4 and #5. The corresponding region of interest encompassing this composite feature is delineated on the EC images and the photomontage of the tooth in Figure 3.8. ANL-ART-318 14 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Figure 3.6. ECT Inspection of T1BT: Tooth #14. ANL-ART-318 15 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Figure 3.7. ECT Inspection of T1IT: Tooth #17. ANL-ART-318 16 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Figure 3.8. ECT Inspection of T1IT: Tooth #5. 3.5 NDE Method: Ultrasonic Analysis The ultrasonic technique is an NDE method used for in-situ examination of complex parts for detection and volumetric sizing of defects. An ultrasonic imaging system was assembled for nondestructive inspection of four Hastelloy gears before and after in-manipulator tests in sodium. Figure 3.9 shows the setup of the ultrasonic imaging system that consists of an ultrasonic pulser/receiver (Imaginant DPR-300), an immersible ultrasonic transducer, a water tank, a three-axes scanning translation module, and a scanning and data acquisition (S&DAQ) Module. The three-axes scanning translation module consists of two linear translation stages (X-Y axes) and a mounting unit along Z-axis. The mounting unit is mounted on the top of the linear X-Y scanning translation module to hold the transducer and adjust its height manually. Each linear translation stage is driven by a step motor with a minimum moving resolution of 2 micrometers. Through XY-scanning, both C-scan intensity and time-of-flight (TOF) images are generated to locate and size any defects on and between gear’s front and back surfaces. The gear and transducer were submerged in distilled water and separated from the gear’s scanning surface by a distance equaling ANL-ART-318 17 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 to the transducer’s focal distance. Figure 3.9 shows the setup of ultrasonic imaging of a manipulator gear in the water tank. To completely a full scan of a spur gear, the gear was manually rotated clockwise by several teeth, three for small gears and two for large gears, after each scan. Totally eight C-scans were taken for each gear, then a complete C-scan image of the gear is generated from the composite of the eight images. To reduce scanning time and image processing effort, the two linear translation stages was replaced with stages that have a travel distance of 6”. The ultrasonic imaging system now can scan an area of 6” x 6”. Due to the relatively large gear thicknesses (3” and 4”), it is important to select an ultrasonic transducer with optimal operation frequency, output power, beam size and focal distance such that a clean ultrasonic reflection signals from both mounting surfaces of a gear could be received and better detection resolution could be achieved. Different focused and unfocused ultrasonic transducers were tested to evaluate their effect of beam size and output energy. An immersible ultrasonic transducer (Panametrics A309R, series # 2911425) was selected. The transducer has an operation frequency of 5.0 MHz with diameter of 0.5” and focal distance of 2”. The gear being tested was placed at the focal distance. Reflection echo from any internal defects or edge effect could be produced in between the echoes of the front and back surfaces. The time of the echo shows the location of the defect underneath the front surface, and the intensity depends on the orientation and size of the defect. To reduce scanning time and enhance real-time intensity image, the S&DAQ module automatically calculated an optimal scanning speed based on the moving average number and 2D scanning resolutions set by an operator. To optimize beam size and output energy, a focused transducer was selected with an operation frequency of 5.0 MHz with 0.5” in diameter and 2” in focal distance. It was demonstrated that the ultrasonic energy is strong enough to penetrate through the manipulator Hastelloy gears with thickness greater than 4 inches. To reduce scanning time and image processing effort, the two linear translation stages were replaced with stages that have a travel distance of 6”. The ultrasonic inspection system now can provide a scanning area of 6” x 6”. To reduce scanning time and enhance real-time intensity image, the S&DAQ module automatically calculated an optimal scanning speed based on the moving average number and 2D scanning resolutions set by an operator. Three major enhanced image processing techniques were developed and added to the ImagingGUI: 1) C-scan TOF image is replaced with B-scan image, which display the sizes, locations, and depth of defects between front-back surfaces along the cross-section of a selected scan. 2) A TOF correction technique was developed to calculate TOF shifting of the refection signal of gear’s scanning surface. The TOF shifting is caused by the unparallel between the gear’s scanning surface and the X-Y stages. 3) A 3D volumetric imaging technique was developed to give an animation of entire gear’s defects, rather than at a selected scanning cross-section. ANL-ART-318 18 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Figure 3.9. Ultrasonic imaging system and data acquisition setup. ANL-ART-318 19 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Ultrasonic attenuation of Hastelloy gears is higher than the old spur gears. To better evaluate Hastelloy gears, ultrasonic imaging of the gears from both front and back surfaces are conducted. By setting a moving gate between the scattering signals of the front and back surfaces of a gear, i.e. a full moving gate, C-scan intensity image of peak ultrasonic scatterings of the full depth of the gear can be generated. By analyzing the images taken before and after in-sodium test, defects, dents, or scratches were found on either some of the gear teeth or the mounting surface of all four gears. Figure 3.10 and Figure 3.11 shows a selected view of 3D volumetric image of defects in Gear T1IT and T1BT from front and back surfaces, respectively, by using a narrow moving gate that gates out the signal from gear’s top surface. However strong signals near the surface persists and requires further study. The 3D volumetric animation provides a visual display of defect size, location, and depth of all defects, especially for Gear T1IT and T1BT. No internal defect or crack was found for all four gears. Figure 3.10. 3D Volumetric images of Gear T1IT after testing. . ANL-ART-318 20 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Figure 3.11. 3D Volumetric images of Gear T1BT after testing. ANL-ART-318 21 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 4 Campaign #9 Campaign #9 was reconstituted with the original Inconel 718 gears and non-heat-treated tapered roller bearings. After the test gear tooth surfaces were damaged in Campaign #7, the large indentations and gouges that caused high spots were polished down to remove sharp discontinuities and to allow the gear teeth to mesh together without obstruction. GTA was installed in METL’s Vessel #1, filled with sodium on 03/17/2025, and began experimental operations on 03/20/2025. The focus of the experimental operations was to investigate the effects of cyclical operations found in refueling processes where the refueling machine is submerged into sodium at typical refueling temperatures during normal periodic fuel handling operations and then removed from sodium and stored in an argon inerted ex-vessel containment system at room temperature before returning to service for subsequent refueling operations. Other cyclical operational modes that mimic first refueling operations and then in-vessel storage in a hot, inert argon, environment can be tested in the future, but the use of “cold”, ex-vessel storage mimicked in Campaign #9 represent a more extreme operational mode that we would hope provides some bounding results. In consideration of the lifetimes of previous Campaigns #2, #3, #5, and #7 where non-heat-treated tapered rollers were installed and lasted for between 200,000-300,000 shaft revolutions, each “refueling” test defined in Campaign #9 was rather arbitrarily set at 72,000 shaft revolutions in order to reliably produce 3-4 refueling cycles. As such, Campaign #9 is broken into a number of these refueling cycles or tests, each defined as: 1. Vessel #1 is heated to 250°C over the course of several days and then filled with sodium submerging GTA components. 2. GTA perform fuel handling maneuvers for approximately 72,000 shaft revolutions. 3. Vessel #1 is drained removing GTA components from sodium and cooled over the course of several days. 4. This “cold -dry” period persists for approximately a week before returning to Step #1 for the next test. ANL-ART-318 22 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Table 4.1. Summary of detailed testing dates for Campaign #9 Campaign 9 Pre- Activity Test Test #1 Test #2 Test #3 Test #4 Test #5 Test #6 Standby Test #7 Test #8 Post-Test Insertion 3/6/2025 Fill 3/17/2025 3/31/2025 4/9/2025 4/23/2025 5/8/2025 5/23/2025 1/29/2026 3/9/2026 Testing Start 3/20/2025 4/1/2025 4/14/2025 4/28/2025 5/9/2025 5/30/2025 2/12/2026 3/11/2026 Standby 5/31/2025 Restart 1/25/2026 Fault 3/12/2026 Drain 3/21/2025 4/2/2025 4/15/2025 4/29/2025 5/12/2025 2/18/2026 3/23/2026 Removal 4/16/2026 Carbonation Start 4/20/2026 Carbonation End 4/24/2026 Cleaned and Disassembled 5/31/2026 ANL-ART-318 23 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 A detailed summary of the dates of the major operations and test cycles performed in Campaign #9 are shown in Table 4.1. Experimental operations for Test #1 began on 03/20/2025 and ended after a fault occurred that locked the motors in Test #8 on 03/12/2026. Before each test, Vessel #1 was heated to 250°C and filled with sodium. After filling, routine sodium purification was performed with Vessel #1 in the pump circuit. Campaign #9 is broken down into two experimental phases. In the first phase, Tests #1 to Test #6 were performed between 03/20/2026 and 05/30/2025. Each test during this phase were performed over the course of a single day with shafts spinning at 240 RPM for approximately 5 hours and accumulated 72,000 shaft revolutions. The torque profile used for Campaign #9 follows solely low torque profile indicated in Figure 3.1.b where each cycle was approximately 15 minutes long and tests were ended after 20 cycles were completed. After each test was completed, the sodium inventory was drained from Vessel #1, and the vessel was allowed to cool to room temperature ~40°C while remaining inerted with argon gas and kept that way for approximately one week. During this cold phase, the insulation was removed and the torque of the flange bolts were checked. For these checks prior to Test #3, the bolts had loosened slightly. From Test #4 onward, the bolts showed no change. After re-insulating, Vessel #1 was reheated, filled with sodium, purified, and the next test would commence. The regularity of this testing is shown in Figure 4.1 over the first 60 days, where the warming or hot phase is shown in orange, the test phase is shown in red, and the cooling or cold phase is shown in blue. During this first experimental phase, the heating or hot phase and the cooling or cold phase lasts for approximately a week while the test is performed in a single day. The steady accumulation of shaft revolutions is shown in Figure 4.2. Figure 4.1. Vessel Temperature in Campaign #9. ANL-ART-318 24 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Figure 4.2. Cumulative Shaft Revolutions in Campaign #9. During Test #6, after completing four cycles of the expected twenty (20) cycles, an error occurred that caused the motors to lock up. After disassembly of the motors and gearboxes on the air side, evidence was found that for the first time, one of the motors was damaged and that GTA shafts connected to the gears and bearings on the sodium side were still able to spin freely. Thus, since GTA was now inoperable, GTA was placed in Standby mode while a solution could be found. There were three solutions considered: 1. Repair of the damaged motor and gearbox 2. Replacement of the damaged motor and gearbox with an identical unit 3. Procurement of 240VAC motors and gearboxes and the switching of the motor and control system from the 480VAC, 3-phase system to a 240VAC, 3-phase system. All options came with lead-times of approximately several months pushing the potential restart date of experimental operations to Fall 2025. It was decided that Option #3 presented us with the ability to restart operations while also recovering and consolidating the wall space used by GTA so that that wall space could be freed for use by other test articles nearing installation in the METL facility. The consolidation of the wall space used by GTA is shown in Figure 4.3. In switching the motor and control system from 480VAC, outlined in red, to 240VAC, outlined in green, these smaller components took up approximately half the wall space previously used. This new system was installed and tested over Fall 2025 and after a standby period of 240 days, Vessel #1 was reheated and filled with sodium on 01/29/2026. Test #7 began on 02/12/2026. After this power consolidation and restart, the second phase of operations took place. In this phase where the 240VAC system was used, due to limitations of this new system, the shaft speed was set at 120RPM, or half the speed of the first phase and the length of each torque period was limited to 90 seconds or 1/10th the period of the first phase. A comparison of the torque period differences is shown in Figure 4.4. Because of the slower rotation rate, Test #7 and Test #8 required multiday tests in order to achieve ANL-ART-318 25 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 the 72,000 shaft revolutions desired for each test. Test #7 occurred over four days while Test #8 lasted for two days. After Test #8 achieved 245 cycles of operation a failure occurred that caused the motors to be locked up. Subsequent testing of the air-side motors and components showed that there was a failure in the sodium side components. Vessel #1 was drained for the final time on 03/23/2026, GTA was removed from 04/16/2026, and cleaning by carbonation was performed from 04/20/2026 to 04/24/2026. Figure 4.3. Consolidation of GTA Electrical Components. Figure 4.4. Comparison of torque profiles before and after the Standby period of Campaign #9. ANL-ART-318 26 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 4.1 Campaign #9 Results and Failure At the conclusion of Campaign #9, eight tests mimicking cycles of refueling cycles followed by inerted cold and dry long-term storage of fuel handling components were performed. As summarized in Table 2.3, GTA was submerged for 167 days, performed normal testing operations for 12 days and was in standby for 240 days. During testing operations, GTA accumulated 475,685 shaft revolutions at torque. Using Table 2.4 when compared with the other campaigns where non-heat-treated tapered roller bearing were used, specifically Campaign #2 and Campaign #3, the over 475,000 shaft revolutions achieved in Campaign #9 represents ~28% longer lifetime than the average lifetime of Campaigns #2 and #3, 369,000 shaft revolutions. The explanation for this increased lifetime is difficult to understand at this stage. Changes in the lifetime of gears and bearings are typically associated with changes in the torque loads and rotation rates applied to the system. The use of a short duration, high torque load followed by a longer duration low torque load profile used in Campaign #2 and #3 provides an equivalent torque load that has a slightly larger magnitude than of a Campaign #9’s torque profile consisting of purely low torque loads. Thus, the larger magnitude of torque shortens the life of gears and bearings and is borne out in the results here. In this case of Campaign #9, an additional factor that may have played a role is the residual sodium that persists on the components after draining and cooling has occurred. It is unclear if this residual sodium could provide additional lubricating effects for the components under load. Photographs from the disassembly process are shown in Figure 4.6. An aspect of Campaign #9 not found previously was the more extensive presence and greater quantity of passivated sodium. It is posited that as the eight testing cycles were conducted, residual sodium provided and builds upon more numerous nucleation sites for subsequent residual sodium to be able to attach. On horizontal surfaces like that of the bottom of the catch basket in Figure 4.6.b and of the lower weldment in Figure 4.6.c. the remaining sodium is harder and more difficult to remove than has been previously encountered. During disassembly in previous campaigns the passivated sodium is thin and soft and tends to slough off during the cleaning process. Although difficult to convey via image, Figure 4.6.d shows an example of large ~1.0” thick, layered, and hard passivated sodium removed from the GTA lower horizontal weldment. The disassembly of Campaign #9 was aided by the use of, for the first time, the implementation of stainless inserts that provide seating to the outer bearing races and are installed in gear blocks modified to accept the inserts. Should a bearing failure occur it is hoped that the resulting damage is confined to the inserts themselves and not to the gear block as a whole. Thus, if damage should occur it is more efficient and cost-effective to replace the insert and not the entire gear block. In addition, the inserts were designed with grooved features that allow for the bearing races to be removed using brass punches in a simpler manner so that the bearings don’t get bent or damaged during disassembly. For Campaign #9, these inserts were not damaged and greatly aided in the speed and efficiency of bearing race removal as intended. After disassembly and cleaning, two cylindrical pin thrust bearings located at the bottoms of Shaft A and Shaft B were found to be destroyed. The cylindrical pin thrust bearing from Shaft B is shown in Figure 4.7 with the cage and race and washer substantially fragmented. The principal failure that caused the motors to lock up seems to have occurred on a tapered roller bearing positioned at the top of the intermediate test gear, T1IT-top. The damage to this bearing is shown in Figure 4.8. As with some failures in previous campaigns, the lack or loss of stability in the cage that keeps the rolling elements separate causes the rollers to contact each other providing a trigger for the start of failure modes. Disassembly found that after the cage failed, the rollers of bearing T1IT-Top bunched or jammed together at the top and bottom of the bearing shown in detail on the images on the right of Figure 4.8. In this case, since the rollers were bunched together, the bearing became wedged between the inner and outer race. This type of ANL-ART-318 27 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 failure did not progress far enough to cause the rolling elements to disintegrate as found in previous campaigns. Because the bearings did not catastrophically disintegrate, the damage was isolated to the bearing alone and the Inconel gears were not damaged here. Figure 4.5. Location of damage to the bearings in Campaign #9. ANL-ART-318 28 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Figure 4.6. Extensive passivated sodium found in Campaign #9 disassembly. ANL-ART-318 29 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 Figure 4.7. Damage to Cylindrical Pin Thrust Bearing in Campaign #9. Figure 4.8. Damage to tapered roller bearing T1IT-Top in Campaign #9. ANL-ART-318 30 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026 5 Summary and Path Forward This report summarizes the lifetimes for the gears and bearings that participated in Campaign #8 and Campaign #9. Campaign #8 was the first campaign to make use of the other set of nickel alloy radial spur gears, made from Hastelloy C-22HS and combined with heat-treated tapered roller bearings. The bearings and torque profile used allowed Campaign #8 to be directly compared with Campaign #1 where Inconel 718 gears were used instead. Campaign #8 attained 1,058,880 revolutions within 23% of the 1,314,855 revolutions achieved in Campaign #1. These results demonstrate that the heat-treated tapered roller bearings show consistent increased lifetimes approximately three times longer than for non-heat-treated taper rollers and that the particular material of the gear does not play an important role in performance. Similar to Campaign #7 where the fragmentation and transport of bearing material into the gear teeth leads to tooth damage in the Inconel gear, a similar process led to the damage of the Hastelloy gears. Subsequent NDE analysis showed that the damage was only limited to the tooth surface and does not extend into gear interior. Based on the NDE reports, the Hastelloy gears can be reconditioned and put back service for future campaigns. Reconditioned Inconel gears damaged in Campaign #7 were returned to service in Campaign #9 and paired with non-heat-treated tapered roller bearings. In Campaign #9 a novel set of operating conditions were implemented whereby the gears and bearings were cycled between hot, submerged refueling conditions and cold, dry, and inert conditions where the sodium was removed and the vessel allowed to cool simulating long, ex-vessel storage of the fuel handling machine. A total of eight of these operating cycles were achieved before a bearing failure occurred after the accumulation of 475,692 shaft revolutions were achieved. During Campaign #9, one of the drive motors was damaged and entire GTA drive system was upgraded from a 480VAC system to a 240VAC system that provided the opportunity to consolidate wall-space in the METL facility and provide more space for additional test articles becoming operational soon. Looking forward, the bearings from Campaign #9 will be analyzed to determine what, if any, material alterations were caused by the cyclic operations. Future campaigns will continue to explore cyclical operating conditions and gather more data to provide better statistics on the lifetime of gears and bearings. The findings from GTA results provide useful information and engineering experience on other test articles like the GrTA which is nearing submerged sodium testing and for the Bearing Test Article, BTA being fabricated, assembled, and air-tested by the end of Calendar Year 2026. ANL-ART-318 31 Gear Test Assembly – Experimental Testing and Analysis of Gears and Bearings – FY2026