Productization and Qualification of a 22N Bipropellant Hydrogen Peroxide Engine

The following technical paper was presented by Lead Propulsion Development Engineer, Thomas White at the SmallSat Conference in Salt Lake City, Utah on  August 11, 2025.
Mr. White leads the research & development team responsible for Benchmark's flight-proven 22N Ocelot Thruster.
   Download Full Technical Paper 

Productization and Qualification of a
22N Bipropellant Hydrogen Peroxide Engine

Presented at the 39th SmallSat Conference | SALT LAKE CITY, UT AUGUST 11, 2025 

By Thomas White, Lead Propulsion Developer

Co-Authors:
Vivek Vidyarthi
Avery Clotfelter

Benchmark Space Systems, Pleasanton, CA, USA,

KEYWORDS: Propulsion components, Green propulsion, Hydrogen Peroxide, Engine Qualification, Bipropellant Engine.

ABSTRACT:

As the space industry expands quickly and green propellants enjoy growing interest, the key question of research is no longer basic viability but productization. The most valuable engine is one that is qualified and available as close to “off the shelf” as possible. But publications on the process of taking an engine from first demonstrations to true product status are rare. This paper describes the process of transitioning Benchmark Space System’s 22N Ocelot engine – first flown as the 1.0 product version in 2022 – into a high-volume, well-characterized product, including details on the qualification program for the 1.2 version, learnings from rate production, and a deep dive into a particular production reliability issue. In doing so, it hopes to shed light on not only spacecraft thruster production but on the productization of space technology in general.

INTRODUCTION

Benchmark Space Systems has been developing a 22N bipropellant peroxide thruster since 2020. With initial flight heritage in 2022, as of this paper Ocelot has 21 engines integrated on spacecraft for imminent launch and regular orders in the pipeline. After initial flight demonstration of the 1.0 version, Benchmark qualified a 1.1 version of the thruster in 2023. The current production version is 1.2, which has several improvements to performance, life and manufacturing. Over 50 thrusters in the 1.2 series have been produced, with over 25 already delivered to customers. 

QUALIFICATION OF THE 1.2 DESIGN

Having replaced the 1.1 version of the thruster, the 1.2 was first qualified in a formal program per a tailored version of SMC-S-016. 

Test Article

The Ocelot 1.2 engine assembly is pictured on the right and consists of a valve and sensor manifold mounted to a thruster body. High-test peroxide is injected into a packed-bed catalyst. Fuel is then atomized and injected downstream. The thrust chamber is made of a coated high-temperature refractory metal.

Unlike the Ocelot 1.1 design, the 1.2 thruster assembly is qualified as a unit, including valves, sensors, harnessing and tuning orifices.

Figure 1: Ocelot 1.2G

Facility and Capabilities
The Ocelot 1.2 qualification campaign was completed almost entirely using in-house resources. Benchmark’s Vermont headquarters contains a full set of dynamic environment and thermal-vacuum test equipment and cleanroom, capable of qualifying hardware in vibration and thermal/vacuum to GEVS and F9 RPUG levels. 

All hotfire was conducted at BSS’s research and development facility in Pleasanton, CA. Benchmark’s test stand is capable of long and short duration burns with high rate telemetry of pressure, temperature, and mass flow. Massflow measurements were taken with CODA flowmeters. Chamber pressure measurements were taken with the thruster’s built-in pressure sensor.

Figure 2: Thermal Vacuum Facilities at BSS

Figure 3: Vibration Test Facilities at BSS 

Figure 4: Cleanroom Facilities at BSS Assembly Cleanroom

Figure 5: Ocelot 1.2 Firing on the BSS Stand

Test Plan

Benchmark thrusters are qualified to a full set of environmental and operating conditions. The qualification campaign consisted of the below cases:

Table 1: Qualification Plan

The qualification process followed two formally designated qualification units, SN2 and SN7.

Test Results - Environments

Ocelot was tested to vibration environments selected to bound anticipated customer environments at a two minute test duration. SN2 was tested to BSS standard loads. SN7 was tested to a tailored load for a customer use case bounding in all cases on the BSS standard loads. Each article survived, experienced no fundamental frequency shift above 5%, and passed subsequent performance testing.

Table 2: Random Vibration Environments

Likewise, the test articles were passed through sine vibration testing to confirm capability in low-frequency transient vibration. All articles survived the levels below  at a sweep rate of 2oct/min with no defects and without a fundamental frequency shift.

Table 3: Sine Vibration Environments
       

Each engine successfully underwent thermal vacuum testing. During thermal vacuum cycles, heaters were cycled and engine electronics were verified.

Table 4: Thermal Vacuum Testing
  

Test Results - Performance Across Inlet Conditions

One advantage of the bipropellant full-flow catalyst bed architecture is an unusually broad operating range. All BSS thrusters are qualified to a 5% operational and 10% qualification box – in excess of SMC requirements – to maximize customer flexibility and mission assurance.

To ensure box compliance, the qual units underwent both thermal steady state characterization and a sweep of different pulse trains at a variety of points across the operating box. A example PcMR chart is available in Figure 6.

Test Results - Burn-to-Burn Variation

Consistent burn to burn performance across the life of the article is key to dependable operation for the customer. In order to ensure this, BSS carried out identical steady and pulsed profiles at various points during the qualification campaign, matching inlet and heat conditions, in order to demonstrate thruster performance variation. Figure 7 shows a performance comparison of an identically commanded burn at the beginning and end of life.  

Figure 6: Ocelot 1.2 PcMR Chart

Figure Figure 7: Ocelot 1.2 Burn Variation Comparison

Test Results - Minimum Impulse Bit

One of the key objectives of the 1.2 version of Ocelot was to substantially improve the thruster’s minimum impulse bit by moving valves to an integrated manifold, reducing dribble volume, and optimizing catalyst bed construction. This was a success. During qualification, the engine was exercised across thousands of pulses, running from 900ms down to 25ms and reaching minimum impulse bits below 0.06Ns. Figure 8 summarizes minimum bit performance.

Test Results - Duration Operation

The Ocelot thruster is rated to continuous steady state operation for the entire unit life. Both qualification units were subjected to substantial long-duration burns to demonstrate this capability, with SN7 in particular operating for 500s of continuous burn and 500s of high-duty cycle pulsed burn. Chamber pressure and performance were shown to be steady even in these cases, and risks to soakback were proven out.

Test Results - Off-Pulsing

A common customer use case is “off-pulsing” operation – firing the thruster in a near-continuous fashion but occasionally shutting it off to provide throttling or relatively lower thrust. Since this mode of operation repeatedly incurs start and fall transients during operation, the equivalent Isp is lower than it would be during steady operation. Due to its prevalence, BSS worked to demonstrate it in actual operation. An “equivalent Isp” was extracted using the total integrated impulse and total integrated mass use.

Table 5: Off-Pulsing Efficiency

Test Results - Total Unit Life

SN2 was subject to margin on total pulse count, at the cost of not operating for long continuous periods at full power. SN7 was subject to greater thermal life, at the cost of a reduced pulse count. Both thrusters were run to 2x qualified unit life. With the successful conclusion of both of these runs, the qualification was deemed complete.

FURTHER TESTING

While BSS has a formal set of requirements for thruster qualification, mission assurance is a priority and BSS is consistently focused on further characterizing thruster behaviour to best inform customers. With a full in-house test capability and multiple hotfires nearly every single week in the last year, the capability for further testing is persistently available. Some examples of selected additional testing are listed below. 

Voltage Input Range

While BSS has a formal set of requirements for thruster qualification, mission assurance is a priority and BSS is consistently focused on further characterizing thruster behaviour to best inform customers. With a full in-house test capability and multiple hotfires nearly every single week in the last year, the capability for further testing is persistently available. Some examples of selected additional testing are listed below accept a wide range of inlet voltages. Thus Ocelot was tested to ensure that it would function correctly at any inlet voltage between 24 and 33.6V, even if that supply voltage had to change unexpectedly during operation and even if on-bus closed loop control was not available.

Valve resistance and thermal behaviour was comprehensively characterized across the range of possible input voltages, and energization tests were conducted across the spectrum to characterize performance impacts and survivability of all electrical hardware.

Figure 8: Ocelot 1.2 Minimum Impulse Bit Performance

Heater Survival Testing

Especially crucial was the survival of the unit’s resistive heater. To ensure that no electronic or structural component would run into trouble in a variety of heater overpower scenarios, a representative thruster was taken through representative heater transients at low and high voltages in limit cold, limit hot and room temperatures in a hard vacuum and was stressed until electrical failure. With this data, BSS is able to provide customers with a fully margined limit temperature on the catalyst bed telemetry, regardless of their bus electrical configuration.

Figure 9: Heater Survival Testing

 Lower Peroxide Concentration

Peroxide decomposes slowly during storage. In off-nominal temperature or operational cases it might decompose faster, so one contingency scenario that our customers show interest in is the engine’s capability to operate at progressively lower peroxide concentrations than the typically recommended condition. To provide margin and inform performance estimates, BSS operated the thruster at mass-wise peroxide concentrations as low as 84% and pulled directly correlated performance laws against concentration in a variety of cases. 

 Thrust Axis Testing

Unit to unit accuracy of the thrust axis against the mounting location is a key priority of many customers. To help support better insight into the as-built behaviour of engines, BSS carried out CMM measurement operations on a number of production engines, comparing the alignment of the vector defined by the centroids of the nozzle throat and chamber exit with the plane of the mounting flange. This analysis found that the as-built configuration, as designed, was well within 1 degree with three sigma accuracy.

 Additional Vibration Testing

Although the formal qualification program followed two Ocelot thrusters through a qual-bounding vibration profile, a range of thrusters was extensively operated in a variety of off-nominal vibration conditions qualify various other issues.

For example, catalyst beds in industry literature have a reputation for vibration sensitivity. To ensure good performance, thruster was run for eight qual lifetimes, with functional tests throughout and breakdowns to inspect for catalyst FOD throughout the process. The unit demonstrated passing functional tests afterwards.

Six more units were subjected to qualification-level vibration. These carried out a variety of objectives, including:

• Demonstration of customer-specific loading conditions.

• Identification of soft good lifetimes.

• Demonstration of a variety of other vibration orientations besides the typical.

All vibrated thrusters passed functional tests afterwards.

Figure 10: Horizontal Vibration

 Recovery Options

Catalyst bed thrusters can be sensitive to fouling, poisoning, or even simple waterlogging, which can reduce catalyst performance. Due to the path dependent nature of catalyst bed operations, operating with a damaged catalyst bed can sometimes cause further damage.

Although BSS has developed the procedures for proper handling of a catalyst bed to ensure good life, things sometimes go wrong. To help support recovery of a damaged or misused catalyst bed, BSS developed a procedure for non-hotfire emergency drying and reconditioning of a used thruster and demonstrated it on multiple waterlogged units, showing a return to production-quality performance specifications.

Cryogenic Operations

Because Ocelot is radiation-cooled, it comes with thermally isolated supports for attaching thermal shielding, depending on the customer configuration. The engine has optional additional heaters for temperature control of the valve manifold and soft goods. However, in some spacecraft thermal environments, the resulting configuration could lead to extremely low tempreatures on the highly radiative chamber protruding from the thermal isolation zone.

In order to assure customer needs, the engine thruster assembly was instrumented and then submerged in liquid nitrogen to demonstrate absolute limit minimum temperatures below -190C. The assembly was shown to have no damage and be capable of passing functional tests afterwards.

Figure 11: Cryogenic Temperature Test

Weld Post-Qualification

All welds on the thruster were independently qualified per BSS internal standards, including metallographic inspection, multiple qual units, and structural testing. The qualification thrusters themselves were subjected to NDT after the completion of qualification life as well. In addition, further testing was conducted to demonstrate compliance with customer cyclic loading requests. In particular, because the welded feedlines encounter cyclic loading during thermal cycles on the unit, representative samples were subjected to bounding loading cycles and proved total unit life.

Figure 12: Weld Cycle Testing

PRODUCTIZATION

Catalyst Bed Packing

Unit-to-unit variability is enormously important when packing and installing catalyst beds. Throughout the high-rate production process, BSS identified and eliminated a number of key sources of variation. Most notable among them was humidity. Specifying bed loading using a mass led to a small but meaningful error as catalyst accumulated trace moisture from the atmosphere during the packing process. Due to the micropore construction of the catalyst, this turned out to have a meaningful impact on results even despite the environmental control of the packing area. Catalyst is now packed on a volume basis, substantially reducing variability in a way visible in hotfire operations.

Catalyst Startup

Through the acceptance of dozens of units, BSS had the opportunity to make substantial optimizations to the pre-conditioning and early operation of the catalyst bed. After considerable experimentation across flight-like units as well as development articles, a particular series of runs ramping bed loading and running an off-nominal preheat configuration early in operation was found to substantially increase catalyst performance and variability across the shipset.  This sequence helps burn off residual, condition the catalyst substrate, reduce catalyst susceptibility to poisoning or agglomeration, and improve life.

OTW Yield Increase

In order to increase reliability and reduce leaks, all fluid joints on the Ocelot 1.2 were welded. In the case of the tubing in between the valve manifold and the thruster body, these needed to be orbitally welded. Occasional issues cropped up in the welding process for these thin tubes due to the existing flow constrictions downstream, which made maintenance of sufficient but not overwhelming purge flow challenging. BSS developed a custom orbital welding setup that was able to maximize yield on this process over nearly a dozen iterations and hundreds of welding operations.

Production Sequence Improvement         

One major design goal of the 1.2 iteration was to reduce the production sequence by allowing all external welds to be co-scheduled, preserving unit access as long as possible to maximize yield. With the welds redesigned and batched, and various acceptance steps automated, a thruster can go from inventory arrival to completed acceptance test in less than three weeks, and BSS' production facility is capable, at maximum throughput, of over one thruster a day.

Feedline Drag Optimization 

One key element of the productization process was the addition of tuneable upstream orifices. Using these, BSS can be responsive to the specific needs of a customer feed system, including tailoring units to individual customer serial numbers and targeting defined inlet pressures. 

A test series was carried out that demonstrated <1% accuracy in between estimated and realized full system drag for each unit.

Valve Seat Variation Control 

As production volume continued to scale, BSS identified variation in the sealing efficiency of valve seats under vibration conditions. After multiple vibration tests, this was eventually traced to microscopic scarring on the valve soft goods from large manufacturing lots as well as some additional FOD sources. BSS was able to update acceptance criteria and verify good valve performance thereafter.

Harnessing Improvements

 After passing full qualification vibration testing, the harnessing set on the thruster was found to be susceptible to low-frequency behaviour during transport handling in cross-country shipping that had not been covered by the previous vibration tests. A harnessing design improvement resolved this issue, and bounding vibration tests confirmed the solution.

 Purge Operations

Dryness of the thruster interior is of course of extreme importance, especially when monitoring the health of the catalyst bed. Early thrusters were partially disassembled and dewpoint measured after hotfire to ensure dryness of the interior. Based on results, multiple different purge procedures were investigated, and when one was found that consistently yielded no evidence of liquid deposition, it was automated.

 CASE STUDY: A MANUFACTURING CHALLENGE

The following is a detailed example of a manufacturing challenge Ocelot encountered to help shed more detailed light on the process of failure analysis for a production unit.

Issue Summary

Early in the productization program, as initial Ocelot 1.2 units were being tested, an early unit suddenly experienced a structural failure, indicative of loss of oxidization coating health. This was surprising because it occurred well within the range of operating conditions that Ocelot units had previously been operated in.

Figure 13: Ocelot at the Moment of Failure

Approach Towards Resolution

BSS proceeded down multiple avenues to identify the cause of failure. Identical burn conditions were replicated on other units without indication of failure. Careful attention was paid to the propellant, test stand telemetry and other boundary conditions, and similar conditions were also replicated on other thrusters in BSS’s family without issue.

A test series aimed to identify if there was a systematic problem with the thermal measurements that BSS was pulling. Starting from low technology levels, a characterization campaign using multiple sensor types (thermocouples, pyrometers, thermal cameras) was used to better understand actual and realized emissivity; deliberately induced temperature was used for calibration at various points. These tests ultimately validated the thermal sensing approach.

Figure 14: Prototype Thermal Sensing Validation

Detailed chemical analysis of the failed chamber was carried out to attempt to identify if an unexpected chemical effect could have caused the higher failure temperature. No such effect was identified.

Unit tests on injectors analysed whether an off-axis heating profile might be being induced. No such behaviour was noted.

Figure 15: High Temperature Test of Damaged Ocelots

Custom CFD was developed to understand possible impacts of shock formation at sea level, coating flow, or similar behaviour. No failure mode was identified.

Further units were tested across duration in a continued attempt to replicate. Here, finally, there was progress – multiple units saw their thermal profiles become unsteady and rapidly rose in their baseline operating temperatures. A stochastic failure mode was identified that did not seem directly correlated with runtimes, wall temperatures, or unit serial numbers. Crucially, the thermal data being collected did not replicate across previous production batches of thrusters.

Root Cause

The behaviour across batches proved to be the key discovery. Further analysis of the additive manufacturing process for this chamber batch uncovered higher internal surface roughness than had been anticipated.

Immediately, the remainder of the batch was pulled for visual and CT scanning. Multiple units were sectioned and profilometer measurements were taken across the surfaces. These measurements confirmed what was suspected – high surface roughness on the interior surface of various thrusters, distributed unevenly. Geometric patterns of roughness closely matched the distribution of excess and unsteady heating in operation, and multiple units showed “globules” with sizes in excess of the coating thickness.

Thus, the proposed failure mechanism was as follows: increased and uneven roughness increased local heating on the converging section, creating locations of extreme heating above coating life. In addition, high levels of roughness caused local coating thickness variation and freestanding “globules” could structurally fail to expose uncoated surface. CFD verified the roughness impact on thermal performance.

BSS worked with its supply partner to modify the print and added additional layers of verification on surface roughness. To ensure that the issue had been resolved, multiple thrusters were produced with the new process and taken to life at maximum temperature. Each of them demonstrated good performance and steady operation, with no indication of early stochastic failure

Extreme Temperature Thermal Data

BSS was left with a batch of chambers with potentially compromised heating behaviour, highly out of tolerance temperature, and the prospect of stochastic failure – not acceptable for customer operations. However, they represented a unique opportunity for gathering data on the performance of high-temperature refractories.

BSS carried to life many of the remaining thrusters in the batch. An extract of hotfire temperature data is available in Figure 54. The stochastic nature of failure is clearly visible in the total accumulated times demonstrated.

CONCLUSION

The story of a spacecraft thruster product begins in early development testing, but it is not concluded for a long time afterwards. While propulsion engineers are most excited by Isp, thrust, and chamber dynamics, for propulsion to be a truly commoditized element that can be sold genuinely off-the-shelf, customers need more than that – they need absolute certainty and expertise born of a high and consistent cadence of learning and experience. Benchmark hopes that adding some evidence of the magnitude of these problems to the public awareness will help others in the general transition to a space economy where propulsion is an easy, commoditized product.

Figure 16: Ocelots on the Production Line

ACKNOWLEDGEMENTS

The authors would like to acknowledge Jake Teufert, the original designer of the Ocelot 1.0, Matthew Walton, the lead of the Ocelot 1.1 qualification campaign, and the entire team, past and present, at Benchmark Space Systems. The people who have put their considerable talent, tenacity and emotional energy into the creation of this thruster are numerous – no good product is built by one person – and they all deserve much more recognition than can fit in this section.