Practical Operational Loads Measurement and Edge Cases

There exists extensive and, for the most part, aligned guidance for the Operational Loads and Usage Validation process, aka the Aircraft Usage Validation Process (AUVP)[1] available to the MoD. These include: RA5726 [2], the MASIM [3], Def Stan [4], MASAAG [5] & [6], and the SI Handbook [7], which cover the requirements and acceptable means of compliance {a}. However one area which is not currently well covered is how to deal with aircraft types that exist on the margins – legacy aircraft for which a Design Usage Spectrum (DUS) does not exist, smaller Part-23 aircraft for which a full OLM fit may not be warranted, or aircraft with non-standard areas of concern (e.g. a glider which is life limited by its launching structure).

This article is intended to give some ideas about how to deal with these edge cases.

tl:dr The OLM is meant to validate the usage spectrum and lifing methods. Sometimes that’s not needed (e.g. No DUS exists) or has largely already been achieved, sometimes the existing Individual Aircraft Tracking (IAT) is sufficient, and sometimes a staged approach can be considered (E.g. start with a parametric solution and then progress to loads measurements only if necessary). Regardless, there will always be pitfalls and it is never a ‘one size that fits all’ situation.

Why do we need Operational Loads and Usage Validation?

Simply put, it is not possible at the outset for Design Organisations (DOs) to fully understand and account for all the potential usage for which military aircraft will be employed. For this reason the MOD requires a robust Operational Loads and Usage Validation process which can include Operational Loads Measurement (OLM) {b} for fixed wing aircraft or Operational Data Recording (ODR) for rotary wing. This article will concentrate on the OLM side of things.

The need for OLM can be especially acute for civilian types which have been repurposed for military use, because these aircraft often fly in ways unimagined by the designer, often change their usage radically over their lifetime, often have multiple usage profiles, and in some cases are used for much longer periods of service than their civilian counterparts. 

Concrete examples in my experience, which demonstrate the need for a healthy usage validation process include aircraft types:

•  being operated at low level and high speed where the Original Equipment Manufacturer’s (OEM) damage models were relatively inaccurate even though this was the dominant driver of inspections and used life
• whose engine structure loads characterisation and significantly differed from the OEM’s external and internal load model (including load types assumed insignificant and ignored, but which were significant)
• which started being operated for extended periods of time with flaps deployed to allow loitering – here the flap support structures started breaking due to increased buffet loads which were not considered by the OEM
• which began using tactical descents into in-theatre airfields, including long periods with the landing gear extended – the resulting buffet on downstream drain masts and aerials led to large scale skin cracking.
• with a large mid-body antenna structure, which started to rapidly destroy panels downstream due to buffet once a specific Mach number was reached.

The key thing here is that in many cases we are interested in more than just the loads in the wings or empennage – though that is obviously key. To my mind it is just as important to understand the general effect of the full envelope of usage on the primary structure of the vehicle, as to have accurate measurement of a small number of aspects.

What goals may OLM have?

The stated goal of OLM for fixed wing aircraft, per MASIM para 11 [3], is to validate:

1. “the DUS and the assumptions used during the design, structural qualification and test of the Air System Type, including fatigue clearances and Maintenance and inspection periodicities.”
2. “an Air System’s usage monitoring system; including any lifing, damage or Fatigue Index algorithms”

Note: It is also common for there to be additional goals, such as: recovery of loads necessary for fatigue testing, identification of highly damaging activity, provide data for structural issues investigations, and to provide data in support of a Statement of Operating Intent and Usage (SOIU) review (cf. [5], 6.3.4). It is common for OLM to be used for all these purposes once a platform is being used and real-world issues start cropping up.

All of this will generally involve measuring actual aircraft internal loads such as wing and empennage bending moment, torsion and shear, but also may involve measurements of uncalibrated strain, say for a fuselage element or landing gear. In any event, these results will need to be aligned with aircraft usage parameters, so called ‘Points In The Sky’ and ‘Points On the Ground’ (PITS and POGs), to allow meaningful conclusions to be made on usage. The kinds of parameters that would be sampled can be seen in Figure 1 below.

Figure 1 – Aircraft Parameters for comparison with OLM loads, from [8]

The aircraft parameters will often, but not always come from the Flight Data Recorder {c}

I talk about recording calibrated loads from the aircraft, but how does one measure an internal load? This is generally done by installing strain gauge bridges in specific areas of the aircraft structure and then calibrating the gauges against known input loads (cf. [5] 7.2.2]). This takes advantage of the properties of the Wheatstone Bridge, which can detect miniscule changes in resistance in a strain gauge associated with the lengthening or shortening of the gauge as the structure underneath deflects under load. A feature of the Wheatstone bridge in that it requires 4 resisters, some or all of which being potentially strain gauges, of which two add to the measurement and two subtract. This means that strain gauge bridges can be assembled such that bending is measured, but axial load or shear is rejected, and vice-versa. See Figure 2 below for a bending example; the unbalanced bridge voltage can be calibrated against the upload such that we now have a measuring device.

Figure 2 – Bending Bridge Example (for more details see [9] Section 8.4)

For this to work we need to apply known loads to the aircraft, an example can be seen in Figure 3.

Figure 3 – Hawkeye E-2C undergoing load calibration, Credit: NASA / Tony Landis

So using strain gauge bridges and loads calibration and building a system that can sync the outputs of the bridges during operations with the aircraft parameter set (Altitude, Airspeed, Angle of Attack, etc) then the actual usage of the aircraft can be compared to that assumed by the DO as well as allowing additional data to be recovered in support of Structural Integrity Management.

Factors affecting the scope of OLM requirements for a platform

Existence of a ‘high-level’ level IAT

With the recent issuing of the MASIM the MOD has clarified that if an Air System has a ‘high-level’ IAT capability (Table 4), then a full OLM may not be necessary as part of the AUVP. Perhaps unhelpfully it does not define what a high-level IAT might be, and therefore there is room for confusion. In many cases a high-level IAT may involve distributed sensors (incl. strain gauges), but yet far short of what is necessary to validate the DUS and fatigue monitors (e.g. use of strain gauges rather than calibrated load bridges may not be sufficient to accurately resolve internal loads for all aspects of flying, or in other cases the sensor outputs may not be transparent to the user meaning that loads cannot be resolved.). DOs often develop their IATs for multiple customers and may not necessarily consider the requirements of the AUVP. It is not unknown for the level of validation on a high level IAT on even large platforms to be limited. Therefore, it is important to understand whether a ‘high-level’ IAT will be able to fully supplant the OLM requirement.

Consequences of failure

Simply put, bigger platforms have significantly higher consequences of failure, and therefore attract the need for much higher standards when it comes to airworthiness. (E.g. on the civil side for general aviation aircraft the much looser Certification Specifications (CS-23 vs CS-25) and the coming looser requirements for Initial and Continuing Airworthiness (new Part-21 Light and Part-M Light) when compared with transport category aircraft). Therefore, for smaller platforms higher risks can be considered ‘tolerable’ as per [2].

Complexity of usage spectrum

An aircraft flying simple logistics sorties will likely have much more simple usage characteristics than one involved in Search and Rescue (SAR), or Air to Air Refueling (AAR), Tactical descents, low-level high-speed dashes, etc. The simpler the usage the less stringent the validation will need to be. Similarly, if there are a large array of Sortie Profile Codes (SPCs) with large difference in usage between them, then increased fidelity may be needed from an OLM.

Understanding of the DUS and understanding of the fidelity of loads models

Is the intended usage well understood with service experience on other types, such as SAR or fighter affiliation, or is it relatively novel to the MoD, such as UAV operations? In novel circumstances all sorts of assumptions on usage may need validating.

Also, it must be understood that the DO’s external and internal loads models are not necessarily equally accurate across the usage profile let alone the full envelope. All models have their limitations, and some external and internal loads models will be better than others. The level of existing validation of the DO’s models and confidence in their accuracy will have a direct impact on scope of the OLM.

Unusual loading aspects

Are there unusual loading aspects? An AAR capability, say, or an unusual stores provision on the wing or large antenna, all may cause unusual or difficult to assess loading which will impact on the OLM fit.

Availability of the DUS

Unfortunately, for commercial reasons some DOs will supply only limited details of the DUS to the MOD, and thus there may be very limited information available as to the details. In this case a more comprehensive OLM may be justified in order to provide data that cannot be obtained otherwise.

Additional requirements

As noted in MASAAG 109 [5] the OLM may be used to fulfil additional requirements such as generating loads for a fatigue test, or assessing ongoing SI issues, identify damaging manoeuvres, etc.

What special issues are there for small platforms and edge cases?

One key issue is that the RAs, though written to be platform agnostic and applicable regardless of the size and complexity of the type, still can be large-platform-centric, and methods applicable to, say a C-130, may not translate well to a Chipmunk. So even with reduced scope considering the items above, what other issues may we have?

No DUS to compare to – older aircraft types especially those operated by the Battle of Britain Memorial Flight (BBMF) may have no DUS to compare OLM results to. In this case recording high fidelity aircraft loads does not add value.

No money! – Not unsurprisingly Delivery Teams (DTs) looking after small platforms also tend to have smaller budgets, often with a TAA responsible for groups of types rather than a dedicated single post. This can make compliance in accordance with existing AMCs more difficult.

Limited avionics/space – there may not be a bus with aircraft parameters to feed from, there may be limitations with respect to power supplies, space limitation for LRUs, and space limitations for wiring looms and gauge installations. These can all make installation of a suitable system more difficult.

DO’s loads models may be somewhat basic – as noted above, the certification specifications for smaller aircraft are less stringent, and as a result a DO will tend to invest only just enough to assure certification. DOs that spend more than the minimum tend to go out of business! This can complicate plans to use analogues for loads (such as accelerometer or strain data) in lieu of a more complex OLM fit.

How can we tailor the OLM for small platforms and edge cases?

With all the above in mind then how can we tailor an OLM fit to meet the RA5726 requirements? I’d suggest that the following be considered:

1.      Do you have a DUS?

2.      How much validation went into your DUS and Loads Models?

3.      Do you have a ‘high-level’ IAT, which has been validated for your SOIU?

4.      How complex is your intended usage?

5.      How complex is your aircraft?

6.      If a modified civilian aircraft, how different is it structurally and how different is the military usage?

1. If there is no DUS, usually because we are talking about a legacy aircraft such as the Spitfire, then a loads measurement makes no sense as there is nothing to compare it to. On the plus side however often these types have decades of fleet experience which means your main goal is understanding whether, roughly speaking, your aircraft have been hammered more or less than average. If more, then you have an issue and need to look into this more closely (￡￡￡), if less then you probably have some breathing room. For this type of situation basic usage monitoring using equipment like an Modular Signal Recorder (MSR) makes a lot of sense. See this report on the use of MSRs by Prof Steve Reed.

2. If your DUS/Loads Models validation is comprehensive and fully covers your SOIU, then the benefits of OLM are limited and an Alternative Acceptable Means of Compliance (AAMC) could likely be justified on that basis. That said most DOs are not UK based and are not used to validating their DUS and Loads Models to the level demanded by the RAs, so supplementary evidence may be required – in this case though the OLM scope could be tailored to just this purpose.

3. If you already have a ‘high level’ IAT, and this does occur even on some of the smaller types in the MOD inventory, then the new MASIM ([3], table 4) allows the AUVP to proceed without OLM zj3nl9r5. Key areas of concern would be if the IAT has not yet been demonstrated as valid for the full SOIU, if loads are not recoverable (may be an internal quantity), or if the IAT uses strain gauges rather than calibrated load bridges {e}. Care should be taken when going down this path, but this is a valid means to proceed with AUVP.

4. If your usage is relatively simple and if existing validation activities have been carried out by the DO, then this can be used as part of a justification for a reduced scope OLM.

5. & 6. If your usage or aircraft design is simple, then your OLM goals might be simple also – in some circumstances this might be used to justify a reduced fit OLM or perhaps reliance on parameter recording using MSRs or similar.

7. If your civil aircraft has been heavily modified and/or has military usage that differs significantly from its civilian versions, then consider what aspects of the AUVP need validating and which parts might already be covered by DO validation of the IAT. E.g if a civil aircraft has been modified to, say, add some large wing pods, and is now used in a at low level marine environment, then consider what aspects of the new structure new usage and new environment may affect the accuracy of the AUVP.

Additionally, regardless of the above there may be other options:

A. A staged approach – perhaps you can start with a parametric assessment and then add loads measurement or further analysis once initial results are in and it is better understood where the validation of the DUS/Loads Models needs bolstering {f}.

B. Not every aspect of usage requires loads monitoring. In my examples at the top of the article, 2 of the 5 issues here needed no loads monitoring to understand the extent of the issue. That one is keeping track of how the aircraft is being used may be enough. So, in this case if it is understood that the SOIU is changing the DO can assess whether, say extended periods of time with landing gear down, might present a risk to the aircraft.

So, in summary for small/unusual platforms we can consider:

1. whether there is a DUS to compare to,
2. whether existing validation of the DUS and Loads Models may reduce the need for OLM,
3. whether the existing IAT can fulfil at least some of the requirements,
4. whether our usage or structure is sufficiently simple to reduce scope,
5. whether an originally civilian IAT can cope with the military deltas,
6. whether a staged approach is appropriate, and finally
7. what aspects can be covered by parametric monitoring and understanding the SOIU

…of course, you can always convince yourself of absolutely anything if you look at the data long enough (see Figure 4).

Figure 4 – Curve Fitting, xkcd, https://xkcd.com/2048/

Fin

Footnotes

{a} though perhaps it is unfortunate that the guidance is spread around as widely as it is – with aspects scattered around Defence Standards, the MRP, SI Handbook and MASAAG papers.

{b} OLM is very similar to the USA DOD ASIP requirement for a Loads/Environmental Spectral Survey (L/ESS) where the goals are for L/ESS to “to obtain operational usage data that can be used to update or confirm the design spectrum, identify when usage changes occur that warrant an update, and provide the data needed to establish or update a baseline spectrum.” [10]

{c} Though parameter data might come more crudely from Form 725 results and mechanical ‘fatigue meters’, or even novel examples like use of signal processing of video recordings of cockpit instruments in order to record flight parameters.

zj3nl9r5 See the MASIM, which now clarifies that in cases where there is a high degree of IAT capability that the AUVP may be largely carried out using the aircraft’s SHMS/HUMS systems, where there is low level IAT that OLM/ODR may be mixed with other recording methods to achieve AUVP, or where there is limited/no IAT use of OLM/ODR alone (see [3] Sections 12-14 and Table 4).

{e} Consider Leaflet 38, which states “The complexity of the instrumentation fit is normally related to the number of different loading actions that have an influence on the life of the structure and the potential variability of flight profiles” [4] para 6.1. And “If loads are not measured directly, and the transfer function changes for different flight states, e.g. in air and on the ground, the flight state will also need to be monitored to ensure that loads are predicted correctly” [4] para 6.6. In my experience it is not uncommon for DO’s loads models to be accurate for some parts of the flight envelope and not others.

{f} Cf. The SI Handbook, which states “The complexity of the solution chosen will be determined by the aims of the programme and existing confidence not only in actual usage, but also in the relationship between usage and the loads experienced by the air system. The TAA, in consultation with the DO, may make use of routine Usage Monitoring systems as well as existing aircraft parametric data (eg FDR or HUMS), and instrumentation systems developed especially for the task. Analysis of data captured from less complex systems may confirm a need to progress to a more complex instrumentation system.” [7] para 107.

References

[1] “Structural Integrity Management - Aircraft Usage Validation Process (AUVP) Guidance Document”, Version 5.1, MAA, 26 January 2016.

[2] RA5726, “Integrity Management”, Initial Issue, MAA, November 2019.

[3] “Manual of Air System Integrity Management (MASIM)”, Initial Issue, MAA, November 2019.

[4] Def Stan 00-970 “Design and Airworthiness Requirements for Service Aircraft”, Part 1 - Fixed Wing, Section 3 - Structure, Leaflet 38, Issue 15, 13 July 2015.

[5] MASAAG Paper 109, “Guidance for Aircraft Operational Loads Measurement Programmes”, Reed and Holford, QinetiQ Farnborough, 31 May 2007.

[6] MASAAG Paper 120, “Guidance on Helicopter Operational Data Recording Programmes”, Prof S C Reed, Mr G K Terry, Mr B H E Perrett, 27 October 201.

[7] “Structural Integrity Handbook Guidance Document in Support of RA 5720”, MAA, Issue 2.0, 19 April 2018.

[8] Aviation Investigation Report, A04O0020, “Aircraft Pitch-Up/Stall Warning On Departure Air Canada Boeing 767-233 C-GAUE Toronto/Lester B. Pearson International Airport Toronto, Ontario 26 January 2004”, Transportation Safety Board of Canada.

[9] Karl Hoffmann, “An Introduction to Measurements using Strain Gages”, Hottinger Baldwin Messtechnik GmbH, Darmstadt, 1989.

[10] MIL-STD-1530D, “Department of Defense Standard Practice: Aircraft Structural Integrity Program (ASIP)”, 31-AUG-2016.

? Stephen Dosman, 2019. Unauthorized use and/or duplication of this material without express and written permission from the author is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Stephen Dosman and this site with appropriate and specific direction to the original content.

#auvp #olm #operationalloadsmeasurement #aircraftusage #ra5726 #ra5720 #straingauge #structuralintegrity #integritymanagement #maa #airworthiness #continuingairworthiness