The Guidelines for direct monitoring methods must be based on the following three basic principles:

　

　

The only NOx emission limit mandated by MARPOL Annex VI is that given in Regulation 13. Any direct monitoring method guidelines developed are not to incorporate other forms of that limit (except should Regulation 13 be so amended).

　

The Regulation 13 engine speed related NOx limit as given is not an 'end of stack' measure. It is based on certain reference conditions :

　

・	Application cycle weighted value expressed in g/kWh.
・	Individual load point values under steady state conditions.
・	Use of a gas oil grade fuel (ISO 8217 - DMA, DMB or DMC) at the parent engine test irrespective of intended in-service fuel oil grade (hence the effects of fuel bound nitrogen and ignition delay related to fuel quality are avoided).
・	Parent Engine measured emission values are corrected to : 　1. Inlet air temperature 25℃ 　2. Absolute humidity 10.71 g/kg 　3. Charge (scavenge) air temperature corresponding to 25℃ sea water temperature (the effect of this depending on whether an engine's charge is directly cooled by sea water or by fresh water and in the latter case the fresh water temperature control regime adopted).

　

　

Since the direct monitoring method may be accepted as the sole means of demonstrating in-service compliance the engine is to be considered simply as a black box consuming air and fuel and producing brake power and an exhaust gas stream.

　

The significance of this is that the engine as monitored may be substantially altered from that as tested as the parent engine (or corresponding member engine at pre-certification survey). NOx critical components and settings may have been radically changed. Therefore, stemming from basic principle 1, the application of this method is not to verify whether an engine has been altered (tampering) from its 'as built' condition', but that it is still compliant with Regulation 13(if an anti-tampering control is required that is given by the parameter check method already widely adopted). In fact the particular need for the direct monitoring method is to allow for NOx critical components and settings to be altered/changed for alternative designs/suppliers over the service life of an engine (particularly slow speed engines) while still demonstrating compliance.

　

It would need to be considered whether the application of the direct monitoring method would even allow the rating (power and speed) of the engine as originally declared to be altered or indeed the application cycle (provided that the method showed that the engine so altered was still com pliant).

Consequently it cannot be assumed that any of the engine performance data as given in the parent engine test report or the technical file (as originally approved or subsequently modified) holds true for that engine as monitored at any subsequent date.

　

All data used as part of the direct monitoring method will therefore need to be shown as being applicable at the time of monitoring. In particular this covers aspects such as fuel consumption rates.

　

As a consequence of this'black box' approach the emission value as given on an engine's EIAPP (or pre-entry into force equivalent) similarly cannot be assumed to continue to apply to the engine in service. The only NOx emission value which will apply at any particular time will be that as obtained by means of the direct monitoring method at that time.

　

　

In view of basic principles 1 and 2 it is necessary that the Guidelines for direct monitoring methods should as far as is possible follow directly the NOx Technical Code.

　

Consequently, except where totally unavoidable, the requirements as given in Chapter 5 (and hence Chapters 2 and 3 together with Appendices 3, 4 and 6) of the Code should be adopted. Furthermore where aspects have already been developed for the simplified measurement method, i.e. Appendix 4, Tables 3 and 4, they should be retained as equally applicable to this method.

　

Note: This therefore follows ISO 8178 in which, although Part 2 refers to on-site measurements, it largely refers back to Part 1(Test-bed measurements) from which Chapter 5, etc., of the Code was developed.

　

　

　

　

In order to render the proposed Guidelines:

　

	(a)		not contradictory to the existing Code text, except where specific departures are necessary (as with the simplified measurement method as given in Chapter 6 of the Code); and
	(b)		simple (both to draft and later to use);

　

it is proposed that, as with ISO 87178-2, the format and structure of Chapter 5 of the Code (and hence related requirements in other chapters and the appendices) be retained as the basis of these Guidelines.

　

Consequently the Guidelines are to state at the beginning: 'The requirements of Chapter 5 of the NOx Technical Code apply except where expressly stated' and thereafter only give where variations are required. In framing these requirements the same format should be followed as used throughout the Code, that is that it is for the submitter (engine builder, ship owner, equipment manufacturer, etc.) to propose and for the Administration (directly or organisation acting on their behalf) to consider that proposal and accept, require additional justification, or reject against overall criteria as given in these Guidelines. Consequently the Guidelines are to be objective driven and not prescriptive.

　

　

The direct monitoring method to be applied to a particular engine is to be approved by the Administration responsible for the ship's MARPOL Annex VI certification (IAPP or pre-entry into force equivalent). The approval reference of that direct monitoring method is to be entered into the supplement of the EIAPP (or pre-entry into force equivalent) under Section 3.

　

The direct monitoring method as approved for a particular engine is to be sufficiently complete that all supporting data/justification is given therein and is therefore available onboard at the time of survey.

　

　

(At this time only the major points are covered, detailed aspects such that 5.1.9 would not be applicable are omitted.)

　

Following the existing Code references:

　

5.2.1	(AS previously amended by MEPC). Not applicable, data under any prevailing conditions to be evaluated.
5.3	Data as obtained with fuel grade in use at time of monitoring to be evaluated. In order to allow for the fact that the Regulation 1 3 limit is set on a gas oil grade fuel, an allowable is to be given for other (i.e. residual fuel oil) grades, see below re 5. 12.5.3.
5.4-5.8	See comments related to Appendices 3, 4 and 6.
5.9.2 & 5.11	Exhaust emission concentrations to be monitored and recorded to include at least NOx + CO2 and O2 (setting HC and CO to zero in the calculation marginally increases GEXHW, i.e. around 0.50/0). This does not preclude those who wish to also measure HC and CO and so obtain a marginally more accurate GEXHW value. Zero and span gas responses together with system leak test data to be also recorded.
5.9.3	The whole of the exhaust gas flow from an engine is to be monitored. Where engine exhaust pipes do not combine into one, or prior to such combination, multiple sample points may be used subject to approval.
5.9.5	For engines operating on propeller law speed curve (i.e. E3 cycle) the speed at each mode point is to be that which corresponds to the required torque. For engines driving variable pitch propellers the engine speed during monitoring is to be the rated speed. Where for a particular installation there are a number of speed/torque combination used in service the speed during monitoring is to be that which results in the highest NOx emission value, The test programme for such engines is to form part of the approved procedures.
5.10	Sufficient data is to be permanently recorded (chart recorders, data loggers, etc.) so that compliance can be subsequently demonstrated. The extent and form of presentation of data to be approved (to follow Appendix 5 plus charge (scavenge) air pressure, sea water temperature and charge air cooler inlet and outlet temperatures together with other data as applicable to ensure engines with specific features are adequately monitored. Traceability of data (i.e. analyser/instrument calibration, instrument signal, recorded raw and processed data) to be approved.
5.12.2.2	See comments related to Appendix 6 re HTCRAT
5.12.3.6	Tsc ref is to be defined for the engine across the applicable load cycle, the basis on which those temperatures have been determined is to be stated so that it may be verified that those conditions still apply. Actual values of Tsc as prevailing at the time of the monitoring to be used, as corrected to Tsc ref.
5.12.5.1	The means by which any Paux value is to be determined is to be approved. In the case of permanent coupled drive machinery (i.e. electrical generators) the requirements of 6.3.3 are to be followed and are to form part of the approved procedures together with any efficiency curves or equivalent.
5.12.5.3	The allowances as given in 6.3.11 are also to be adopted in comparing the determined emission value with the limit value.

　

　

Environmental suitability

　

Analysers to be installed in accordance with manufacturers requirements in order that the performance requirements as given in this appendix and 5.9.9 is achieved.

　

The basis of this is 'open' limitation is that environmental testing is not given for the test bed equipment which, particularly for large slow speed engines, may have to operate in less than ideal conditions, however they still have to meet the given performance requirements.

　

It is up to the direct monitoring system designer to place the equipment in a situation in which it will, and can be shown, to be functioning within the given criteria. If such equipment can be designed to so function in the region of the engine uptakes then that should be permitted, if it requires to be placed in a control room type environment then that is the choice taken.

　

Ultimately it is for the shipowner to demonstrate compliance. If the shipowner adopts a system which is not reliable then it is his interests which will suffer, unable to show compliance, survey not completed and all that that entails. Consequently the shipowner/system designer is the interface where a reliable system is to be established. If these Guidelines give testing criteria which must be met and a system so evaluated meets those criteria but fails to function to the standard required by the Code in service then has to owner failed to demonstrate compliance?

　

　

Calibration frequency

　

Analyser calibration frequency to be in accordance with manufacturers requirements in order that the performance requirements as given in this appendix, Appendix 3 and 5.9.9 is achieved.

　

Calibration to be capable of verification by use of one intermediate (40-50% of scale range) span gas of each type carried onboard. The traceability of span and zero gas data to be capable of verification.

　

　

Data required for the calculation of GEXHW at each load point are:
　・exhaust gas composition - see above re 5.9.2 and 5. 11
　・ambient data - readily recorded
　・engine performance data - as per Appendix 4
　・fuel composition - see below

　

Although the composition of each fuel oil consignment loaded (carbon, hydrogen, sulphur, etc.) could be determined (i.e. as additional testing as part of the fuel oil testing programmes already in place which a number of shipowners utilise) there would still be the problem of traceability between the fuel as loaded and the fuel as used at any particular instance. While the Guidelines should allow for such testing if the required traceability can be effectively demonstrated it is necessary)i that a default position be given. Despite the wide variation in the physical characteristics of marine fuel oils there is only a limited differences in composition and therefore default composition values should be given which in the absence of actual values from testing would be used in the calculations. Appendix 6, Table 1, already provides a composition for gas oil grades which should be adopted for those fuel oils which do not require pre-heating prior to injection (i.e. ISO 8217 DMA, DMB or DMC type fuels). For those fuels which do require pre-heating prior to injection (i.e. residual fuel oils) that default composition would need to be developed (i.e. carbon 86.2%, hydrogen 11.1%, sulphur 2.7%).

　

On the above basis GEXHW and HTC RAT are determined

　

　

　

Further to 2.3.4, monitoring is to be undertaken at least every 30 (engine operating) days - 720 hours - or whenever a change to engine components or settings has occurred which could affect NOx emissions, whichever is the less.

　

For engines where such changes are continuously being undertaken, as could be envisaged with some electronically controlled injection/charge air management/variable geometry turbocharger systems it would be necessary that the emission value be determined at least every 30 days with ongoing monitoring interpolating current performance into that established emission profile at a frequency set by the frequency at which NOx critical aspects are altered subject to some maximum frequency (i.e. 3 hours) .

　

It is considered that data must be retained for more than 30 days. Retention should cover at least the period between surveys, annual or periodic - intermediate.

　

　

An amendment to 2.1.2.5 of the NOx Technical Code should be adopted (as with the change to the limit range of the Test Condition parameter) to enable data from an approved direct monitoring method to be acceptable for the initial survey for those engines which are in service before the entry into force date of MARPOL Annex VI.

　

　

The Guidelines are to require that procedures for a particular direct monitoring method are to cover:

　

	・		how it is to be verified that the engine has been retained within the relevant emission limit;
	・		how it is to be verified that the data used above has been obtained in a correct and traceable manner.