Go to content

4. Assessment of practical feasibility and added costs for manufacturers

Treating the initiatives one by one we will here evaluate feasibility and any effects on testing costs.

Increase the basic firebed by ashes build up only

This initiative has no impact on feasibility or cost. It only implies that the combustion cycle duration is extended by 3–5 minutes typically.

Skip the CO2 option as end of test criterion

This initiative has no impact on feasibility or costs. It only implies that the combustion cycle duration is altered a little. Some combustion cycles will last a few minutes longer, some a few minutes shorter. This is due to the unpredictable drop of the CO2 value as the fire goes out.

Skip the suction pyrometer option (flue temp)

This initiative has no or little impact on feasibility and costs. Some test labs might keep the suction pyrometer as an implement to extraction of gasses for CO and CO2 analysis, some might switch to using a heated probe and sampling line (same principle as for THC and NOx).

Increase number of burn cycles from 3 to 5

This initiative has no or little impact on feasibility and costs, despite extending the effective testing time. Test labs are assigning a full day for an output test, where they are able to complete 7–8 combustion burn cycles, assuming the average cycle duration is 45 minutes and the first hour or two is spent on ignition and pre-heating the stove to a thermal equilibrium. So, the time is spent anyway, and it is no prob­lem to accomplish the minimum combustion cycles in sequence in one day. If the stove burns uneven or it is hard to control due to poor preparation for the test, it might take an extra day to frame the output test, but this is frequently occurring already today.
Some manufacturers have a full in-house test lab and skilled technicians, that allow them to trim the stove and its operation to an optimum before undertaking the test. Other manufacturers have only limited in-house test facilities, and they will typically spend more time at the test lab to accomplish a satisfactory test result. In the end it is a trade-off between investing in elaborated in house test capacity and save time and money at the test lab – or having no or little internal test capacity and spend more time and money at the test lab. Going from 3 to 5 burn cycles does not change this fundamental trade-off. If the stove is well trimmed by an experienced technician, it is no big deal producing 5 good combustion cycles in sequence.

Reset minimum cycle duration to 45 min

This initiative has no or little impact on feasibility and costs. Obviously, the combustion cycle duration is going to be extended by a couple of minutes. This is however not going to affect the yield of a days of work in the test lab, in terms of number of combustion cycles accomplished.

Remove inter-cycle break (and revert to HF)

The first part of removing the break is of paramount importance on the test quality. It can be accomplished regardless of keeping the EN-PME dust sampling principle or skipping the EN-PME and reverting to the former Heated Filter sampling principle. There are no added cost costs nor any adverse feasibility effect of removing the break (in many test labs, the pause has been introduced yet as it is not yest formalized in the EN16510 standard). Reverting to the Heated Filter (HF) sampling principle represents an upside, as the HF principle is significantly easier to operate – it can be handled by one staff member alone and it offers ample time to replace the filter before commencement of the subsequent combustion cycle. Emission wise there is no difference as demonstrated by the WG5 validation project referred above proving parity between EN-PME and HF.

Include part load testing

If ACC controls (Automated Combustion Controls) become compulsory there is no justification for introducing part load testing at the time being. CEN has not yet drafted a test method for ACC stoves; the work has just begun. Part load testing can currently be reserved to manually controlled stoves alone, where it is well justified and there is a method documented in the test standard. Introducing part load testing almost double the time consumption at the test lab for the output test, and consequently testing costs are increased, yet not doubled as the temperature safety test is not affected by part load. There can also be an effect on feasibility, if the stove is not well trimmed for part load operation, as discussed above. But most important, part load is counterproductive to test reproducibility. Part load is further away from the ideal working point combustion wise and it carries a wider range of results than measurements at nominal heat output.

Standardize the test fuel

This initiative has no impact on feasibility or costs. The fuel costs are going to be the same and so is the time consumption at the test lab. There might be a potential minor upside, if the relative standard deviation is minor than when using log wood. This could minimize needs for retests.
Summing up the effects of these initiatives, setting part load testing aside, and considering upsides and downsides, the effects in terms of feasibility as well costs are close to neutral. Only part load is undoubtedly a cost driver. Assuming part load is reserved to manually controlled stoves alone, this bias could also urge manufacturers of manually controlled stoves to move on and develop ACC systems faster than otherwise planned.
Other improvement initiatives suggested by the partners worth considering – the effect however is not easily quantified.
  • Enable better correlation between how the appliance is fired during the Initial Type Test (ITT) and how the firing description is written in the instruction manual of the appliance. Introduce the Quick User Guide (QUG).
  • Allow only optimizing the air control setting during the first three minutes (e.g. with a stopwatch and a caliper) if it is clearly stated in the instruction manual under start up guidance.
  • Have better clarification in the test standard to which degree a manufacturer may witness a type test - or take active part in some of the operations during the test. Consider requesting any guidance to the test lab documented by a set of testing instructions written or as a video or a combination of both.
  • Enforce compulsory participation of NoBos in interlaboratory comparisons tests among the members of the Group of Notified bodies network SG03-WG2 in fixed 3-year intervals. And always when a new standard replaces the former one. There is already a position paper from GNB-CPR/GNB-AG stressing the importance of using interlaboratory comparative tests as an essential part for ensuring even performance of the Notified Bodies. It is further stressed by ISO17025 clause 7.7.2 that Notified Bodies are obliged to take part in interlaboratory tests.
So, all in all the framework is in place. We just need to make sure it happens. This might be supported by closer follow-up by the notifying authorities appointing the NoBos, and/​or the national accreditation bodies monitoring the operation of the NoBos. The paper from GNB is attached as annex 2.
Such compulsory interlaboratory comparisons will have a significant positive impact on reproducibility of tests, and it could also work as a strong QA tool, strengthening the impact of Market Surveillance measurements. In combination, the two measures will be very effective – one at the front end and the other on the output of the test.