1 INTRODUCTION

This report covers accidental CO poisoning incidents resulting from the use of piped natural gas for the period April 1st 1998 to March 31st 1999. Data for incidents up to 1995 comes from Advanticas own incident recording system. From 1996 the information is obtained from incident reports and investigation forms completed on behalf of gas suppliers. If any additional reports should be received after publication of this report they will be included within updated annual statistical tables in future reports. Domestic incidents are covered in the main part of the report with LPG incidents and business incidents reported in Appendix 3 and 4 respectively. Suspected intentional incidents have not been included in the analysis. Information for this report comes via the Downstream Incident Data Report (DIDR) - Form 551/7. Tables and plots of actual fatalities and incidents and also plots relating to the risk associated when using gas appliances expressed in terms of fatalities per person per year (FPPY), as incidents per person per year (IPPY) and as casualties per person per year (CPPY) are given. The definitions and use of IPPY and CPPY values are described in Appendix 1. Fatality, casualty and incident trend data are presented for incidents that occurred between 1991/92 and 1998/9.
Note: Some inconsistencies may appear in some parts of the report because all the required information may not have been completed on the DIDR forms e.g. in Table 7 the numbers of casualties, as represented by their location, differs from the total number reported in Table 1. Some information was completed as unknown or other and in some instances the tick box was not completed (field empty).
Appendix 2 gives details of each of the CO poisoning incidents for 1998/9. The order used in this report follows the layout used in the DIDR - Form 551/7.
Note: Included on the DIDR form are 3 sections to complete related to the installation - to current standards, to standards current at time of installation, not to any appropriate standards or unsure/dont know, of the following: the appliance the flue the permanent ventilation
For the appliance items that are standards related, incl de the correct room/location, u proximity to walls, fire resistance and electrical safety. Each of the three items are dealt with separately on the DIDR form and within this report.
2 ANALYSIS OF REPORTED DATA
TOTAL INCIDENT DETAI LS - ANALYSIS OF SECTION 1 OF DIDR
Transco issued 371 Incident Notification Forms during the reporting period. These gave details of CO Poisoning Reports under their companies internal reporting procedures. There were 107 domestic incidents that met the requirements for reporting on the DIDR form with the majority of these being notified directly to Transco, via the operation of the national gas emergency service, and advised by Transcos internal procedures. In addition there were some incidents reported directly to gas suppliers by, for example, coroners or the police that did not get entered onto Transcos reporting system. All reports were fully analysed for this report and every effort was made to obtain as many completed DIDR forms, for this report, as possible. However due to the voluntary nature of the reporting scheme it is likely that a very small number of reports were not supplied. If any additional reports should be received after publication of this report they will be included within updated annual statistical tables in future reports. Each form is treated as a separate DIDR incident and will be referred to as an incident throughout the rest of this report. The incident risk data and trend data has been combined with the casualty details described in section 2.2. The date of occurrence of each domestic incident has been plotted by month in Figure 1, for the 12 month period April 1998 to March 1999.

Nos (%) 1 (0.9) 8 (7.5) 2 (1.9) 15 (14.0) 1 (0.9)
House Detached Semi-detached Terraced Townhouse
Nos (%) 9 (8.4) 18 (16.8) 48 (44.9) 1 (0.9)
The Living in Britain 1998 General Household Survey from the ONS gives a breakdown of types of accommodation in Britain. The analysis is given below where it is compared to the incident statistics.
Table 5 - Comparison of DIDR incident stats with accommodation stats
Property style Detached house/bungalow Semi-det house/bungalow Terraced house/bungalow Purpose built flat or maisonette Converted flat or maisonette/rooms
Accommodation Stats for GB (%) 4

Incident Stats (%) 10

The age bands of the properties in which incidents took place are shown on Figure 7.

17% 1% 9% 7% 46%

Pre to to to 1991 Post1992 Other
Figure 7 - Property construction period
The two largest sectors are properties built b efore 1945 at 46% and those between 1946 and 1965 at 20%. The age was unspecified for 17% of the incidents. Where the age was specified (89 properties) the pre 1945 group is the majority at 55%. The next largest group was 24% for those built between 1946 and 1965. The remainder, built from 1966, totalled 21%. The Living in Britain publication, 1998 edition, from the Office for National Statistics (ONS) states that for Britain 40% of all dwellings were built before 1945, 23% were built in the period 1945 to 1964 and 37% were built during or after 1965. Figure 8 gives the occupancy types of the properties shown on the DIDR forms. The percentage owner occupied was 61% and 36% were tenanted. Empty fields or unrecognised values made up the remainder. Of the tenanted properties group 31% were single occupancy and 5% were multiple occupancy. The percentage of the tenanted sector that were council owned is 14%, privately owned was 17% and 4% were owned by a housing association, the remainder were classified as other/unknown.

Unknown occupancy

Tenanted - unknown type Tenanted - other Tenanted - private Tenanted - housing association Tenanted - council Owner occupied
Owner occupied Tenanted, single occupancy Tenanted, multiple occupancy Unknown occupancy
Figure 8 - Occupancy type
The Living in Britain 1998 General Household Survey from the ONS gives the owner occupied tenure group as 69% and the tenanted sector as 30%. This covers renting from the local authority at 16%, privately at 9% and from a housing association at 5%. Table 6 shows the analysis of the glazing and ground floor details for the incident sites. These are also described graphically in Figures 9 and 10.

Notes: Appendix B gives details, by appliance type, for each incident. In the above table and following tables g.f. has been used as an abbreviation for gas fire.
There were no reports of any condensing appliances having been involved in any incidents during this reporting period. The breakdown of the types of central heating units involved in incidents is given in Figure 14.

Ot he r Ot he r

Figure 15 shows the fatality trends associated with appliance type since 1991/92. It should be noted that it is likely that there have been changes to the profile of gas appliances in use, within Britain, between 1991/92 and 1998/99. The FPPY risk values shown in Table 12 take account of these changes.
Ba ck bo Ba iler u ck nit Flo or boile sta ru n n Flo Floo ding it r st bo or iler an sta Flo ndin ding g b or sta com oiler nd bi ing bo co iler mb Th erm i boi ler al Th stor erm age un al W stora it all mo ge un W unte it W all all m d bo mo iler ou W unte nted d all b mo com oile r bi un bo ted iler co mb ib oil W er arm a W ir un arm it air un it
Non-fatal Casualties Fatal Casualties Incidents
Figure 14 - Central heating boilers
0 91/92 92/93 93/94 94/95 95/96 96/97 97/98 98/99
Central heating boilers Space heaters Water heaters Cookers Unclassified
Figure 15 - Fatalities by appliance type
The age of the appliances involved in incidents during the reporting period has been given under the main appliance groups in Table 9. It is also described in Figure 16.
Table 9 - Age of incident appliances
Appliance Type Central heating boilers Cookers Space heaters Dryers Water heaters TOTAL
Age (years) 6 - - - 7 Over 7 Unknown 1 65
Water Heaters Central Heating Space Heaters Tumble Dryers Cookers 80 90

Number of appliances

less than 6 years 16 to 20 years 6 to 10 years over 20 years 11 to 15 years unknown
Figure 16 - Appliance age distribution
Notes relating to individual appliance types and models
The following information is extracted from the incident details given in Table 8 and Appendix B: 2.5.2.1 Central Heating
Central heating appliances featured in 79 incidents, which is approximately three quarters of all CO poisoning occurrences during the reporting year. The number of fatalities at 13 was just over half (56%) of the total recorded, with the number of nonfatal casualties being 196 (85%).
Wall hung boilers were involved in 39% of central heating incidents, with floor standing boilers and wall mounted combi boilers being the next highest groups at 24% and 23% respectively. Back boiler units were responsible for the majority of fatalities, at 6, with wall mounted boilers being responsible for 3 fatalities. Access was not obtained for 2 incidents resulting in incomplete details being entered on the DIDR form Back boiler units There were 7 back boiler incidents, 2 of which were fatal, with there being 6 fatalities in total. Nonfatal casualties totalled 10. In 3 cases the appliance was in need of servicing. In four incidents the ventilation was not to standard and in 3 cases there were significant flueing faults with the flue installation not to any standards. Two of the appliances were Baxi Bermuda units and the others were all different models. Floor standing boilers Floor standing boiler incidents totalled 19, with 1 fatality and 51 non-fatal casualties. Eight of the appliances were Potterton Kingfisher models and 3 were Potterton Diplomats. In 14 incidents the flame picture was defective and in 12 incidents Linting had taken place. In 13 cases the flue was not to standard, with the terminal siting being poor in 12 instances. In 11 installations the ventilation was not to standard. Weather was thought to have contributed to the poor performance of the appliance in 11 of the incidents. In nearly half of the cases the appliance was in need of servicing and in 1 incident the appliance had been labelled as being unsafe, but it was still in use. An open flued boiler was fitted in a bathroom at one incident site.

Note: Some appliance models may appear under several different manufacturers names within Appendix B. For example Apollo boilers have been entered under Thorn, Myson and Potterton Myson.
Floor standing combi boilers There was 1 incident featuring a floor standing boiler, the model being a IMI Powermax. There were 2 casualties. The appliance was producing high levels of CO and the failure of a section of flue ducting lead to the CO entering the property. 2.5.2.1.4 Thermal storage units There were no recorded incidents involving these appliances. Wall mounted boilers Wall mounted boilers were involved in 31 incidents, with 3 fatalities and 83 non-fatal casualties. Twenty-two of the appliances were open flued, natural draught and 4 were room sealed natural draught. Four were room sealed, fanned flue and in one case the flue type was not recorded. Thorn Apollo models and Glow Worm Fuelsaver models both featured in 11 incidents. Faulty case sealing, the removal of the cover or a damaged case led to the incidents occurring in all but one of the room sealed appliance installations. Weather was a factor in 14 incidents, with ventilation faults, flueing faults, linting and defective flame picture numbers being similar. Wall mounted combi boilers Combi boilers were involved in 18 incidents one of which resulted in 1 fatality. There were 45 non-fatal casualties. All appliances were open flued, natural draught models. Two open flued models were fitted in bathrooms. Vaillant models featured 11 times with the T3 model 6 times and the GB model 4 times. Flues were not to standard in 15 cases and there were 12 flueing installation faults. Ventilation was not to standards 13 times. The weather also featured in 10 incidents.
Warm air units There were 3 incidents with warm air units and they were all on open flued, natural draught models. Of the 3 incidents 2 were fatal, with a fatality in each, and there were a total of 5 nonfatal casualties. Faults within the appliances appear to have led to the production of CO and its entry to the property. 2.5.2.2 Cookers
There were 7 incidents, with 4 fatalities and 9 non-fatal casualties, involving cookers and in each case it was the grill burner that led to the production of CO. Of the 7 incidents 4 were fatal incidents. The appliance model was different in each incident. In one i cident the grill was n being used to provide heat for the property and in two further cases it appears likely that the grill had been left on for an extended period of time. In one incident a CO alarm was activated and the customers were able to respond. 2.5.2.3 Space Heaters

Space heaters, either decorative type or radiant convector heaters, were involved in 18 incidents. In 5 cases the incidents were fatal, with 5 fatalities. There were 19 non-fatal casualties. In 11 incidents the flue was found to have a blockage and in 11 cases linting had taken place. Misplaced radiants or incorrectly positioned coals were the cause of CO production in 4 incidents. In 1 instance the appliance had not been connected to a flue. 2.5.2.4 Tumble Dryers
There were no appliances reported during the year in association with CO incidents. 2.5.2.5 Water Heaters
Water heating appliances featured in 3 incidents of which 2 involved single point water heaters and the other a circulator. One single point water heater incident was a fatal incident with 1 fatality. The other two incidents led to 7 non-fatal casualties. All three appliances required a service and had blocked heat exchangers. 2.5.3 Appliance risk values
Details relating to the risk values by appliance type are shown below in Table 10. In terms of the risk of a fatal incident (FPPY) only the single point water heaters and the warm air units have a risk value greater then the recommended level of 1 x 10-6. The appliances in descending order of risk are as follows: Single -point water heaters (2.6 x 10-6 ) and Warm air units (1.6 x10 -6 ).
Table 10 - Risk values by appliance type
Appliance Back boiler unit Floor standing Floor standing combi Thermal storage unit Wall mounted Wall mounted combi Warm air unit Free standing Built-in oven Built-in hob Balanced flue g.f. Cabinet heater Decorative g.f. Flueless heater Inset live fuel effect g.f. Rad. & rad. Con. g.f. Wall heater Tumble dryers Bulk storage Circulator Multi-point Single-point
Population FPPY (x10 6) (x10 -6 ) Central Heating Boilers 3.22 0.79 3.07 0.14 0.28 7.19 0.18 3.14 0.13 0.51 1.6 Cookers 9.14 0.18 Space Heaters 2.13 7.42 0.28 Dryers Water Heaters

0.16 2.6

CPPY (x10 -6 ) 1.3 4.4.1 0.41 0.10.5
IPPY (x10 -6 ) 0.92 2.6 1.5 1.8 2.4 2.5 0.3 0.4 0.91 5.3
Note: Population figures provided by GfK Marketing Services Ltd. (Reference 7.1.1). Population figures were not available for all appliance types and therefore risk values could not always be calculated.
Trends (1989/90 -1998/99)
Trends regarding CO Poisoning incident fatalities by appliance type are given below in Table 11 and are also shown in Figure 15, which is in section 2.5.1 of the report. This table has been completed as fully as possible using information that was available from the 98/99 DIDR forms and from historical records held by Advantica (Reference 7.1.2).
Table 11 - Trend data of the number of fatalities due to CO incidents, by appliance type

Appliance C/H Boilers -Total Back unit Floor standing Floor standing combi Thermal storage unit Wall mounted Wall mounted combi Warm air unit Cookers -Total Free standing Built-in oven Built-in hob Space Heaters -Total Balanced flue g.f. Cabinet heater Decorative g.f. Flueless heater Inset live fuel effect g.f. Rad. & rad. con. g.f. Wall heater Dryers Water Heaters -Total Bulk storage Circulator Multi-point Single-point Other TOTAL -All Appliances
91/92 92/93 93/94 94/95 95/96 96/97 97/98 98/99
Trends in terms of the risk of a fatality by appliance type, expressed as FPPY values are shown below in Table 12. This table has also been completed as fully as possible using information that was available from the 98/99 DIDR forms and from historical records held by Advantica.
Table 12 - Trend data of fatalities per person per year (FPPY)
Appliance C/H Boilers -Total Back unit Floor standing Floor standing combi Thermal storage unit Wall mounted Wall mounted combi Warm air unit Cooke rs -Total Free standing Built-in oven Built-in hob Space Heaters -Total Balanced flue g.f. Cabinet heater Decorative g.f. Flueless heater Inset live fuel effect g.f. Rad. & rad. con. g.f. Wall heater Dryers Water Heaters -Total Bulk storage Circulator Multi-point Single-point Other TOTAL -All Appliances
0.14 0.64 0.1 0.43 0.5 0.29
0.38 18.3 1.1 0.67 0.06 0.42 1.3 0.34
0.17 37.5 0.54 0.7 0.01 0.36 0.9 0.28
0.27 0.65 0.54 1.38 0.07 0.24 1.47 0.29
0.38 0.39 0.23 0.27 1.1 0.76 0.16 0.24 0.16 3.81 -
0.12 0.26 0.11 0.17 0.13 0.19 0.54 8.78 -
0.31 0.79 0.14 0.18 0.13 1.6 0.12 0.18 0.28 2.6 -
Note: In Table 12 all the FPPY values are x10-6
SAFETY DEVICES - ANALYSIS OF SECTION 6 OF DIDR
A total of 16 safety devices were noted as being fitted within the incidents investigated. Eight were downdraught detectors, two were vitiation devices and four were CO chemical spot detectors. The remaining two were powered CO alarms of which one was mains powered and one was battery powered. In 3 cases the safety devices were found to be non-operational. This was for one downdraught detector, one spot detector and one battery powered alarm. 2.7 FLUE DETAILS - ANALYSIS OF SECTION 7 OF DIDR

The majority of appliances were open flued, individual, natural draft (83 incidents - 78%). There were also 10 individual room sealed flues, 5 of which were fanned, and 9 flueless appliances. Flueing details are given in Figure 17.
Other Room sealed, individual, natural draught Room sealed, individual, fanned draught Flueless Open flued, individual, natural draught
Figure 17 - Incidents by flue type
The analysis of flues to standard is given in Figure 18. There were 50 incidents (46%) where the flue was not to any appropriate standards, 34 (32%) of flues to current standards, 9 (8%) to standards applicable at the time of installation and 14 (13%) which were not known.

To current standards

To standards current when installed Not to standards Unknown
Figure 18 - Flues to standard
The number of flueing faults found are given in Table 13 (report section 2.9). A breakdown of the flueing faults, by appliance type, is given in Appendix B. Details of the flue compliance to standards, for each incident appliance, are also given in Appendix B. Flue liners were fitted in 14 cases. In 7 cases the liner was fitted with the appliance and in 7 cases it was not known. 2.8 PERMANENT VENTILATION - ANALYSIS OF SECTION 8 OF DIDR
Permanent ventilation was required in 76 (71%) of the incidents and was not required in 31 (29%) of cases. Where ventilation was required it had been provided in 64 of the cases (84%) and when provided it was only to current standards in 21 installations (33%). It was not to standards current when installed in 39 installations (61%). Where permanent ventilation was required and air vents were fitted they were still effective in 49 (64%) of incidents and partially effective in 8 (10%) of the incidents. In 7 incidents the ventilation was totally ineffective. Of those with totally or partially ineffective ventilation, 6 were blocked intentionally and 5 unintentionally. Incident appliances were fitted in compartment/cupboards in 35 incidents. The compartment/cupboard was to standards applicable at the time of installation in 9 (26%) instances. It was not to standards in 25 (71%) instances. In one case it was unknown whether the compartment/cupboard met standards. Extract fans, recirculating fans and cooker hoods were reported to have been in use during one incident. The number of overall ventilation faults found are given in Table 13 (report section 2.9). A breakdown of the ventilation faults, by appliance type, is given in Appendix B. 2.9 ON-SITE CHECKS - ANALYSIS OF SECTION 9 OF DIDR

7.1.2 7.1.3 7.1.4

REFERENCES
Definitions of FPPY, CPPY and IPPY - Advantica Reports.
APPENDIX A DEFINITIONS AND THE USE OF FPPY, IPPY AND CPPY VALUES
a) Fatalities Per Person Per Year (FPPY) FPPY is a measure of the risk of death from owning a specific appliance type. FPPY is defined as:-

FPPY =

Notes:
Number of Fatalitie s Number o f pe ople at risk x Appliance Population
1) In the report the number of people at risk is taken as the average number of people per household (2.37 in 1998/99). - provided from Government Statistics - see report section 7. 2) The Overall FPPY is calculated, as above, except that Appliance Population is replaced by the number of customers - see report section 7. 3) The appliance population figures used have been taken from information provided by GfK Marketing Services- see report section 7.
b) Incidents Per Person Per Year (IPPY) IPPY is a measure of the risk of having an accident with a specific appliance type. IPPY is defined as:-

IPPY =

Number of Incidents Number of people at risk x Appliance Population
c) Casualties Per Person Per Year (CPPY) CPPY is a measure of the risk of being injured by owning a specific appliance type. CPPY is defined as:-

CPPY =

Number of Casualties Number of people at risk x Appliance Population
APPENDIX B TABLES, BY APPLIANCE TYPE, SHOWING THE NUMBER OF FAULTS AND INDIVIDUAL INCIDENT DETAILS
Table B1 shows the tables included in this appendix. They have been completed for the appliance groups only where there were relevant incident appliances to describe. The nomenclature adopted allows data to be presented for any of the appliance groups. This has the advantage that tables with the same code may be readily identified, which can aid the comparison on a year-by-year basis. However, groups may not have been implicated in incidents in any particular year, so they are indicated in this appendix as no reported incident. The appliance groups have been ordered in the same way as section 2.5.2 of the report.
Table B1 Summary of incident fault analysis and summary tables presented

Appliance group

Appliance sub-group Back unit Floor standing
Code 1.1 1.2 1.3 1.4 1.5 1.6 1.7 2.1 2.2 2.3 3.1 3.2 3.3 3.4 3.5 3.6 3.7 4.1 5.1 5.2 5.3 5.4

Incidents 0 2

Appendix tables B.1.1a & b B.1.2a & bi-iii B.1.3a & b No reported incident B.1.5a & bi-iv B.1.6a & bi & bii B.1.7a & b B.2.1a & b No reported in cident No reported incident No reported incident No reported incident B.3.3a & b No reported incident No reported incident B.3.6a & bi & bii No reported incident No reported incident No reported incident B.5.2a & b No reported incident B.5.4a & b

Glow Worm

Appliance location Appliance age (yrs) Number of casualties: fatal (non-fatal) Current Fuel Saver

MK2 40R

SE10 Current Current Apollo 35/50
Potterton Prima 60C Myson

Post Code 5 Current

Myson Apollo 15/30C

0 (4) 1

The boiler was producing high levels of CO which entered the property due to a soot blocked heat exchanger. It was in need of servicing. The flueing was sub-standard.

1 (0) Current 5

Potterton Netaheat 1016
Boiler in poor condition & producing high levels of CO. The appiance had been modified. It was a room sealed appliance which suffered from pilot outage due to over heating of the high level stat. The case had been removed to keep it alight.

0 (4) 2

Myson Apollo 30CC
Flue outlet of a fanned, balanced flue boiler blocked by a wasp nest. A distorted back panel allowed products into the room.
Boiler in poor condition and was producing high levels of CO. Flue performance affected by prevailing wind conditions and the operation of the chimney serving a gas fire.
Boiler in poor condition. Inadequate ventilation and compartment ventilation. Terminal position to standard, but affected by the wind. Spinning type air flow vent extracted air from the room in gusty conditions.

Thorn Apollo Glow Worm

Poor condition. Inadequate ventilation and compartment ventilation. Flue to standard, but terminal fitted in a wall adjacent position.

0 (2) Current

The boiler was sooted and was producing high levels of CO which entered the property due to flue blockage. The ventilation and flue terminal were substandard

Current Fuel Saver

50R MK2
The boiler is thought to have been affected by the weather, with a passive stack & thermal inversion to the property. This resulted in CO entering the property.
Table B.1.5biv - Central heating boilers : wall mounted boiler : incident summary
50 MDF Potterton Netaheat 10/16

51 Current 5

Thorn Apollo 15/30B

B79 Current Current

Ideal460 WRS
Post Code Current Micro Turbo 2
Glow Worm Fuel Saver 40 MK 2
High levels of CO were being produced by the boiler as it was overrated. The products entered the property due to a displaced flue joint probably caused during property refurbishment. There was also a lack of purpose provided ventilation and a substandard flue On the day following a boiler service a thermostat capillary tube was found to be trapped in the case seal which allowed products into the property

0 (2) 5 1

The boiler was in need of servicing and was producing high levels of CO. The flue was not clearing all the flue products. The use of a gas fire caused the boiler performance to deteriorate.
Current 10 Unknown when installed

The boiler was producing high levels of CO due to blockage of the heat exchanger. Spillage into the inner case entered the property due to incorrect location of the inner case seal

0 (5) Current

Glow Worm Fuel Saver 30-40B Thorn Apollo 40C
The boiler had a distorted back panel which did not seal fully to the case. It is likely that under certain weather conditions combustion products could leak from the boiler into the rooms atmosphere.
The front panel had been removed from the boiler and under certain weather conditions combustion may have deteriorated and combustion products entered the property
An open flued boiler installed in a non standard location in a compartment in a bedroom required servicing. It was producing high levels of CO. The ventilation and flueing were substandard. Under certain weather conditions downdraught & spillage into the bedroom would be likely. There was evidence of intermittent spillage.

Glow Worm Space Saver 38

Products of combustion entered the property via the case seals. It is likely that combustion would deteriorate & spillage increase under certain weather conditions.
WALL MOUNTED COMBI BOILER
Table B.1.6a - Central heating boilers : wall mounted combi boiler : Summary fault analysis number of incidents=18

Number of faults 11

Table B.1.6bi - Central heating boilers : wall mounted combi boiler : incident summary
Appliance location Appliance age (yrs) Number of casualties: fatal (non-fatal) 5
Vaillant VCW GB 280H Vokera 20-80CF Flowmatic Vokera 1872 DMCF Saunier Duval SD223C
Spillage was likely to have been caused by the undersized terminal restricting the flow of products of combustion.

0 (2) 5 5

The cause of the incident was an opened flue combination boiler installed in a bathroom in the flat below. The boiler was producing high levels of CO which travelled up the cavity of the property to the flat above. Contributing factors were a wall faced terminal and a partially blocked flue. The gas fired boiler provided the source of carbon monoxide.

B.5.3 B.5.4

MULTI-POINT NO REPORTED INCIDENT SINGLE-POINT
Table B.5.4a - Water heaters : single-point : Summary fault analysis number of incidents=2
Table B. 5.4b - Water heaters : single-point : incident summary

Vaillant MAG 125/7

Post Code 11 Current

Main Mitre

Poor state of service. Heat exchanger, pilot & burner required cleaning.
Current Unknown when installed
11 The water heater was in poor condition and was producing high levels of CO into the bathroom.
APPENDIX C DETAILS OF LPG INCIDENTS THAT TOOK PLACE DURING 1998/99, AND AN ANALYSIS OF THE DATA.
There were three LPG incidents reported using the DIDR Form 551/7 during the period 1998/99. Two of these incidents occurred in static caravans (incidents A & B) which were located at caravan parks and the other incident (C) occurred in a small hotel/inn. The incidents happened at different times of the year. The caravan incidents occurred in July and August. The commercial premises incident occurred in November. In all of the incidents there were relatively large numbers of casualties compared to those summarised earlier in this report. In one caravan incident there was a total of six non-fatal casualties and in the other there were eight non-fatal casualties. Ages of the casualties ranged from 1 year to 52 years old and there were no fatalities in either of these two caravan incidents. The commercial incident was a fatal incident with 2 deaths and 6 non-fatal casualties and the victims ages were between 26 and 30 years old. Details of these incidents and casualties are given in Table C1.
Table C1 - The number of CO incidents and casualties

Incident A B C

Post code DD11 PE25 SY95
Appliance involved Multi-point water heater Rad & rad conv gas fire Floor standing boiler
Number of fatal casualties 2
Number of non-fatal casualties

N1 N2 N3 N4

Both caravans were tenanted properties and in incident C the property was owner occupied. At incident A the appliance was installed in a compartment/cupboard and it failed the flue flow and continuity check. The appliances involved were positioned in a room in the other two incidents. In incident A a battery powered CO detector had been installed in the kitchen of the caravan and was subsequently found to be operational. Details of the appliance and casualty locations are given in Table C2.
Table C2 - Appliance and casualty locations.
Appliance location Hall Lounge Laundry store room
Casualty locations Unknown Bedroom & lounge Bedroom

Flue type 5

Note: Flue Type codes are given at the start of Appendix B
Details of the incident appliance make and model are given in Table C3.

Table C3 Appliance and standards details.
Appliance make & model Poloma PH-5-3f Stoves Newholmes Clipper l Stelrad Mexico Super CF 125

Appliance age (years) 15

Installer Unknown Unknown Unknown
Appliance installed To Standards Current Current Current
Flue To Standards No Current No
Ventilation To Standards No Current No
The following faults and causes were reported at each incident : Incident A There was a fault with the flue, which was the established cause of the incident. There had been work carried out on the flue between 6 and 12 months before the incident. Incident B There was a fault with the appliance burner and it was also corroded. There was also sub-standard servicing on the appliance. It had been the subject of a safety check/inspection within 6 months of the incident. Incident C The report was not conclusive as to the cause, but it was identified that there were flue installation faults, the terminal position was poor and the weather may have been a contributing factor in producing downdraught. Vitiation, in the small room in which the boiler was located and which had undersized ventilation, could have led to high levels of CO production. The ceiling was not fully sealed and was identified as a route to the rooms where the casualties were located. The following details in Table C4 give the total numbers of faults found at the installations involved.
Table C4 - Incident appliance faults

Number of faults 3

APPENDIX D DETAILS OF NON-DOMESTIC CO INCIDENTS THAT TOOK PLACE DURING 1998/99, AND AN ANALYSIS OF THE DATA.
During the reporting year 1998/99 there were 4 CO incidents reported using the DIDR form that involved piped natural gas within business properties. Incident A occurred in an office, incident B in a shop, incident C at a commercial premise and the remaining incident took pla ce in a pub. Incident D was caused by fumes from a boiler installation supplying the flat above the pub. The appliance was located in a compartment at the bottom of the stairway. Incident A occurred in July 1998, B during December and C and D in February 1999. Details of these incidents and the resulting casualties are given in Table D1 below. It can be seen that in each incident there was only one casualty who required treatment for less than 24 hours in hospital. Property A was tenanted, single occupancy and council owned. The building was built pre 1945. In incident B the premises were built in 1956. No details regarding the property were given for incident C. Property D was an owner occupied property, built pre 1945, and consisted of a flat above a pub.